The problem of bias in training data in regression problems in medical decision support

Artificial Intelligence in Medicine

Volumn 24, Issue 1, 2002, Pages 51-70

  Mac Namee, B ^a   Cunningham, P ^a   Byrne, S ^a   Corrigan, O I ^a  

^a TRINITY COLLEGE DUBLIN   (Ireland)

Author keywords

Anticoagulant drug therapy; Artificial neural networks; Medical decision support; Regression

Indexed keywords

DECISION SUPPORT SYSTEMS; DRUG THERAPY; LEARNING SYSTEMS; NEURAL NETWORKS; REGRESSION ANALYSIS; SAMPLING;

ANTICOAGULANT DRUG THERAPY;

BIOMEDICAL ENGINEERING;

WARFARIN;

ABALONE; AGE DETERMINATION; ANALYTICAL ERROR; ANTICOAGULANT THERAPY; ARTICLE; ARTIFICIAL NEURAL NETWORK; BLOOD CLOTTING TIME; CONTROLLED STUDY; CORONARY RISK; EVALUATION; HEART INFARCTION; HUMAN; INTERNATIONAL STANDARD UNIT; LEARNING; MEDICAL DECISION MAKING; MEDICAL EDUCATION; NONHUMAN; PREDICTION; PRIORITY JOURNAL; REGRESSION ANALYSIS; SAMPLING; STRATIFICATION; TREATMENT OUTCOME;

BIAS (EPIDEMIOLOGY); COMPUTER SIMULATION; DECISION MAKING, COMPUTER-ASSISTED; DECISION SUPPORT TECHNIQUES; HUMANS; REGRESSION ANALYSIS;

EID: 0036146454     PISSN: 09333657     EISSN: None     Source Type: Journal    
DOI: 10.1016/S0933-3657(01)00092-6     Document Type: Article

References (19)

1. Boosted mixture of experts: An ensemble learning scheme
2. Guidelines on oral anticoagulation, 2nd edition
3. Stratifying samples to improve learning
4. Using neural nets for decision support in prescription and outcome prediction in anticoagulation drug therapy
5. Improving minority class prediction using case-specific feature weights
6. Toward scalable learning with non-uniform class and cost distributions: A case study in credit card fraud detection
7. Stability problems with artificial neural networks and the ensemble solution
8. Boosting using neural networks
9. Experiments with a new boosting algorithm
10. Machine learning for the detection of oil spills in satellite radar images
11. Heterogeneous uncertainty sampling for supervised learning
12. Reducing misclassification costs
13. Robust classification for imprecise environments
14. The case against accuracy estimation for comparing induction algorithms
15. A brief introduction to boosting
16. Measuring the accuracy of diagnostic systems
17. Learning algorithms for key phrase extraction