Application of the fuzzy ARTMAP neural network model to medical pattern classification tasks

被引:40
作者
Downs, J
Harrison, RF
Kennedy, RL
Cross, SS
机构
[1] UNIV SHEFFIELD, DEPT AUTOMAT CONTROL & SYST ENGN, SHEFFIELD S1 3JD, S YORKSHIRE, ENGLAND
[2] UNIV EDINBURGH, WESTERN GEN HOSP, DEPT MED, EDINBURGH EH4 2XU, MIDLOTHIAN, SCOTLAND
[3] UNIV SHEFFIELD, SCH MED, DEPT PATHOL, SHEFFIELD S10 2RX, S YORKSHIRE, ENGLAND
关键词
artificial neural networks; fuzzy ARTMAP; breast cancer; coronary care; myocardial infarction;
D O I
10.1016/0933-3657(95)00044-5
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
This paper presents research into the application of the fuzzy ARTMAP neural network model to medical pattern classification tasks. A number of domains, both diagnostic and prognostic, are considered. Each such domain highlights a particularly useful aspect of the model. The first, coronary care patient prognosis, demonstrates the ARTMAP voting strategy involving 'pooled' decision-making using a number of networks, each of which has learned a slightly different mapping of input features to pattern classes. The second domain, breast cancer diagnosis, demonstrates the model's symbolic rule extraction capabilities which support the validation and explanation of a network's predictions. The final domain, diagnosis of acute myocardial infarction, demonstrates a novel category pruning technique allowing the performance of a trained network to be altered so as to favour predictions of one class over another (e.g. trading sensitivity for specificity or vice versa). It also introduces a 'cascaded' variant of the voting strategy intended to allow identification of a subset of cases which the network has a very high certainty of classifying correctly.
引用
收藏
页码:403 / 428
页数:26
相关论文
共 36 条
  • [1] EARLIEST ELECTROCARDIOGRAPHIC EVIDENCE OF MYOCARDIAL-INFARCTION - IMPLICATIONS FOR THROMBOLYTIC TREATMENT
    ADAMS, J
    TRENT, R
    RAWLES, J
    [J]. BMJ-BRITISH MEDICAL JOURNAL, 1993, 307 (6901): : 409 - 413
  • [2] BIOCHEMICAL MARKERS OF MYOCARDIAL INJURY - IS MB CREATINE-KINASE THE CHOICE FOR THE 1990S
    ADAMS, JE
    ABENDSCHEIN, DR
    JAFFE, AS
    [J]. CIRCULATION, 1993, 88 (02) : 750 - 763
  • [3] APOLLONI B, 1990, P INT JOINT C NEURAL, V2, P571
  • [4] ASFOUR YR, 1993, P WORLD C NEUR NETW, V2, P210
  • [5] A COMPARISON OF NEURAL NETWORK AND OTHER PATTERN-RECOGNITION APPROACHES TO THE DIAGNOSIS OF LOW-BACK DISORDERS
    BOUNDS, DG
    LLOYD, PJ
    MATHEW, BG
    [J]. NEURAL NETWORKS, 1990, 3 (05) : 583 - 591
  • [6] BRAHAMS D, 1989, LANCET, V2, P632
  • [7] Carpenter G., 1991, Pattern recognition by self-organizing neural networks
  • [8] THE ART OF ADAPTIVE PATTERN-RECOGNITION BY A SELF-ORGANIZING NEURAL NETWORK
    CARPENTER, GA
    GROSSBERG, S
    [J]. COMPUTER, 1988, 21 (03) : 77 - 88
  • [9] FUZZY ART - FAST STABLE LEARNING AND CATEGORIZATION OF ANALOG PATTERNS BY AN ADAPTIVE RESONANCE SYSTEM
    CARPENTER, GA
    GROSSBERG, S
    ROSEN, DB
    [J]. NEURAL NETWORKS, 1991, 4 (06) : 759 - 771
  • [10] FUZZY ARTMAP - A NEURAL NETWORK ARCHITECTURE FOR INCREMENTAL SUPERVISED LEARNING OF ANALOG MULTIDIMENSIONAL MAPS
    CARPENTER, GA
    GROSSBERG, S
    MARKUZON, N
    REYNOLDS, JH
    ROSEN, DB
    [J]. IEEE TRANSACTIONS ON NEURAL NETWORKS, 1992, 3 (05): : 698 - 713