Predicting Individual Disease Risk Based on Medical

DaffyDoc (United States of America)
303 words 4 learners

Learn words with Flashcards and other activities

Definition First极速168赛车 377936e02ce27ce724b98272ce20c0e0
Print Flashcards极速168赛车 902a441ff479f3adacbe4135eb3a3334

Other learning activities

Answer a few questions about each word. Use this to prep for your next quiz! Compete with other teams in real time to see who answers the most questions correctly! Test your spelling acumen. Read the definition, listen to the word and try spelling it!

Teaching tools

Create and assign quizzes to your students to test their vocabulary. Assign learning activities including Practice, Vocabulary Jams and Spelling Bees to your students, and monitor their progress in real-time.

Full list of words from this list:

words only definitions & notes
  1. descriptor
    the phonological or orthographic sound or appearance of a word that can be used to describe or identify something
    Categories and Subject Descriptors
    I.2.1 [Arti cial Intelligence]: Applications and Expert
    Systems|Medicine and science; I.5.1 [Pattern Recogni-
    tion]: Models|statistical ; J.3 [Computer Applications]:
    Life and Medical Sciences|Medical information systems
    General Terms
    Algorithms, Experimentation
    Keywords
    collaborative ltering, disease risk prediction, ensemble, prospective
    health care
    Corresponding Author
    Permission to make digital or hard copies of all or part of this w...
  2. phenotype
    observable characteristics produced by genes and environment
    Indeed, technologies
    ranging from linkage equilibrium and candidate gene
    association studies to genome wide associations have provided
    an extensive list of disease-gene associations, o
    ering
    us detailed information on mutations, SNPs, and the associated
    likelihood of developing speci c disease phenotypes
    [4].
  3. combinatorial
    relating to or involving combinations
    However,
    the combinatorial problem generated by the di
    erent
    disease factors and the previous medical history of a patient
    is so complex that no single health care professional
    can fully comprehend it all.
  4. iterative
    marked by repetition
    We also describe an Iterative version,
    ICARE, which incorporates ensemble concepts for improved
    performance.
  5. medical history
    the case history of a medical patient as recalled by the patient
    We propose CARE,
    a Collaborative Assessment and Recommendation Engine,
    which relies only on a patient's medical history using ICD-
    9-CM codes in order to predict future diseases risks.
  6. information system
    system consisting of the network of all communication channels used within an organization
    Categories and Subject Descriptors
    I.2.1 [Arti cial Intelligence]: Applications and Expert
    Systems|Medicine and science; I.5.1 [Pattern Recogni-
    tion]: Models|statistical ; J.3 [Computer Applications]:
    Life and Medical Sciences|Medical information systems
    General Terms
    Algorithms, Experimentation
    Keywords
    collaborative ltering, disease risk prediction, ensemble, prospective
    health care
    Corresponding Author
    Permission to make digital or hard copies of all or part of this w...
  7. hard copy
    (computer science) matter that is held in a computer and is typed or printed on paper
    Categories and Subject Descriptors
    I.2.1 [Arti cial Intelligence]: Applications and Expert
    Systems|Medicine and science; I.5.1 [Pattern Recogni-
    tion]: Models|statistical ; J.3 [Computer Applications]:
    Life and Medical Sciences|Medical information systems
    General Terms
    Algorithms, Experimentation
    Keywords
    collaborative ltering, disease risk prediction, ensemble, prospective
    health care
    Corresponding Author
    Permission to make digital or hard copies of all or part of this w...
  8. baby boomer
    a member of the baby boom generation in the 1950s
    The rapidly increasing
    medical concerns of the baby boomer generation is one major
    factor stressing the health care system.
  9. reactive
    participating in processes changing substances into others
    Health care, thus, needs to become more proactive than
    reactive in recognizing the onset of disease and risk.
  10. collaborative
    accomplished by working jointly
    We propose CARE,
    a Collaborative Assessment and Recommendation Engine,
    which relies only on a patient's medical history using ICD-
    9-CM codes in order to predict future diseases risks.
  11. proactive
    causing something to happen rather than waiting to respond
    Health care, thus, needs to become more proactive than
    reactive in recognizing the onset of disease and risk.
  12. mutation
    a change or alteration in form or qualities
    Indeed, technologies
    ranging from linkage equilibrium and candidate gene
    association studies to genome wide associations have provided
    an extensive list of disease-gene associations, o
    ering
    us detailed information on mutations, SNPs, and the associated
    likelihood of developing speci c disease phenotypes
    [4].
  13. republish
    publish again
    To copy otherwise, to
    republish, to post on servers or to redistribute to lists, requires prior specific
    permission and/or a fee.
  14. redistribute
    disperse anew
    To copy otherwise, to
    republish, to post on servers or to redistribute to lists, requires prior specific
    permission and/or a fee.
  15. predictive
    relating to prediction
    The underlying hypothesis behind this line of research
    is that once we catalogue all disease-related mutations, we
    will be able to predict the susceptibility of each individual
    to future diseases using various molecular biomarkers, ushering
    us into an era of predictive medicine.
  16. genome
    the full DNA sequence of an organism
    Today the prevailing model of prospective heath care is
    rmly based on the genome revolution.
  17. preventative
    tending to hinder
    This
    crisis has motivated the drive towards preventative medicine,
    where the primary concern is recognizing disease risk and
    taking action at the earliest signs.
  18. algorithm
    a precise rule specifying how to solve some problem
    Categories and Subject Descriptors
    I.2.1 [Arti cial Intelligence]: Applications and Expert
    Systems|Medicine and science; I.5.1 [Pattern Recogni-
    tion]: Models|statistical ; J.3 [Computer Applications]:
    Life and Medical Sciences|Medical information systems
    General Terms
    Algorithms, Experimentation
    Keywords
    collaborative ltering, disease risk prediction, ensemble, prospective
    health care
    Corresponding Author
    Permission to make digital or hard copies of all or part of this w...
  19. napa
    plant with an elongated head of broad stalked leaves resembling celery; used as a vegetable in east Asia
    CIKM’08, October 26-30, 2008, Napa Valley, California, USA.
  20. linkage
    the act of connecting things
    Indeed, technologies
    ranging from linkage equilibrium and candidate gene
    association studies to genome wide associations have provided
    an extensive list of disease-gene associations, o
    ering
    us detailed information on mutations, SNPs, and the associated
    likelihood of developing speci c disease phenotypes
    [4].
  21. motivate
    give an incentive for action
    This
    crisis has motivated the drive towards preventative medicine,
    where the primary concern is recognizing disease risk and
    taking action at the earliest signs.
  22. boomer
    a member of the large generation born in the 1950s
    The rapidly increasing
    medical concerns of the baby boomer generation is one major
    factor stressing the health care system.
  23. qualitative
    involving distinguishing attributes
    However, these
    sporadic and qualitative `risk assessments' generally focus on
    only a few diseases and are limited by a particular doctor's
    experience, memory, and time.
  24. predict
    make a guess about what will happen in the future
    Predicting Individual Disease Risk Based on Medical
    History
    Darcy A. Davis
    University of Notre Dame
    [email protected]
    Nitesh V. Chawla
    University of Notre Dame
    [email protected]
    Nicholas Blumm
    Northeastern University
    [email protected]
    Nicholas Christakis
    Harvard Medical School
    [email protected]
    Albert-László Barabási
    Northeastern University
    [email protected]
    ABSTRACT
    The monumental cost of health care, especially for chronic
    disease treatment, is quickly becoming u...
  25. underlie
    be underneath
    The underlying hypothesis behind this line of research
    is that once we catalogue all disease-related mutations, we
    will be able to predict the susceptibility of each individual
    to future diseases using various molecular biomarkers, ushering
    us into an era of predictive medicine.
  26. specialize
    become more focused on an area of activity or field of study
    These novel systems require no specialized information
    and provide predictions for medical conditions of
    all kinds in a single run.
  27. disease
    an impairment of health
    Predicting Individual Disease Risk Based on Medical
    History
    Darcy A. Davis
    University of Notre Dame
    [email protected]
    Nitesh V. Chawla
    University of Notre Dame
    [email protected]
    Nicholas Blumm
    Northeastern University
    [email protected]
    Nicholas Christakis
    Harvard Medical School
    [email protected]
    Albert-László Barabási
    Northeastern University
    [email protected]
    ABSTRACT
    The monumental cost of health care, especially for chronic
    disease treatment, is quickly becoming u...
  28. citation
    an official award usually given as formal public statement
    Categories and Subject Descriptors
    I.2.1 [Arti cial Intelligence]: Applications and Expert
    Systems|Medicine and science; I.5.1 [Pattern Recogni-
    tion]: Models|statistical ; J.3 [Computer Applications]:
    Life and Medical Sciences|Medical information systems
    General Terms
    Algorithms, Experimentation
    Keywords
    collaborative ltering, disease risk prediction, ensemble, prospective
    health care
    Corresponding Author
    Permission to make digital or hard copies of all or part of this work for...
  29. health care
    the preservation of mental and physical health by preventing or treating illness through services offered by the health profession
    Predicting Individual Disease Risk Based on Medical
    History
    Darcy A. Davis
    University of Notre Dame
    [email protected]
    Nitesh V. Chawla
    University of Notre Dame
    [email protected]
    Nicholas Blumm
    Northeastern University
    [email protected]
    Nicholas Christakis
    Harvard Medical School
    [email protected]
    Albert-László Barabási
    Northeastern University
    [email protected]
    ABSTRACT
    The monumental cost of health care, especially for chronic
    disease treatment, is quickly becoming u...
  30. northeastern
    situated in or oriented toward the northeast
    Predicting Individual Disease Risk Based on Medical
    History
    Darcy A. Davis
    University of Notre Dame
    [email protected]
    Nitesh V. Chawla
    University of Notre Dame
    [email protected]
    Nicholas Blumm
    Northeastern University
    [email protected]
    Nicholas Christakis
    Harvard Medical School
    [email protected]
    Albert-László Barabási
    Northeastern University
    [email protected]
    ABSTRACT
    The monumental cost of health care, especially for chronic
    disease treatment, is quickly becoming u...
  31. medical science
    the science of dealing with the maintenance of health and the prevention and treatment of disease
    Categories and Subject Descriptors
    I.2.1 [Arti cial Intelligence]: Applications and Expert
    Systems|Medicine and science; I.5.1 [Pattern Recogni-
    tion]: Models|statistical ; J.3 [Computer Applications]:
    Life and Medical Sciences|Medical information systems
    General Terms
    Algorithms, Experimentation
    Keywords
    collaborative ltering, disease risk prediction, ensemble, prospective
    health care
    Corresponding Author
    Permission to make digital or hard copies of all or part of this w...
  32. CDC
    a federal agency in the Department of Health and Human Services; located in Atlanta; investigates and diagnoses and tries to control or prevent diseases (especially new and unusual diseases)
    A CDC study
    estimates that 880.5 million visits were made to physician
    oces, about 3.1 visits per patient, in 2001 [3].
  33. ensemble
    an assemblage of parts considered as forming a whole
    We also describe an Iterative version,
    ICARE, which incorporates ensemble concepts for improved
    performance.
  34. experimentation
    the testing of an idea
    Categories and Subject Descriptors
    I.2.1 [Arti cial Intelligence]: Applications and Expert
    Systems|Medicine and science; I.5.1 [Pattern Recogni-
    tion]: Models|statistical ; J.3 [Computer Applications]:
    Life and Medical Sciences|Medical information systems
    General Terms
    Algorithms, Experimentation
    Keywords
    collaborative ltering, disease risk prediction, ensemble, prospective
    health care
    Corresponding Author
    Permission to make digital or hard copies of all or part of this w...
  35. server
    a person who waits on tables in a restaurant
    To copy otherwise, to
    republish, to post on servers or to redistribute to lists, requires prior specific
    permission and/or a fee.
  36. medical
    relating to the study or practice of medicine
    Predicting Individual Disease Risk Based on Medical
    History
    Darcy A. Davis
    University of Notre Dame
    [email protected]
    Nitesh V. Chawla
    University of Notre Dame
    [email protected]
    Nicholas Blumm
    Northeastern University
    [email protected]
    Nicholas Christakis
    Harvard Medical School
    [email protected]
    Albert-László Barabási
    Northeastern University
    [email protected]
    ABSTRACT
    The monumental cost of health care, especially for chronic
    disease treatment, is quickly becoming u...
  37. sporadic
    recurring in scattered or unpredictable instances
    However, these
    sporadic and qualitative `risk assessments' generally focus on
    only a few diseases and are limited by a particular doctor's
    experience, memory, and time.
  38. medical care
    professional treatment for illness or injury
    INTRODUCTION
    Medical care and research are literally the most vital part
    of science for humans, as none of us are immune to physical
    ailments and biological deterioration.
  39. unmanageable
    hard to control
    Predicting Individual Disease Risk Based on Medical
    History
    Darcy A. Davis
    University of Notre Dame
    [email protected]
    Nitesh V. Chawla
    University of Notre Dame
    [email protected]
    Nicholas Blumm
    Northeastern University
    [email protected]
    Nicholas Christakis
    Harvard Medical School
    [email protected]
    Albert-László Barabási
    Northeastern University
    [email protected]
    ABSTRACT
    The monumental cost of health care, especially for chronic
    disease treatment, is quickly becoming unman...
  40. prediction
    a statement made about the future
    These novel systems require no specialized information
    and provide predictions for medical conditions of
    all kinds in a single run.
  41. susceptibility
    the state of being easily affected
    The underlying hypothesis behind this line of research
    is that once we catalogue all disease-related mutations, we
    will be able to predict the susceptibility of each individual
    to future diseases using various molecular biomarkers, ushering
    us into an era of predictive medicine.
  42. unravel
    become or cause to become undone by separating the fibers of
    Yet, these rapid
    advances have also unraveled the limitations of the genome
    based approaches [12].
  43. prospective
    of or concerned with or related to the future
    Categories and Subject Descriptors
    I.2.1 [Arti cial Intelligence]: Applications and Expert
    Systems|Medicine and science; I.5.1 [Pattern Recogni-
    tion]: Models|statistical ; J.3 [Computer Applications]:
    Life and Medical Sciences|Medical information systems
    General Terms
    Algorithms, Experimentation
    Keywords
    collaborative ltering, disease risk prediction, ensemble, prospective
    health care
    Corresponding Author
    Permission to make digital or hard copies of all or part of this w...
  44. usher
    someone employed to conduct others
    The underlying hypothesis behind this line of research
    is that once we catalogue all disease-related mutations, we
    will be able to predict the susceptibility of each individual
    to future diseases using various molecular biomarkers, ushering
    us into an era of predictive medicine.
  45. ailment
    an often persistent bodily disorder or disease
    INTRODUCTION
    Medical care and research are literally the most vital part
    of science for humans, as none of us are immune to physical
    ailments and biological deterioration.
  46. incorporate
    make into a whole or make part of a whole
    We also describe an Iterative version,
    ICARE, which incorporates ensemble concepts for improved
    performance.
  47. gene
    part of DNA controlling physical characteristics and growth
    Indeed, technologies
    ranging from linkage equilibrium and candidate gene
    association studies to genome wide associations have provided
    an extensive list of disease-gene associations, o
    ering
    us detailed information on mutations, SNPs, and the associated
    likelihood of developing speci c disease phenotypes
    [4].
  48. assessment
    the act of judging a person or situation or event
    We propose CARE,
    a Collaborative Assessment and Recommendation Engine,
    which relies only on a patient's medical history using ICD-
    9-CM codes in order to predict future diseases risks.
  49. medical school
    a graduate school offering study leading to a medical degree
    Predicting Individual Disease Risk Based on Medical
    History
    Darcy A. Davis
    University of Notre Dame
    [email protected]
    Nitesh V. Chawla
    University of Notre Dame
    [email protected]
    Nicholas Blumm
    Northeastern University
    [email protected]
    Nicholas Christakis
    Harvard Medical School
    [email protected]
    Albert-László Barabási
    Northeastern University
    [email protected]
    ABSTRACT
    The monumental cost of health care, especially for chronic
    disease treatment, is quickly becoming u...
  50. deterioration
    the process of changing to an inferior state
    INTRODUCTION
    Medical care and research are literally the most vital part
    of science for humans, as none of us are immune to physical
    ailments and biological deterioration.
  51. molecular
    relating to the simplest units of an element or compound
    The underlying hypothesis behind this line of research
    is that once we catalogue all disease-related mutations, we
    will be able to predict the susceptibility of each individual
    to future diseases using various molecular biomarkers, ushering
    us into an era of predictive medicine.
  52. assess
    estimate the nature, quality, ability or significance of
    Currently, physicians can use
    family and health history and physical examination to approximate
    the risk of a patient, guiding laboratory tests to
    further assess the patient's stage of health.
  53. risk
    a source of danger
    Predicting Individual Disease Risk Based on Medical
    History
    Darcy A. Davis
    University of Notre Dame
    [email protected]
    Nitesh V. Chawla
    University of Notre Dame
    [email protected]
    Nicholas Blumm
    Northeastern University
    [email protected]
    Nicholas Christakis
    Harvard Medical School
    [email protected]
    Albert-László Barabási
    Northeastern University
    [email protected]
    ABSTRACT
    The monumental cost of health care, especially for chronic
    disease treatment, is quickly becoming u...
  54. motivated
    strongly driven to succeed or achieve something
    This
    crisis has motivated the drive towards preventative medicine,
    where the primary concern is recognizing disease risk and
    taking action at the earliest signs.
  55. chronic
    long-lasting or characterized by long suffering
    Predicting Individual Disease Risk Based on Medical
    History
    Darcy A. Davis
    University of Notre Dame
    [email protected]
    Nitesh V. Chawla
    University of Notre Dame
    [email protected]
    Nicholas Blumm
    Northeastern University
    [email protected]
    Nicholas Christakis
    Harvard Medical School
    [email protected]
    Albert-László Barabási
    Northeastern University
    [email protected]
    ABSTRACT
    The monumental cost of health care, especially for chronic
    disease treatment, is quickly becoming u...
  56. overwhelm
    overcome, as with emotions or perceptual stimuli
    Annual health
    care expenditure in the U.S. alone is an overwhelming sum,
    with a strong majority of this money used for chronic disease
    treatment.
  57. approximate
    not quite exact or correct
    Currently, physicians can use
    family and health history and physical examination to approximate
    the risk of a patient, guiding laboratory tests to
    further assess the patient's stage of health.
  58. patient
    enduring trying circumstances with even temper
    We propose CARE,
    a Collaborative Assessment and Recommendation Engine,
    which relies only on a patient's medical history using ICD-
    9-CM codes in order to predict future diseases risks.
  59. classroom
    a room in a school where lessons take place
    Categories and Subject Descriptors
    I.2.1 [Arti cial Intelligence]: Applications and Expert
    Systems|Medicine and science; I.5.1 [Pattern Recogni-
    tion]: Models|statistical ; J.3 [Computer Applications]:
    Life and Medical Sciences|Medical information systems
    General Terms
    Algorithms, Experimentation
    Keywords
    collaborative ltering, disease risk prediction, ensemble, prospective
    health care
    Corresponding Author
    Permission to make digital or hard copies of all or part of this work for...
  60. statistical
    of or relating to the interpretation of quantitative data
    Categories and Subject Descriptors
    I.2.1 [Arti cial Intelligence]: Applications and Expert
    Systems|Medicine and science; I.5.1 [Pattern Recogni-
    tion]: Models|statistical ; J.3 [Computer Applications]:
    Life and Medical Sciences|Medical information systems
    General Terms
    Algorithms, Experimentation
    Keywords
    collaborative ltering, disease risk prediction, ensemble, prospective
    health care
    Corresponding Author
    Permission to make digital or hard copies of all or part of this w...
  61. immune
    of the condition in which an organism can resist disease
    INTRODUCTION
    Medical care and research are literally the most vital part
    of science for humans, as none of us are immune to physical
    ailments and biological deterioration.
  62. monumental
    of outstanding significance
    Predicting Individual Disease Risk Based on Medical
    History
    Darcy A. Davis
    University of Notre Dame
    [email protected]
    Nitesh V. Chawla
    University of Notre Dame
    [email protected]
    Nicholas Blumm
    Northeastern University
    [email protected]
    Nicholas Christakis
    Harvard Medical School
    [email protected]
    Albert-László Barabási
    Northeastern University
    [email protected]
    ABSTRACT
    The monumental cost of health care, especially for chronic
    disease treatment, is quickly becoming u...
  63. onset
    the beginning or early stages
    Health care, thus, needs to become more proactive than
    reactive in recognizing the onset of disease and risk.
  64. specialized
    developed or designed for a particular activity or function
    These novel systems require no specialized information
    and provide predictions for medical conditions of
    all kinds in a single run.
  65. symptom
    a sensation associated with a particular disease
    Therefore, current medical
    care is reactive, stepping in once the symptoms of a disease
    have emerged, rather than proactive, treating or eliminating
    a disease at the earliest signs.
  66. equilibrium
    a stable situation in which forces cancel one another
    Indeed, technologies
    ranging from linkage equilibrium and candidate gene
    association studies to genome wide associations have provided
    an extensive list of disease-gene associations, o
    ering
    us detailed information on mutations, SNPs, and the associated
    likelihood of developing speci c disease phenotypes
    [4].
  67. likelihood
    the probability of a specified outcome
    Indeed, technologies
    ranging from linkage equilibrium and candidate gene
    association studies to genome wide associations have provided
    an extensive list of disease-gene associations, o
    ering
    us detailed information on mutations, SNPs, and the associated
    likelihood of developing speci c disease phenotypes
    [4].
  68. fee
    a fixed charge for a privilege or for professional services
    Categories and Subject Descriptors
    I.2.1 [Arti cial Intelligence]: Applications and Expert
    Systems|Medicine and science; I.5.1 [Pattern Recogni-
    tion]: Models|statistical ; J.3 [Computer Applications]:
    Life and Medical Sciences|Medical information systems
    General Terms
    Algorithms, Experimentation
    Keywords
    collaborative ltering, disease risk prediction, ensemble, prospective
    health care
    Corresponding Author
    Permission to make digital or hard copies of all or part of this work for...
  69. generate
    bring into existence
    However,
    the combinatorial problem generated by the di
    erent
    disease factors and the previous medical history of a patient
    is so complex that no single health care professional
    can fully comprehend it all.
  70. Medicare
    federally administered health care for older individuals
    We present experimental results on
    a Medicare dataset, demonstrating that CARE and ICARE
    perform well at capturing future disease risks.
  71. biological
    pertaining to life and living things
    INTRODUCTION
    Medical care and research are literally the most vital part
    of science for humans, as none of us are immune to physical
    ailments and biological deterioration.
  72. base
    lowest support of a structure
    Predicting Individual Disease Risk Based on Medical
    History
    Darcy A. Davis
    University of Notre Dame
    [email protected]
    Nitesh V. Chawla
    University of Notre Dame
    [email protected]
    Nicholas Blumm
    Northeastern University
    [email protected]
    Nicholas Christakis
    Harvard Medical School
    [email protected]
    Albert-László Barabási
    Northeastern University
    [email protected]
    ABSTRACT
    The monumental cost of health care, especially for chronic
    disease treatment, is quickly becoming u...
  73. limitation
    an act of restricting (as by regulation)
    Yet, these rapid
    advances have also unraveled the limitations of the genome
    based approaches [12].
  74. eliminate
    end, take out, or do away with
    Therefore, current medical
    care is reactive, stepping in once the symptoms of a disease
    have emerged, rather than proactive, treating or eliminating
    a disease at the earliest signs.
  75. associate
    bring or come into action
    Indeed, technologies
    ranging from linkage equilibrium and candidate gene
    association studies to genome wide associations have provided
    an extensive list of disease-gene associations, o
    ering
    us detailed information on mutations, SNPs, and the associated
    likelihood of developing speci c disease phenotypes
    [4].
  76. health
    the general condition of body and mind
    Predicting Individual Disease Risk Based on Medical
    History
    Darcy A. Davis
    University of Notre Dame
    [email protected]
    Nitesh V. Chawla
    University of Notre Dame
    [email protected]
    Nicholas Blumm
    Northeastern University
    [email protected]
    Nicholas Christakis
    Harvard Medical School
    [email protected]
    Albert-László Barabási
    Northeastern University
    [email protected]
    ABSTRACT
    The monumental cost of health care, especially for chronic
    disease treatment, is quickly becoming u...
  77. care
    providing treatment for or attending to someone or something
    Predicting Individual Disease Risk Based on Medical
    History
    Darcy A. Davis
    University of Notre Dame
    [email protected]
    Nitesh V. Chawla
    University of Notre Dame
    [email protected]
    Nicholas Blumm
    Northeastern University
    [email protected]
    Nicholas Christakis
    Harvard Medical School
    [email protected]
    Albert-László Barabási
    Northeastern University
    [email protected]
    ABSTRACT
    The monumental cost of health care, especially for chronic
    disease treatment, is quickly becoming u...
  78. humans
    all of the living human inhabitants of the earth
    INTRODUCTION
    Medical care and research are literally the most vital part
    of science for humans, as none of us are immune to physical
    ailments and biological deterioration.
  79. correspond
    take the place of or be parallel or equivalent to
    Categories and Subject Descriptors
    I.2.1 [Arti cial Intelligence]: Applications and Expert
    Systems|Medicine and science; I.5.1 [Pattern Recogni-
    tion]: Models|statistical ; J.3 [Computer Applications]:
    Life and Medical Sciences|Medical information systems
    General Terms
    Algorithms, Experimentation
    Keywords
    collaborative ltering, disease risk prediction, ensemble, prospective
    health care
    Corresponding Author
    Permission to make digital or hard copies of all or part of this w...
  80. Nicholas
    a bishop in Asia Minor who is associated with Santa Claus
    Predicting Individual Disease Risk Based on Medical
    History
    Darcy A. Davis
    University of Notre Dame
    [email protected]
    Nitesh V. Chawla
    University of Notre Dame
    [email protected]
    Nicholas Blumm
    Northeastern University
    [email protected]
    Nicholas Christakis
    Harvard Medical School
    [email protected]
    Albert-László Barabási
    Northeastern University
    [email protected]
    ABSTRACT
    The monumental cost of health care, especially for chronic
    disease treatment, is quickly becoming u...
  81. guiding
    showing the way by conducting or leading
    Currently, physicians can use
    family and health history and physical examination to approximate
    the risk of a patient, guiding laboratory tests to
    further assess the patient's stage of health.
  82. copy
    a thing made to be similar or identical to another thing
    Categories and Subject Descriptors
    I.2.1 [Arti cial Intelligence]: Applications and Expert
    Systems|Medicine and science; I.5.1 [Pattern Recogni-
    tion]: Models|statistical ; J.3 [Computer Applications]:
    Life and Medical Sciences|Medical information systems
    General Terms
    Algorithms, Experimentation
    Keywords
    collaborative ltering, disease risk prediction, ensemble, prospective
    health care
    Corresponding Author
    Permission to make digital or hard copies of all or part of this w...
  83. factor
    anything that contributes causally to a result
    The rapidly increasing
    medical concerns of the baby boomer generation is one major
    factor stressing the health care system.
  84. medicine
    the profession devoted to alleviating diseases and injuries
    This
    crisis has motivated the drive towards preventative medicine,
    where the primary concern is recognizing disease risk and
    taking action at the earliest signs.
  85. underlying
    in the nature of something though not readily apparent
    The underlying hypothesis behind this line of research
    is that once we catalogue all disease-related mutations, we
    will be able to predict the susceptibility of each individual
    to future diseases using various molecular biomarkers, ushering
    us into an era of predictive medicine.
  86. based
    having a base
    Predicting Individual Disease Risk Based on Medical
    History
    Darcy A. Davis
    University of Notre Dame
    [email protected]
    Nitesh V. Chawla
    University of Notre Dame
    [email protected]
    Nicholas Blumm
    Northeastern University
    [email protected]
    Nicholas Christakis
    Harvard Medical School
    [email protected]
    Albert-László Barabási
    Northeastern University
    [email protected]
    ABSTRACT
    The monumental cost of health care, especially for chronic
    disease treatment, is quickly becoming u...
  87. expert
    a person with special knowledge who performs skillfully
    Categories and Subject Descriptors
    I.2.1 [Arti cial Intelligence]: Applications and Expert
    Systems|Medicine and science; I.5.1 [Pattern Recogni-
    tion]: Models|statistical ; J.3 [Computer Applications]:
    Life and Medical Sciences|Medical information systems
    General Terms
    Algorithms, Experimentation
    Keywords
    collaborative ltering, disease risk prediction, ensemble, prospective
    health care
    Corresponding Author
    Permission to make digital or hard copies of all or part of this w...
  88. dame
    a woman of refinement
    Predicting Individual Disease Risk Based on Medical
    History
    Darcy A. Davis
    University of Notre Dame
    [email protected]
    Nitesh V. Chawla
    University of Notre Dame
    [email protected]
    Nicholas Blumm
    Northeastern University
    [email protected]
    Nicholas Christakis
    Harvard Medical School
    [email protected]
    Albert-László Barabási
    Northeastern University
    [email protected]
    ABSTRACT
    The monumental cost of health care, especially for chronic
    disease treatment, is quickly becoming u...
  89. category
    a general concept that marks divisions or coordinations
    Categories and Subject Descriptors
    I.2.1 [Arti cial Intelligence]: Applications and Expert
    Systems|Medicine and science; I.5.1 [Pattern Recogni-
    tion]: Models|statistical ; J.3 [Computer Applications]:
    Life and Medical Sciences|Medical information systems
    General Terms
    Algorithms, Experimentation
    Keywords
    collaborative ltering, disease risk prediction, ensemble, prospective
    health care
    Corresponding Author
    Permission to make digital or hard copies of all or part of this w...
  90. focus on
    center upon
    However, these
    sporadic and qualitative `risk assessments' generally focus on
    only a few diseases and are limited by a particular doctor's
    experience, memory, and time.
  91. emerge
    come out into view, as from concealment
    Therefore, current medical
    care is reactive, stepping in once the symptoms of a disease
    have emerged, rather than proactive, treating or eliminating
    a disease at the earliest signs.
  92. demonstrate
    give an exhibition of to an interested audience
    We present experimental results on
    a Medicare dataset, demonstrating that CARE and ICARE
    perform well at capturing future disease risks.
  93. distribute
    give to several people
    Categories and Subject Descriptors
    I.2.1 [Arti cial Intelligence]: Applications and Expert
    Systems|Medicine and science; I.5.1 [Pattern Recogni-
    tion]: Models|statistical ; J.3 [Computer Applications]:
    Life and Medical Sciences|Medical information systems
    General Terms
    Algorithms, Experimentation
    Keywords
    collaborative ltering, disease risk prediction, ensemble, prospective
    health care
    Corresponding Author
    Permission to make digital or hard copies of all or part of this work for...
  94. concept
    an abstract or general idea inferred from specific instances
    We also describe an Iterative version,
    ICARE, which incorporates ensemble concepts for improved
    performance.
  95. experimental
    of the nature of or undergoing a trial
    We present experimental results on
    a Medicare dataset, demonstrating that CARE and ICARE
    perform well at capturing future disease risks.
  96. ranging
    wandering freely
    Indeed, technologies
    ranging from linkage equilibrium and candidate gene
    association studies to genome wide associations have provided
    an extensive list of disease-gene associations, o
    ering
    us detailed information on mutations, SNPs, and the associated
    likelihood of developing speci c disease phenotypes
    [4].
  97. step in
    act as a substitute
    Therefore, current medical
    care is reactive, stepping in once the symptoms of a disease
    have emerged, rather than proactive, treating or eliminating
    a disease at the earliest signs.
  98. recommendation
    praise of a person or thing as worthy or desirable
    We propose CARE,
    a Collaborative Assessment and Recommendation Engine,
    which relies only on a patient's medical history using ICD-
    9-CM codes in order to predict future diseases risks.
  99. heath
    a low evergreen shrub of the family Ericaceae
    Today the prevailing model of prospective heath care is
    rmly based on the genome revolution.
  100. earliest
    more early than; most early
    This
    crisis has motivated the drive towards preventative medicine,
    where the primary concern is recognizing disease risk and
    taking action at the earliest signs.
  101. expenditure
    the act of spending money for goods or services
    Annual health
    care expenditure in the U.S. alone is an overwhelming sum,
    with a strong majority of this money used for chronic disease
    treatment.
  102. hypothesis
    a tentative insight that is not yet verified or tested
    The underlying hypothesis behind this line of research
    is that once we catalogue all disease-related mutations, we
    will be able to predict the susceptibility of each individual
    to future diseases using various molecular biomarkers, ushering
    us into an era of predictive medicine.
  103. USA
    North American republic containing 50 states - 48 conterminous states in North America plus Alaska in northwest North America and the Hawaiian Islands in the Pacific Ocean; achieved independence in 1776
    CIKM’08, October 26-30, 2008, Napa Valley, California, USA.
  104. increase
    a process of becoming larger or longer or more numerous
    Experts expect the burden on the system to
    continually increase in coming years.
  105. provide
    give something useful or necessary to
    These novel systems require no specialized information
    and provide predictions for medical conditions of
    all kinds in a single run.
  106. catalogue
    a complete list of things, usually arranged systematically
    The underlying hypothesis behind this line of research
    is that once we catalogue all disease-related mutations, we
    will be able to predict the susceptibility of each individual
    to future diseases using various molecular biomarkers, ushering
    us into an era of predictive medicine.
  107. physician
    a licensed medical practitioner
    A CDC study
    estimates that 880.5 million visits were made to physician
    oces, about 3.1 visits per patient, in 2001 [3].
  108. prevailing
    most frequent or common
    Today the prevailing model of prospective heath care is
    rmly based on the genome revolution.
  109. recognize
    perceive to be something or something you can identify
    This
    crisis has motivated the drive towards preventative medicine,
    where the primary concern is recognizing disease risk and
    taking action at the earliest signs.
  110. association
    a formal organization of people or groups of people
    Indeed, technologies
    ranging from linkage equilibrium and candidate gene
    association studies to genome wide associations have provided
    an extensive list of disease-gene associations, o
    ering
    us detailed information on mutations, SNPs, and the associated
    likelihood of developing speci c disease phenotypes
    [4].
  111. application
    the action of putting something into operation
    Categories and Subject Descriptors
    I.2.1 [Arti cial Intelligence]: Applications and Expert
    Systems|Medicine and science; I.5.1 [Pattern Recogni-
    tion]: Models|statistical ; J.3 [Computer Applications]:
    Life and Medical Sciences|Medical information systems
    General Terms
    Algorithms, Experimentation
    Keywords
    collaborative ltering, disease risk prediction, ensemble, prospective
    health care
    Corresponding Author
    Permission to make digital or hard copies of all or part of this w...
  112. testing
    experimentation to determine how well something works
    However, universal testing
    is neither time nor cost ecient.
  113. laboratory
    a workplace for the conduct of scientific research
    Currently, physicians can use
    family and health history and physical examination to approximate
    the risk of a patient, guiding laboratory tests to
    further assess the patient's stage of health.
  114. prevail
    be larger in number, quantity, power, status or importance
    Today the prevailing model of prospective heath care is
    rmly based on the genome revolution.
  115. university
    an institution of higher learning that grants degrees
    Predicting Individual Disease Risk Based on Medical
    History
    Darcy A. Davis
    University of Notre Dame
    [email protected]
    Nitesh V. Chawla
    University of Notre Dame
    [email protected]
    Nicholas Blumm
    Northeastern University
    [email protected]
    Nicholas Christakis
    Harvard Medical School
    [email protected]
    Albert-László Barabási
    Northeastern University
    [email protected]
    ABSTRACT
    The monumental cost of health care, especially for chronic
    disease treatment, is quickly becoming u...
  116. era
    a period marked by distinctive character
    The underlying hypothesis behind this line of research
    is that once we catalogue all disease-related mutations, we
    will be able to predict the susceptibility of each individual
    to future diseases using various molecular biomarkers, ushering
    us into an era of predictive medicine.
  117. model
    a representation of something, often on a smaller scale
    Categories and Subject Descriptors
    I.2.1 [Arti cial Intelligence]: Applications and Expert
    Systems|Medicine and science; I.5.1 [Pattern Recogni-
    tion]: Models|statistical ; J.3 [Computer Applications]:
    Life and Medical Sciences|Medical information systems
    General Terms
    Algorithms, Experimentation
    Keywords
    collaborative ltering, disease risk prediction, ensemble, prospective
    health care
    Corresponding Author
    Permission to make digital or hard copies of all or part of this w...
  118. permission
    approval to do something
    Categories and Subject Descriptors
    I.2.1 [Arti cial Intelligence]: Applications and Expert
    Systems|Medicine and science; I.5.1 [Pattern Recogni-
    tion]: Models|statistical ; J.3 [Computer Applications]:
    Life and Medical Sciences|Medical information systems
    General Terms
    Algorithms, Experimentation
    Keywords
    collaborative ltering, disease risk prediction, ensemble, prospective
    health care
    Corresponding Author
    Permission to make digital or hard copies of all or part of this w...
  119. overwhelming
    very intense
    Annual health
    care expenditure in the U.S. alone is an overwhelming sum,
    with a strong majority of this money used for chronic disease
    treatment.
  120. digital
    relating to or performed with the fingers
    Categories and Subject Descriptors
    I.2.1 [Arti cial Intelligence]: Applications and Expert
    Systems|Medicine and science; I.5.1 [Pattern Recogni-
    tion]: Models|statistical ; J.3 [Computer Applications]:
    Life and Medical Sciences|Medical information systems
    General Terms
    Algorithms, Experimentation
    Keywords
    collaborative ltering, disease risk prediction, ensemble, prospective
    health care
    Corresponding Author
    Permission to make digital or hard copies of all or part of this w...
  121. rely
    have confidence or faith in
    We propose CARE,
    a Collaborative Assessment and Recommendation Engine,
    which relies only on a patient's medical history using ICD-
    9-CM codes in order to predict future diseases risks.
  122. concern
    something that interests you because it is important
    This
    crisis has motivated the drive towards preventative medicine,
    where the primary concern is recognizing disease risk and
    taking action at the earliest signs.
  123. comprehend
    get the meaning of something
    However,
    the combinatorial problem generated by the di
    erent
    disease factors and the previous medical history of a patient
    is so complex that no single health care professional
    can fully comprehend it all.
  124. abstract
    existing only in the mind
    Predicting Individual Disease Risk Based on Medical
    History
    Darcy A. Davis
    University of Notre Dame
    [email protected]
    Nitesh V. Chawla
    University of Notre Dame
    [email protected]
    Nicholas Blumm
    Northeastern University
    [email protected]
    Nicholas Christakis
    Harvard Medical School
    [email protected]
    Albert-László Barabási
    Northeastern University
    [email protected]
    ABSTRACT
    The monumental cost of health care, especially for chronic
    disease treatment, is quickly becoming u...
  125. relate
    give an account of
    The underlying hypothesis behind this line of research
    is that once we catalogue all disease-related mutations, we
    will be able to predict the susceptibility of each individual
    to future diseases using various molecular biomarkers, ushering
    us into an era of predictive medicine.
  126. currently
    at this time or period
    Currently, physicians can use
    family and health history and physical examination to approximate
    the risk of a patient, guiding laboratory tests to
    further assess the patient's stage of health.
  127. system
    a group of independent elements comprising a unified whole
    These novel systems require no specialized information
    and provide predictions for medical conditions of
    all kinds in a single run.
  128. increasingly
    advancing in amount or intensity
    Given the weak signals that most disease associated SNPs
    or mutations o
    er, it is increasingly clear that the promise
    769
  129. stress
    special emphasis attached to something
    The rapidly increasing
    medical concerns of the baby boomer generation is one major
    factor stressing the health care system.
  130. detailed
    developed with careful treatment of particulars
    Indeed, technologies
    ranging from linkage equilibrium and candidate gene
    association studies to genome wide associations have provided
    an extensive list of disease-gene associations, o
    ering
    us detailed information on mutations, SNPs, and the associated
    likelihood of developing speci c disease phenotypes
    [4].
  131. corresponding
    similar especially in position or purpose
    Categories and Subject Descriptors
    I.2.1 [Arti cial Intelligence]: Applications and Expert
    Systems|Medicine and science; I.5.1 [Pattern Recogni-
    tion]: Models|statistical ; J.3 [Computer Applications]:
    Life and Medical Sciences|Medical information systems
    General Terms
    Algorithms, Experimentation
    Keywords
    collaborative ltering, disease risk prediction, ensemble, prospective
    health care
    Corresponding Author
    Permission to make digital or hard copies of all or part of this w...
  132. require
    have need of
    These novel systems require no specialized information
    and provide predictions for medical conditions of
    all kinds in a single run.
  133. research
    a seeking for knowledge
    INTRODUCTION
    Medical care and research are literally the most vital part
    of science for humans, as none of us are immune to physical
    ailments and biological deterioration.
  134. developing
    of societies with low levels of industrial capability
    Indeed, technologies
    ranging from linkage equilibrium and candidate gene
    association studies to genome wide associations have provided
    an extensive list of disease-gene associations, o
    ering
    us detailed information on mutations, SNPs, and the associated
    likelihood of developing speci c disease phenotypes
    [4].
  135. visit
    the act of going to see some person or place or thing for a short time
    A CDC study
    estimates that 880.5 million visits were made to physician
    oces, about 3.1 visits per patient, in 2001 [3].
  136. Harvard
    American philanthropist who left his library and half his estate to the Massachusetts college that now bears his name (1607-1638)
    Predicting Individual Disease Risk Based on Medical
    History
    Darcy A. Davis
    University of Notre Dame
    [email protected]
    Nitesh V. Chawla
    University of Notre Dame
    [email protected]
    Nicholas Blumm
    Northeastern University
    [email protected]
    Nicholas Christakis
    Harvard Medical School
    [email protected]
    Albert-László Barabási
    Northeastern University
    [email protected]
    ABSTRACT
    The monumental cost of health care, especially for chronic
    disease treatment, is quickly becoming u...
  137. science
    a branch of study or knowledge involving the observation, investigation, and discovery of general laws or truths that can be tested systematically
    Categories and Subject Descriptors
    I.2.1 [Arti cial Intelligence]: Applications and Expert
    Systems|Medicine and science; I.5.1 [Pattern Recogni-
    tion]: Models|statistical ; J.3 [Computer Applications]:
    Life and Medical Sciences|Medical information systems
    General Terms
    Algorithms, Experimentation
    Keywords
    collaborative ltering, disease risk prediction, ensemble, prospective
    health care
    Corresponding Author
    Permission to make digital or hard copies of all or part of this w...
  138. test
    standardized procedure for measuring sensitivity or aptitude
    However, universal testing
    is neither time nor cost ecient.
  139. distributed
    spread out or scattered about or divided up
    Categories and Subject Descriptors
    I.2.1 [Arti cial Intelligence]: Applications and Expert
    Systems|Medicine and science; I.5.1 [Pattern Recogni-
    tion]: Models|statistical ; J.3 [Computer Applications]:
    Life and Medical Sciences|Medical information systems
    General Terms
    Algorithms, Experimentation
    Keywords
    collaborative ltering, disease risk prediction, ensemble, prospective
    health care
    Corresponding Author
    Permission to make digital or hard copies of all or part of this work for...
  140. treatment
    the management of someone or something
    Predicting Individual Disease Risk Based on Medical
    History
    Darcy A. Davis
    University of Notre Dame
    [email protected]
    Nitesh V. Chawla
    University of Notre Dame
    [email protected]
    Nicholas Blumm
    Northeastern University
    [email protected]
    Nicholas Christakis
    Harvard Medical School
    [email protected]
    Albert-László Barabási
    Northeastern University
    [email protected]
    ABSTRACT
    The monumental cost of health care, especially for chronic
    disease treatment, is quickly becoming u...
  141. pattern
    a repeated design, structure, or arrangement
    Categories and Subject Descriptors
    I.2.1 [Arti cial Intelligence]: Applications and Expert
    Systems|Medicine and science; I.5.1 [Pattern Recogni-
    tion]: Models|statistical ; J.3 [Computer Applications]:
    Life and Medical Sciences|Medical information systems
    General Terms
    Algorithms, Experimentation
    Keywords
    collaborative ltering, disease risk prediction, ensemble, prospective
    health care
    Corresponding Author
    Permission to make digital or hard copies of all or part of this w...
  142. focus
    the concentration of attention or energy on something
    However, these
    sporadic and qualitative `risk assessments' generally focus on
    only a few diseases and are limited by a particular doctor's
    experience, memory, and time.
  143. history
    a record or narrative description of past events
    Predicting Individual Disease Risk Based on Medical
    History
    Darcy A. Davis
    University of Notre Dame
    [email protected]
    Nitesh V. Chawla
    University of Notre Dame
    [email protected]
    Nicholas Blumm
    Northeastern University
    [email protected]
    Nicholas Christakis
    Harvard Medical School
    [email protected]
    Albert-László Barabási
    Northeastern University
    [email protected]
    ABSTRACT
    The monumental cost of health care, especially for chronic
    disease treatment, is quickly becoming u...
  144. literally
    without exaggeration
    INTRODUCTION
    Medical care and research are literally the most vital part
    of science for humans, as none of us are immune to physical
    ailments and biological deterioration.
  145. computer
    a machine for performing calculations automatically
    Categories and Subject Descriptors
    I.2.1 [Arti cial Intelligence]: Applications and Expert
    Systems|Medicine and science; I.5.1 [Pattern Recogni-
    tion]: Models|statistical ; J.3 [Computer Applications]:
    Life and Medical Sciences|Medical information systems
    General Terms
    Algorithms, Experimentation
    Keywords
    collaborative ltering, disease risk prediction, ensemble, prospective
    health care
    Corresponding Author
    Permission to make digital or hard copies of all or part of this w...
  146. propose
    present for consideration, examination, or criticism
    We propose CARE,
    a Collaborative Assessment and Recommendation Engine,
    which relies only on a patient's medical history using ICD-
    9-CM codes in order to predict future diseases risks.
  147. specific
    stated explicitly or in detail
    To copy otherwise, to
    republish, to post on servers or to redistribute to lists, requires prior specific
    permission and/or a fee.
  148. Albert
    prince consort of Queen Victoria of England (1819-1861)
    Predicting Individual Disease Risk Based on Medical
    History
    Darcy A. Davis
    University of Notre Dame
    [email protected]
    Nitesh V. Chawla
    University of Notre Dame
    [email protected]
    Nicholas Blumm
    Northeastern University
    [email protected]
    Nicholas Christakis
    Harvard Medical School
    [email protected]
    Albert-László Barabási
    Northeastern University
    [email protected]
    ABSTRACT
    The monumental cost of health care, especially for chronic
    disease treatment, is quickly becoming u...
  149. information
    knowledge acquired through study or experience
    These novel systems require no specialized information
    and provide predictions for medical conditions of
    all kinds in a single run.
  150. complex
    complicated in structure
    However,
    the combinatorial problem generated by the di
    erent
    disease factors and the previous medical history of a patient
    is so complex that no single health care professional
    can fully comprehend it all.
  151. estimate
    judge tentatively
    A CDC study
    estimates that 880.5 million visits were made to physician
    oces, about 3.1 visits per patient, in 2001 [3].
  152. physical
    involving the body as distinguished from the mind or spirit
    INTRODUCTION
    Medical care and research are literally the most vital part
    of science for humans, as none of us are immune to physical
    ailments and biological deterioration.
  153. primary
    of first rank or importance or value
    This
    crisis has motivated the drive towards preventative medicine,
    where the primary concern is recognizing disease risk and
    taking action at the earliest signs.
  154. using
    an act that exploits or victimizes someone
    We propose CARE,
    a Collaborative Assessment and Recommendation Engine,
    which relies only on a patient's medical history using ICD-
    9-CM codes in order to predict future diseases risks.
  155. list
    a database containing an ordered array of items
    To copy otherwise, to
    republish, to post on servers or to redistribute to lists, requires prior specific
    permission and/or a fee.
  156. improve
    to make better
    We also describe an Iterative version,
    ICARE, which incorporates ensemble concepts for improved
    performance.
  157. develop
    progress or evolve through a process of natural growth
    Indeed, technologies
    ranging from linkage equilibrium and candidate gene
    association studies to genome wide associations have provided
    an extensive list of disease-gene associations, o
    ering
    us detailed information on mutations, SNPs, and the associated
    likelihood of developing speci c disease phenotypes
    [4].
  158. Davis
    American statesman
    Predicting Individual Disease Risk Based on Medical
    History
    Darcy A. Davis
    University of Notre Dame
    [email protected]
    Nitesh V. Chawla
    University of Notre Dame
    [email protected]
    Nicholas Blumm
    Northeastern University
    [email protected]
    Nicholas Christakis
    Harvard Medical School
    [email protected]
    Albert-László Barabási
    Northeastern University
    [email protected]
    ABSTRACT
    The monumental cost of health care, especially for chronic
    disease treatment, is quickly becoming u...
  159. prior
    earlier in time
    To copy otherwise, to
    republish, to post on servers or to redistribute to lists, requires prior specific
    permission and/or a fee.
  160. code
    a set of rules or principles or laws
    We propose CARE,
    a Collaborative Assessment and Recommendation Engine,
    which relies only on a patient's medical history using ICD-
    9-CM codes in order to predict future diseases risks.
  161. use
    put into service
    We propose CARE,
    a Collaborative Assessment and Recommendation Engine,
    which relies only on a patient's medical history using ICD-
    9-CM codes in order to predict future diseases risks.
  162. copyright
    the exclusive right to sell a work
    Copyright 2008 ACM 978-1-59593-991-3/08/10 ...$5.00.
  163. capture
    seize as if by hunting, snaring, or trapping
    We present experimental results on
    a Medicare dataset, demonstrating that CARE and ICARE
    perform well at capturing future disease risks.
  164. vital
    performing an essential function in the living body
    INTRODUCTION
    Medical care and research are literally the most vital part
    of science for humans, as none of us are immune to physical
    ailments and biological deterioration.
  165. increased
    made greater in size or amount or degree
    Since 1992,
    the average age increased to 45 years, and the visit rate for
    persons 45 years of age and over increased by 17% from 407.3
    to 478.2 visits per 100 persons [3].
  166. extensive
    large in spatial extent or range or scope or quantity
    Indeed, technologies
    ranging from linkage equilibrium and candidate gene
    association studies to genome wide associations have provided
    an extensive list of disease-gene associations, o
    ering
    us detailed information on mutations, SNPs, and the associated
    likelihood of developing speci c disease phenotypes
    [4].
  167. limit
    as far as something can go
    However, these
    sporadic and qualitative `risk assessments' generally focus on
    only a few diseases and are limited by a particular doctor's
    experience, memory, and time.
  168. grant
    let have
    Categories and Subject Descriptors
    I.2.1 [Arti cial Intelligence]: Applications and Expert
    Systems|Medicine and science; I.5.1 [Pattern Recogni-
    tion]: Models|statistical ; J.3 [Computer Applications]:
    Life and Medical Sciences|Medical information systems
    General Terms
    Algorithms, Experimentation
    Keywords
    collaborative ltering, disease risk prediction, ensemble, prospective
    health care
    Corresponding Author
    Permission to make digital or hard copies of all or part of this work for...
  169. improved
    become or made better in quality
    We also describe an Iterative version,
    ICARE, which incorporates ensemble concepts for improved
    performance.
  170. candidate
    someone who is considered for something
    Indeed, technologies
    ranging from linkage equilibrium and candidate gene
    association studies to genome wide associations have provided
    an extensive list of disease-gene associations, o
    ering
    us detailed information on mutations, SNPs, and the associated
    likelihood of developing speci c disease phenotypes
    [4].
  171. of age
    having attained a specific age;
    Since 1992,
    the average age increased to 45 years, and the visit rate for
    persons 45 years of age and over increased by 17% from 407.3
    to 478.2 visits per 100 persons [3].
  172. engine
    motor that converts energy into work or motion
    We propose CARE,
    a Collaborative Assessment and Recommendation Engine,
    which relies only on a patient's medical history using ICD-
    9-CM codes in order to predict future diseases risks.
  173. continually
    seemingly without interruption
    Experts expect the burden on the system to
    continually increase in coming years.
  174. perform
    get done
    We present experimental results on
    a Medicare dataset, demonstrating that CARE and ICARE
    perform well at capturing future disease risks.
  175. individual
    being or characteristic of a single thing or person
    Predicting Individual Disease Risk Based on Medical
    History
    Darcy A. Davis
    University of Notre Dame
    [email protected]
    Nitesh V. Chawla
    University of Notre Dame
    [email protected]
    Nicholas Blumm
    Northeastern University
    [email protected]
    Nicholas Christakis
    Harvard Medical School
    [email protected]
    Albert-László Barabási
    Northeastern University
    [email protected]
    ABSTRACT
    The monumental cost of health care, especially for chronic
    disease treatment, is quickly becoming u...
  176. future
    the time yet to come
    We propose CARE,
    a Collaborative Assessment and Recommendation Engine,
    which relies only on a patient's medical history using ICD-
    9-CM codes in order to predict future diseases risks.
  177. describe
    give a statement representing something
    We also describe an Iterative version,
    ICARE, which incorporates ensemble concepts for improved
    performance.
  178. examination
    the act of scrutinizing something closely (as for mistakes)
    Currently, physicians can use
    family and health history and physical examination to approximate
    the risk of a patient, guiding laboratory tests to
    further assess the patient's stage of health.
  179. detail
    a small part considered separately from the whole
    Indeed, technologies
    ranging from linkage equilibrium and candidate gene
    association studies to genome wide associations have provided
    an extensive list of disease-gene associations, o
    ering
    us detailed information on mutations, SNPs, and the associated
    likelihood of developing speci c disease phenotypes
    [4].
  180. introduction
    the act of beginning something new
    INTRODUCTION
    Medical care and research are literally the most vital part
    of science for humans, as none of us are immune to physical
    ailments and biological deterioration.
  181. sign
    a visible clue that something has happened or is present
    This
    crisis has motivated the drive towards preventative medicine,
    where the primary concern is recognizing disease risk and
    taking action at the earliest signs.
  182. cost
    be priced at
    Predicting Individual Disease Risk Based on Medical
    History
    Darcy A. Davis
    University of Notre Dame
    [email protected]
    Nitesh V. Chawla
    University of Notre Dame
    [email protected]
    Nicholas Blumm
    Northeastern University
    [email protected]
    Nicholas Christakis
    Harvard Medical School
    [email protected]
    Albert-László Barabási
    Northeastern University
    [email protected]
    ABSTRACT
    The monumental cost of health care, especially for chronic
    disease treatment, is quickly becoming u...
  183. technology
    the practical application of science to commerce or industry
    Indeed, technologies
    ranging from linkage equilibrium and candidate gene
    association studies to genome wide associations have provided
    an extensive list of disease-gene associations, o
    ering
    us detailed information on mutations, SNPs, and the associated
    likelihood of developing speci c disease phenotypes
    [4].
  184. burden
    weight to be carried or borne
    Experts expect the burden on the system to
    continually increase in coming years.
  185. professional
    of or relating to or suitable as an occupation
    However,
    the combinatorial problem generated by the di
    erent
    disease factors and the previous medical history of a patient
    is so complex that no single health care professional
    can fully comprehend it all.
  186. signal
    any action or gesture that encodes a message
    Given the weak signals that most disease associated SNPs
    or mutations o
    er, it is increasingly clear that the promise
    769
  187. treat
    apply a process to with the aim of preparing for a purpose
    Therefore, current medical
    care is reactive, stepping in once the symptoms of a disease
    have emerged, rather than proactive, treating or eliminating
    a disease at the earliest signs.
  188. novel
    an extended fictional work in prose
    These novel systems require no specialized information
    and provide predictions for medical conditions of
    all kinds in a single run.
  189. increasing
    becoming greater or larger
    The rapidly increasing
    medical concerns of the baby boomer generation is one major
    factor stressing the health care system.
  190. version
    something a little different from others of the same type
    We also describe an Iterative version,
    ICARE, which incorporates ensemble concepts for improved
    performance.
  191. profit
    the advantageous quality of being beneficial
    Categories and Subject Descriptors
    I.2.1 [Arti cial Intelligence]: Applications and Expert
    Systems|Medicine and science; I.5.1 [Pattern Recogni-
    tion]: Models|statistical ; J.3 [Computer Applications]:
    Life and Medical Sciences|Medical information systems
    General Terms
    Algorithms, Experimentation
    Keywords
    collaborative ltering, disease risk prediction, ensemble, prospective
    health care
    Corresponding Author
    Permission to make digital or hard copies of all or part of this work for...
  192. page
    one side of one leaf of a book or other document
    Categories and Subject Descriptors
    I.2.1 [Arti cial Intelligence]: Applications and Expert
    Systems|Medicine and science; I.5.1 [Pattern Recogni-
    tion]: Models|statistical ; J.3 [Computer Applications]:
    Life and Medical Sciences|Medical information systems
    General Terms
    Algorithms, Experimentation
    Keywords
    collaborative ltering, disease risk prediction, ensemble, prospective
    health care
    Corresponding Author
    Permission to make digital or hard copies of all or part of this work for...
  193. rapid
    characterized by speed
    Yet, these rapid
    advances have also unraveled the limitations of the genome
    based approaches [12].
  194. commercial
    connected with or engaged in the exchange of goods
    Categories and Subject Descriptors
    I.2.1 [Arti cial Intelligence]: Applications and Expert
    Systems|Medicine and science; I.5.1 [Pattern Recogni-
    tion]: Models|statistical ; J.3 [Computer Applications]:
    Life and Medical Sciences|Medical information systems
    General Terms
    Algorithms, Experimentation
    Keywords
    collaborative ltering, disease risk prediction, ensemble, prospective
    health care
    Corresponding Author
    Permission to make digital or hard copies of all or part of this work for...
  195. granted
    acknowledged as a supposition
    Categories and Subject Descriptors
    I.2.1 [Arti cial Intelligence]: Applications and Expert
    Systems|Medicine and science; I.5.1 [Pattern Recogni-
    tion]: Models|statistical ; J.3 [Computer Applications]:
    Life and Medical Sciences|Medical information systems
    General Terms
    Algorithms, Experimentation
    Keywords
    collaborative ltering, disease risk prediction, ensemble, prospective
    health care
    Corresponding Author
    Permission to make digital or hard copies of all or part of this work for...
  196. annual
    occurring every year
    Annual health
    care expenditure in the U.S. alone is an overwhelming sum,
    with a strong majority of this money used for chronic disease
    treatment.
  197. generation
    group of genetically related organisms in a line of descent
    The rapidly increasing
    medical concerns of the baby boomer generation is one major
    factor stressing the health care system.
  198. related
    connected logically or causally or by shared characteristics
    The underlying hypothesis behind this line of research
    is that once we catalogue all disease-related mutations, we
    will be able to predict the susceptibility of each individual
    to future diseases using various molecular biomarkers, ushering
    us into an era of predictive medicine.
  199. post
    piece of timber or metal fixed firmly in an upright position
    To copy otherwise, to
    republish, to post on servers or to redistribute to lists, requires prior specific
    permission and/or a fee.
  200. study
    applying the mind to learning and understanding a subject
    A CDC study
    estimates that 880.5 million visits were made to physician
    oces, about 3.1 visits per patient, in 2001 [3].
  201. crisis
    a crucial stage or turning point in the course of something
    This
    crisis has motivated the drive towards preventative medicine,
    where the primary concern is recognizing disease risk and
    taking action at the earliest signs.
  202. limited
    subject to restrictions or constraints
    However, these
    sporadic and qualitative `risk assessments' generally focus on
    only a few diseases and are limited by a particular doctor's
    experience, memory, and time.
  203. universal
    applicable to or common to all members of a group or set
    However, universal testing
    is neither time nor cost ecient.
  204. single
    existing alone or consisting of one entity or part or aspect or individual
    These novel systems require no specialized information
    and provide predictions for medical conditions of
    all kinds in a single run.
  205. performance
    the act of doing something successfully
    We also describe an Iterative version,
    ICARE, which incorporates ensemble concepts for improved
    performance.
  206. sum
    a quantity obtained by the addition of a group of numbers
    Annual health
    care expenditure in the U.S. alone is an overwhelming sum,
    with a strong majority of this money used for chronic disease
    treatment.
  207. revolution
    a single complete turn
    Today the prevailing model of prospective heath care is
    rmly based on the genome revolution.
  208. California
    a state in the western United States on the Pacific
    CIKM’08, October 26-30, 2008, Napa Valley, California, USA.
  209. becoming
    displaying or setting off to best advantage
    Predicting Individual Disease Risk Based on Medical
    History
    Darcy A. Davis
    University of Notre Dame
    [email protected]
    Nitesh V. Chawla
    University of Notre Dame
    [email protected]
    Nicholas Blumm
    Northeastern University
    [email protected]
    Nicholas Christakis
    Harvard Medical School
    [email protected]
    Albert-László Barabási
    Northeastern University
    [email protected]
    ABSTRACT
    The monumental cost of health care, especially for chronic
    disease treatment, is quickly becoming u...
  210. major
    greater in scope or effect
    The rapidly increasing
    medical concerns of the baby boomer generation is one major
    factor stressing the health care system.
  211. majority
    the main part
    Annual health
    care expenditure in the U.S. alone is an overwhelming sum,
    with a strong majority of this money used for chronic disease
    treatment.
  212. intelligence
    the ability to comprehend
    Categories and Subject Descriptors
    I.2.1 [Arti cial Intelligence]: Applications and Expert
    Systems|Medicine and science; I.5.1 [Pattern Recogni-
    tion]: Models|statistical ; J.3 [Computer Applications]:
    Life and Medical Sciences|Medical information systems
    General Terms
    Algorithms, Experimentation
    Keywords
    collaborative ltering, disease risk prediction, ensemble, prospective
    health care
    Corresponding Author
    Permission to make digital or hard copies of all or part of this w...
  213. baby
    a very young mammal
    The rapidly increasing
    medical concerns of the baby boomer generation is one major
    factor stressing the health care system.
  214. approach
    move towards
    Yet, these rapid
    advances have also unraveled the limitations of the genome
    based approaches [12].
  215. range
    a variety of different things or activities
    Indeed, technologies
    ranging from linkage equilibrium and candidate gene
    association studies to genome wide associations have provided
    an extensive list of disease-gene associations, o
    ering
    us detailed information on mutations, SNPs, and the associated
    likelihood of developing speci c disease phenotypes
    [4].
  216. average
    an intermediate scale value regarded as normal or usual
    Since 1992,
    the average age increased to 45 years, and the visit rate for
    persons 45 years of age and over increased by 17% from 407.3
    to 478.2 visits per 100 persons [3].
  217. guide
    someone employed to conduct others
    Currently, physicians can use
    family and health history and physical examination to approximate
    the risk of a patient, guiding laboratory tests to
    further assess the patient's stage of health.
  218. rapidly
    with quick movements
    The rapidly increasing
    medical concerns of the baby boomer generation is one major
    factor stressing the health care system.
  219. October
    the month following September and preceding November
    CIKM’08, October 26-30, 2008, Napa Valley, California, USA.
  220. term
    a limited period of time during which something lasts
    Categories and Subject Descriptors
    I.2.1 [Arti cial Intelligence]: Applications and Expert
    Systems|Medicine and science; I.5.1 [Pattern Recogni-
    tion]: Models|statistical ; J.3 [Computer Applications]:
    Life and Medical Sciences|Medical information systems
    General Terms
    Algorithms, Experimentation
    Keywords
    collaborative ltering, disease risk prediction, ensemble, prospective
    health care
    Corresponding Author
    Permission to make digital or hard copies of all or part of this w...
  221. weak
    wanting in physical strength
    Given the weak signals that most disease associated SNPs
    or mutations o
    er, it is increasingly clear that the promise
    769
  222. advance
    move forward
    Yet, these rapid
    advances have also unraveled the limitations of the genome
    based approaches [12].
  223. needs
    in such a manner as could not be otherwise
    Health care, thus, needs to become more proactive than
    reactive in recognizing the onset of disease and risk.
  224. previous
    just preceding something else in time or order
    However,
    the combinatorial problem generated by the di
    erent
    disease factors and the previous medical history of a patient
    is so complex that no single health care professional
    can fully comprehend it all.
  225. problem
    a question raised for consideration or solution
    However,
    the combinatorial problem generated by the di
    erent
    disease factors and the previous medical history of a patient
    is so complex that no single health care professional
    can fully comprehend it all.
  226. 100
    ten 10s
    Since 1992,
    the average age increased to 45 years, and the visit rate for
    persons 45 years of age and over increased by 17% from 407.3
    to 478.2 visits per 100 persons [3].
  227. drive
    operate or control a vehicle
    This
    crisis has motivated the drive towards preventative medicine,
    where the primary concern is recognizing disease risk and
    taking action at the earliest signs.
  228. current
    occurring in or belonging to the present time
    Therefore, current medical
    care is reactive, stepping in once the symptoms of a disease
    have emerged, rather than proactive, treating or eliminating
    a disease at the earliest signs.
  229. otherwise
    in another and different manner
    To copy otherwise, to
    republish, to post on servers or to redistribute to lists, requires prior specific
    permission and/or a fee.
  230. age
    how long something has existed
    Since 1992,
    the average age increased to 45 years, and the visit rate for
    persons 45 years of age and over increased by 17% from 407.3
    to 478.2 visits per 100 persons [3].
  231. person
    a human being
    Since 1992,
    the average age increased to 45 years, and the visit rate for
    persons 45 years of age and over increased by 17% from 407.3
    to 478.2 visits per 100 persons [3].
  232. advantage
    the quality of having a superior or more favorable position
    Categories and Subject Descriptors
    I.2.1 [Arti cial Intelligence]: Applications and Expert
    Systems|Medicine and science; I.5.1 [Pattern Recogni-
    tion]: Models|statistical ; J.3 [Computer Applications]:
    Life and Medical Sciences|Medical information systems
    General Terms
    Algorithms, Experimentation
    Keywords
    collaborative ltering, disease risk prediction, ensemble, prospective
    health care
    Corresponding Author
    Permission to make digital or hard copies of all or part of this work for...
  233. fully
    to the greatest degree or extent; completely or entirely;
    However,
    the combinatorial problem generated by the di
    erent
    disease factors and the previous medical history of a patient
    is so complex that no single health care professional
    can fully comprehend it all.
  234. valley
    a long depression in the surface of the land
    CIKM’08, October 26-30, 2008, Napa Valley, California, USA.
  235. promise
    a verbal commitment agreeing to do something in the future
    Given the weak signals that most disease associated SNPs
    or mutations o
    er, it is increasingly clear that the promise
    769
  236. however
    in whatever way or manner
    However, universal testing
    is neither time nor cost ecient.
  237. notice
    the act of paying attention
    Categories and Subject Descriptors
    I.2.1 [Arti cial Intelligence]: Applications and Expert
    Systems|Medicine and science; I.5.1 [Pattern Recogni-
    tion]: Models|statistical ; J.3 [Computer Applications]:
    Life and Medical Sciences|Medical information systems
    General Terms
    Algorithms, Experimentation
    Keywords
    collaborative ltering, disease risk prediction, ensemble, prospective
    health care
    Corresponding Author
    Permission to make digital or hard copies of all or part of this work for...
  238. stage
    any distinct time period in a sequence of events
    Currently, physicians can use
    family and health history and physical examination to approximate
    the risk of a patient, guiding laboratory tests to
    further assess the patient's stage of health.
  239. early
    at or near the beginning of a period of time or course of events or before the usual or expected time
    This
    crisis has motivated the drive towards preventative medicine,
    where the primary concern is recognizing disease risk and
    taking action at the earliest signs.
  240. memory
    the cognitive process whereby past experience is remembered
    However, these
    sporadic and qualitative `risk assessments' generally focus on
    only a few diseases and are limited by a particular doctor's
    experience, memory, and time.
  241. expect
    regard something as probable or likely
    Experts expect the burden on the system to
    continually increase in coming years.
  242. today
    on this day as distinct from yesterday or tomorrow
    Today the prevailing model of prospective heath care is
    rmly based on the genome revolution.
  243. conditions
    the context that influences the performance of a process
    These novel systems require no specialized information
    and provide predictions for medical conditions of
    all kinds in a single run.
  244. become
    come into existence
    Predicting Individual Disease Risk Based on Medical
    History
    Darcy A. Davis
    University of Notre Dame
    [email protected]
    Nitesh V. Chawla
    University of Notre Dame
    [email protected]
    Nicholas Blumm
    Northeastern University
    [email protected]
    Nicholas Christakis
    Harvard Medical School
    [email protected]
    Albert-László Barabási
    Northeastern University
    [email protected]
    ABSTRACT
    The monumental cost of health care, especially for chronic
    disease treatment, is quickly becoming u...
  245. similar
    having the same or nearly the same characteristics
    CARE
    uses collaborative ltering to predict each patient's greatest
    disease risks based on their own medical history and that
    of similar patients.
  246. year
    the period of time that it takes for a planet (as, e.g., Earth or Mars) to make a complete revolution around the sun
    Experts expect the burden on the system to
    continually increase in coming years.
  247. wide
    having great extent from one side to the other
    Indeed, technologies
    ranging from linkage equilibrium and candidate gene
    association studies to genome wide associations have provided
    an extensive list of disease-gene associations, o
    ering
    us detailed information on mutations, SNPs, and the associated
    likelihood of developing speci c disease phenotypes
    [4].
  248. terms
    status with respect to the relations between people or groups
    Categories and Subject Descriptors
    I.2.1 [Arti cial Intelligence]: Applications and Expert
    Systems|Medicine and science; I.5.1 [Pattern Recogni-
    tion]: Models|statistical ; J.3 [Computer Applications]:
    Life and Medical Sciences|Medical information systems
    General Terms
    Algorithms, Experimentation
    Keywords
    collaborative ltering, disease risk prediction, ensemble, prospective
    health care
    Corresponding Author
    Permission to make digital or hard copies of all or part of this w...
  249. greatest
    highest in quality
    CARE
    uses collaborative ltering to predict each patient's greatest
    disease risks based on their own medical history and that
    of similar patients.
  250. author
    a person who writes professionally
    Categories and Subject Descriptors
    I.2.1 [Arti cial Intelligence]: Applications and Expert
    Systems|Medicine and science; I.5.1 [Pattern Recogni-
    tion]: Models|statistical ; J.3 [Computer Applications]:
    Life and Medical Sciences|Medical information systems
    General Terms
    Algorithms, Experimentation
    Keywords
    collaborative ltering, disease risk prediction, ensemble, prospective
    health care
    Corresponding Author
    Permission to make digital or hard copies of all or part of this w...
  251. step
    the act of changing location by raising the foot and setting it down
    Therefore, current medical
    care is reactive, stepping in once the symptoms of a disease
    have emerged, rather than proactive, treating or eliminating
    a disease at the earliest signs.
  252. particular
    unique or specific to a person or thing or category
    However, these
    sporadic and qualitative `risk assessments' generally focus on
    only a few diseases and are limited by a particular doctor's
    experience, memory, and time.
  253. rate
    a quantity considered as a proportion of another quantity
    Since 1992,
    the average age increased to 45 years, and the visit rate for
    persons 45 years of age and over increased by 17% from 407.3
    to 478.2 visits per 100 persons [3].
  254. generally
    usually; as a rule
    However, these
    sporadic and qualitative `risk assessments' generally focus on
    only a few diseases and are limited by a particular doctor's
    experience, memory, and time.
  255. condition
    a mode of being or form of existence of a person or thing
    These novel systems require no specialized information
    and provide predictions for medical conditions of
    all kinds in a single run.
  256. quickly
    with little or no delay
    Predicting Individual Disease Risk Based on Medical
    History
    Darcy A. Davis
    University of Notre Dame
    [email protected]
    Nitesh V. Chawla
    University of Notre Dame
    [email protected]
    Nicholas Blumm
    Northeastern University
    [email protected]
    Nicholas Christakis
    Harvard Medical School
    [email protected]
    Albert-László Barabási
    Northeastern University
    [email protected]
    ABSTRACT
    The monumental cost of health care, especially for chronic
    disease treatment, is quickly becoming u...
  257. experience
    the content of observation or participation in an event
    However, these
    sporadic and qualitative `risk assessments' generally focus on
    only a few diseases and are limited by a particular doctor's
    experience, memory, and time.
  258. personal
    concerning an individual or his or her private life
    Categories and Subject Descriptors
    I.2.1 [Arti cial Intelligence]: Applications and Expert
    Systems|Medicine and science; I.5.1 [Pattern Recogni-
    tion]: Models|statistical ; J.3 [Computer Applications]:
    Life and Medical Sciences|Medical information systems
    General Terms
    Algorithms, Experimentation
    Keywords
    collaborative ltering, disease risk prediction, ensemble, prospective
    health care
    Corresponding Author
    Permission to make digital or hard copies of all or part of this work for...
  259. various
    having great diversity or variety
    The underlying hypothesis behind this line of research
    is that once we catalogue all disease-related mutations, we
    will be able to predict the susceptibility of each individual
    to future diseases using various molecular biomarkers, ushering
    us into an era of predictive medicine.
  260. in order
    in a state of proper readiness or preparation or arrangement
    We propose CARE,
    a Collaborative Assessment and Recommendation Engine,
    which relies only on a patient's medical history using ICD-
    9-CM codes in order to predict future diseases risks.
  261. result
    something that follows as a consequence
    We present experimental results on
    a Medicare dataset, demonstrating that CARE and ICARE
    perform well at capturing future disease risks.
  262. bear
    be pregnant with
    Categories and Subject Descriptors
    I.2.1 [Arti cial Intelligence]: Applications and Expert
    Systems|Medicine and science; I.5.1 [Pattern Recogni-
    tion]: Models|statistical ; J.3 [Computer Applications]:
    Life and Medical Sciences|Medical information systems
    General Terms
    Algorithms, Experimentation
    Keywords
    collaborative ltering, disease risk prediction, ensemble, prospective
    health care
    Corresponding Author
    Permission to make digital or hard copies of all or part of this work for...
  263. especially
    to a distinctly greater extent or degree than is common
    Predicting Individual Disease Risk Based on Medical
    History
    Darcy A. Davis
    University of Notre Dame
    [email protected]
    Nitesh V. Chawla
    University of Notre Dame
    [email protected]
    Nicholas Blumm
    Northeastern University
    [email protected]
    Nicholas Christakis
    Harvard Medical School
    [email protected]
    Albert-László Barabási
    Northeastern University
    [email protected]
    ABSTRACT
    The monumental cost of health care, especially for chronic
    disease treatment, is quickly becoming u...
  264. years
    a prolonged period of time
    Experts expect the burden on the system to
    continually increase in coming years.
  265. action
    something done (usually as opposed to something said)
    This
    crisis has motivated the drive towards preventative medicine,
    where the primary concern is recognizing disease risk and
    taking action at the earliest signs.
  266. make
    perform or carry out
    Categories and Subject Descriptors
    I.2.1 [Arti cial Intelligence]: Applications and Expert
    Systems|Medicine and science; I.5.1 [Pattern Recogni-
    tion]: Models|statistical ; J.3 [Computer Applications]:
    Life and Medical Sciences|Medical information systems
    General Terms
    Algorithms, Experimentation
    Keywords
    collaborative ltering, disease risk prediction, ensemble, prospective
    health care
    Corresponding Author
    Permission to make digital or hard copies of all or part of this w...
  267. neither
    used to indicate something also does not apply
    However, universal testing
    is neither time nor cost ecient.
  268. million
    the number that is represented as a one followed by 6 zeros
    A CDC study
    estimates that 880.5 million visits were made to physician
    oces, about 3.1 visits per patient, in 2001 [3].
  269. none
    not at all or in no way
    INTRODUCTION
    Medical care and research are literally the most vital part
    of science for humans, as none of us are immune to physical
    ailments and biological deterioration.
  270. clear
    readily apparent to the mind
    Given the weak signals that most disease associated SNPs
    or mutations o
    er, it is increasingly clear that the promise
    769
  271. further
    to or at a greater extent or degree or a more advanced stage
    Currently, physicians can use
    family and health history and physical examination to approximate
    the risk of a patient, guiding laboratory tests to
    further assess the patient's stage of health.
  272. part
    one of the portions into which something is regarded as divided and which together constitute a whole
    Categories and Subject Descriptors
    I.2.1 [Arti cial Intelligence]: Applications and Expert
    Systems|Medicine and science; I.5.1 [Pattern Recogni-
    tion]: Models|statistical ; J.3 [Computer Applications]:
    Life and Medical Sciences|Medical information systems
    General Terms
    Algorithms, Experimentation
    Keywords
    collaborative ltering, disease risk prediction, ensemble, prospective
    health care
    Corresponding Author
    Permission to make digital or hard copies of all or part of this w...
  273. each
    separately for every person or thing
    CARE
    uses collaborative ltering to predict each patient's greatest
    disease risks based on their own medical history and that
    of similar patients.
  274. taking
    the act of someone who picks up or takes something
    This
    crisis has motivated the drive towards preventative medicine,
    where the primary concern is recognizing disease risk and
    taking action at the earliest signs.
  275. subject
    some situation or event that is thought about
    Categories and Subject Descriptors
    I.2.1 [Arti cial Intelligence]: Applications and Expert
    Systems|Medicine and science; I.5.1 [Pattern Recogni-
    tion]: Models|statistical ; J.3 [Computer Applications]:
    Life and Medical Sciences|Medical information systems
    General Terms
    Algorithms, Experimentation
    Keywords
    collaborative ltering, disease risk prediction, ensemble, prospective
    health care
    Corresponding Author
    Permission to make digital or hard copies of all or part of this w...
  276. school
    an educational institution
    Predicting Individual Disease Risk Based on Medical
    History
    Darcy A. Davis
    University of Notre Dame
    [email protected]
    Nitesh V. Chawla
    University of Notre Dame
    [email protected]
    Nicholas Blumm
    Northeastern University
    [email protected]
    Nicholas Christakis
    Harvard Medical School
    [email protected]
    Albert-László Barabási
    Northeastern University
    [email protected]
    ABSTRACT
    The monumental cost of health care, especially for chronic
    disease treatment, is quickly becoming u...
  277. run
    move fast by using one's feet
    These novel systems require no specialized information
    and provide predictions for medical conditions of
    all kinds in a single run.
  278. once
    on one occasion
    Therefore, current medical
    care is reactive, stepping in once the symptoms of a disease
    have emerged, rather than proactive, treating or eliminating
    a disease at the earliest signs.
  279. strong
    having strength or power greater than average or expected
    Annual health
    care expenditure in the U.S. alone is an overwhelming sum,
    with a strong majority of this money used for chronic disease
    treatment.
  280. able
    having the necessary means or skill to do something
    The underlying hypothesis behind this line of research
    is that once we catalogue all disease-related mutations, we
    will be able to predict the susceptibility of each individual
    to future diseases using various molecular biomarkers, ushering
    us into an era of predictive medicine.
  281. human
    a person; a hominid with a large brain and articulate speech
    INTRODUCTION
    Medical care and research are literally the most vital part
    of science for humans, as none of us are immune to physical
    ailments and biological deterioration.
  282. line
    a length between two points
    The underlying hypothesis behind this line of research
    is that once we catalogue all disease-related mutations, we
    will be able to predict the susceptibility of each individual
    to future diseases using various molecular biomarkers, ushering
    us into an era of predictive medicine.
  283. order
    logical arrangement of different elements
    We propose CARE,
    a Collaborative Assessment and Recommendation Engine,
    which relies only on a patient's medical history using ICD-
    9-CM codes in order to predict future diseases risks.
  284. hard
    resisting weight or pressure
    Categories and Subject Descriptors
    I.2.1 [Arti cial Intelligence]: Applications and Expert
    Systems|Medicine and science; I.5.1 [Pattern Recogni-
    tion]: Models|statistical ; J.3 [Computer Applications]:
    Life and Medical Sciences|Medical information systems
    General Terms
    Algorithms, Experimentation
    Keywords
    collaborative ltering, disease risk prediction, ensemble, prospective
    health care
    Corresponding Author
    Permission to make digital or hard copies of all or part of this w...
  285. need
    require or want
    Health care, thus, needs to become more proactive than
    reactive in recognizing the onset of disease and risk.
  286. coming
    of the relatively near future
    Experts expect the burden on the system to
    continually increase in coming years.
  287. U.S.
    North American republic containing 50 states - 48 conterminous states in North America plus Alaska in northwest North America and the Hawaiian Islands in the Pacific Ocean; achieved independence in 1776
    Annual health
    care expenditure in the U.S. alone is an overwhelming sum,
    with a strong majority of this money used for chronic disease
    treatment.
  288. therefore
    as a result; from that fact or reason
    Therefore, current medical
    care is reactive, stepping in once the symptoms of a disease
    have emerged, rather than proactive, treating or eliminating
    a disease at the earliest signs.
  289. alone
    isolated from others
    Annual health
    care expenditure in the U.S. alone is an overwhelming sum,
    with a strong majority of this money used for chronic disease
    treatment.
  290. also
    in addition
    We also describe an Iterative version,
    ICARE, which incorporates ensemble concepts for improved
    performance.
  291. behind
    in or to or toward the rear
    The underlying hypothesis behind this line of research
    is that once we catalogue all disease-related mutations, we
    will be able to predict the susceptibility of each individual
    to future diseases using various molecular biomarkers, ushering
    us into an era of predictive medicine.
  292. family
    a group of people related to one another
    Currently, physicians can use
    family and health history and physical examination to approximate
    the risk of a patient, guiding laboratory tests to
    further assess the patient's stage of health.
  293. indeed
    in truth (often tends to intensify)
    Indeed, technologies
    ranging from linkage equilibrium and candidate gene
    association studies to genome wide associations have provided
    an extensive list of disease-gene associations, o
    ering
    us detailed information on mutations, SNPs, and the associated
    likelihood of developing speci c disease phenotypes
    [4].
  294. kind
    having a tender and considerate and helpful nature
    These novel systems require no specialized information
    and provide predictions for medical conditions of
    all kinds in a single run.
  295. present
    happening or existing now
    We present experimental results on
    a Medicare dataset, demonstrating that CARE and ICARE
    perform well at capturing future disease risks.
  296. used
    previously owned by another
    Annual health
    care expenditure in the U.S. alone is an overwhelming sum,
    with a strong majority of this money used for chronic disease
    treatment.
  297. full
    containing as much or as many as is possible or normal
    Categories and Subject Descriptors
    I.2.1 [Arti cial Intelligence]: Applications and Expert
    Systems|Medicine and science; I.5.1 [Pattern Recogni-
    tion]: Models|statistical ; J.3 [Computer Applications]:
    Life and Medical Sciences|Medical information systems
    General Terms
    Algorithms, Experimentation
    Keywords
    collaborative ltering, disease risk prediction, ensemble, prospective
    health care
    Corresponding Author
    Permission to make digital or hard copies of all or part of this work for...
  298. money
    the most common medium of exchange
    Annual health
    care expenditure in the U.S. alone is an overwhelming sum,
    with a strong majority of this money used for chronic disease
    treatment.
  299. rather
    more readily or willingly
    Therefore, current medical
    care is reactive, stepping in once the symptoms of a disease
    have emerged, rather than proactive, treating or eliminating
    a disease at the earliest signs.
  300. given
    acknowledged as a supposition
    Given the weak signals that most disease associated SNPs
    or mutations o
    er, it is increasingly clear that the promise
    769
  301. a few
    more than one but indefinitely small in number
    However, these
    sporadic and qualitative `risk assessments' generally focus on
    only a few diseases and are limited by a particular doctor's
    experience, memory, and time.
  302. thus
    from that fact or reason or as a result
    Health care, thus, needs to become more proactive than
    reactive in recognizing the onset of disease and risk.
  303. most
    used to indicate the greatest amount or degree of a quality
    INTRODUCTION
    Medical care and research are literally the most vital part
    of science for humans, as none of us are immune to physical
    ailments and biological deterioration.
Created on 十一月 7, 2011

Sign up now (it’s free!)

Whether you’re a teacher or a learner 全国统一开奖网址, soonshin.com can put you or your class on the path to systematic vocabulary improvement.