Artificial Intelligence: Bayesian versus Heuristic Method for Diagnostic Decision Support

Abstract

Evoking strength is one of the important contributions of the field of Biomedical Informatics to the discipline of Artificial Intelligence. The University at Buffalo's Orthopedics Department wanted to create an expert system to assist patients with self-diagnosis of knee problems and to thereby facilitate referral to the right orthopedic subspecialist. They had two independent sports medicine physicians review 469 cases. A board-certified orthopedic sports medicine practitioner, L.B., reviewed any disagreements until a gold standard diagnosis was reached. For each case, the patients entered 126 potential answers to 26 questions into a Web interface. These were modeled by an expert sports medicine physician and the answers were reviewed by L.B. For each finding, the clinician specified the sensitivity (term frequency) and both specificity (Sp) and the heuristic evoking strength (ES). Heuristics are methods of reasoning with only partial evidence. An expert system was constructed that reflected the posttest odds of disease-ranked list for each case. We compare the accuracy of using Sp to that of using ES (original model, p < 0.0008; term importance * disease importance [DItimesTI] model, p < 0.0001: Wilcoxon ranked sum test). For patient referral assignment, Sp in the DItimesTI model was superior to the use of ES. By the fifth diagnosis, the advantage was lost and so there is no difference between the techniques when serving as a reminder system.

Keywords

clinical decision support - process management tools - specific types - clinical information systems - computer-assisted diagnosis - new diagnosis - disease management - clinical informatics - professional training - education - decision support algorithm - computer-assisted decision making - knowledge modeling and representation - knowledge management - clinical information

Protection of Human and Animal Subjects

The study obtained IRB approval # 1690612E.

• References

• 1 Shortliffe EH, Buchanan BG. Artificial intelligence. N Engl J Med 1980; 302 (26) 1482
• 2 Zhou SM, Fernandez-Gutierrez F, Kennedy J. , et al; UK Biobank Follow-up and Outcomes Group. Defining disease phenotypes in primary care electronic health records by a machine learning approach: a case study in identifying rheumatoid arthritis. PLoS One 2016; 11 (05) e0154515
• 3 Warner HR, Olmsted CM, Rutherford BD. HELP–a program for medical decision-making. Comput Biomed Res 1972; 5 (01) 65-74
• 4 Ledley RS, Lusted LB. The role of computers in medical diagnosis. Med Dok 1961; 5: 70-78
• 5 de Dombal T, Clamp S, Margulies M, Chan M. Computer training for doctors and students. BMJ 1994; 309 (6963): 1234-1235
• 6 Shortliffe EH, Axline SG, Buchanan BG, Merigan TC, Cohen SN. An artificial intelligence program to advise physicians regarding antimicrobial therapy. Comput Biomed Res 1973; 6 (06) 544-560
• 7 Shortliffe EH, Davis R, Axline SG, Buchanan BG, Green CC, Cohen SN. Computer-based consultations in clinical therapeutics: explanation and rule acquisition capabilities of the MYCIN system. Comput Biomed Res 1975; 8 (04) 303-320
• 8 Pryor TA, Gardner RM, Clayton PD, Warner HR. The HELP system. J Med Syst 1983; 7 (02) 87-102
• 9 Hripcsak G, Wigertz OB, Kahn MG, Clayton PD, Pryor TA. ASTM E31.15 on health knowledge representation: the Arden Syntax. Stud Health Technol Inform 1993; 6: 105-112
• 10 Pryor TA. The use of medical logic modules at LDS hospital. Comput Biol Med 1994; 24 (05) 391-395
• 11 Miller R, Masarie FE, Myers JD. Quick medical reference (QMR) for diagnostic assistance. MD Comput 1986; 3 (05) 34-48
• 12 Bankowitz RA, McNeil MA, Challinor SM, Parker RC, Kapoor WN, Miller RA. A computer-assisted medical diagnostic consultation service. Implementation and prospective evaluation of a prototype. Ann Intern Med 1989; 110 (10) 824-832
• 13 Giuse DA, Giuse NB, Miller RA. A tool for the computer-assisted creation of QMR medical knowledge base disease profiles. Proc Annu Symp Comput Appl Med Care 1991; 978-979
• 14 Barnett GO, Hoffer EP, Packer MS. , et al. DXplain - important issues in the development of a computer based decision support system. Proc Annu Symp Comput Appl Med Care 1990; 1013
• 15 Barnett GO, Cimino JJ, Hupp JA, Hoffer EP. DXplain. An evolving diagnostic decision-support system. JAMA 1987; 258 (01) 67-74
• 16 Packer MS, Hoffer EP, Barnett GO. , et al. Evolution of DXplain: a decision support system. Proc Annu Symp Comput Appl Med Care 1989; 949-951
• 17 Elkin PL, Barnett GO, Famiglietti KT, Kim RJ. Closing the loop on diagnostic decision support systems. Proc Annu Symp Comput Appl Med Care 1990; 589-593
• 18 Elkin PL, McLatchey J, Packer M. , et al. Automated batch searching of MEDLINE for DXplain. Proc Annu Symp Comput Appl Med Care 1989; 436-440
• 19 Barnett GO, Famiglietti KT, Kim RJ, Hoffer EP, Feldman MJ. DXplain on the Internet. Proc AMIA Symp 1998; 607-611
• 20 Bouhaddou O, Warner H. An interactive patient information and education system (Medical HouseCall) based on a physician expert system (Iliad). Medinfo 1995; 8 (Pt 2): 1181-1185
• 21 Urban MK, Chiu T, Wolfe S, Magid S. Electronic ordering system improves postoperative pain management after total knee or hip arthroplasty. Appl Clin Inform 2015; 6 (03) 591-599
• 22 Farion KJ, Wilk S, Michalowski W, O'Sullivan D, Sayyad-Shirabad J. Comparing predictions made by a prediction model, clinical score, and physicians: pediatric asthma exacerbations in the emergency department. Appl Clin Inform 2013; 4 (03) 376-391
• 23 Qureshi MN, Min B, Jo HJ, Lee B. Multiclass classification for the differential diagnosis on the ADHD subtypes using recursive feature elimination and hierarchical extreme learning machine: structural MRI study. PLoS One 2016; 11 (08) e0160697
• 24 Ye Z, Tafti AP, He KY, Wang K, He MM. SparkText: biomedical text mining on big data framework. PLoS One 2016; 11 (09) e0162721