Clinical Decisions for Psychiatric Inpatients and Their Evaluation by a Trained Neural Network

Abstract

Ninety-two consecutive treatment decisions regarding psychotic and depressed inpatients were evaluated by a trained neural network. Simultaneous evaluations according to the Brief Psychiatric Rating Scale (BPRS), the Hamilton scale, and the neural network were performed. A 15-point decrease in the BPRS for psychotic patients or a 10-point decrease in the Hamilton scale for depressed patients was the cut-off point for treatment success. The neural network performed similarly to the clinicians. The combined clinician-network success rate reached 79% and was significantly higher than that for each alone. Clinicians and neural network disagreed on 62% of decisions made for depressed patients and on 55% of decisions made for psychotic patients. The proportion of agreement was lower in both diagnostic groups than expected by chance. However, the high success of the combined clinician and neural network decisions and the high rate of mutual disagreement may imply that a combined decision is fruitful.

Key-Words

Decision - Neural Network - Psychiatry

• REFERENCES

• 1 Leaning MS, Ng KE, Cramp DG. Decision support for patient management in oncology. Med Inform 1992; 17: 35-46.
• 2 Heckerman DE, Horvitz EJ, Nathwani BN. Towards normative expert systems: Part I. The pathfinder project. Meth Inform Med 1992; 31: 90-105.
• 3 Campr B, Bednar B. A computer expert system for malignant lymphoma. Cesk Pathol 1992; 28: 102-5.
• 4 Gentile FP, Chiatti L, Mauro F. et al. Introduction of cytotoxic agents: a rule-based system for computer assisted cell survival analysis. Anticancer Res 1992; 12: 637-43.
• 5 Schewe S. Computer expert system for diagnosis of joint pain. Differential diagnosis per computer anamnesis. Fortschr Med 1992; 110: 213-4.
• 6 Boone JM, Sigillito VG, Shaber GS. Neural networks in radiology: An introduction and evaluation in a signal detection task. Med Physics 1990; 17: 234-41.
• 7 Baxt G. Use of artificial neural networks for the diagnosis of myocardial infarction. Ann Intern Med 1991; 115: 843-8.
• 8 Mann NH, Brown MD. Artificial intelligence in the diagnosis of low back pain. Orthop Clin North Amer 1991; 22: 304-14.
• 9 Wesson DR, Hink RH. Expert systems in treating substance abuse. West J Med 1990; 152: 585-7.
• 10 Plugge LA, Verhey FR, Jolles J. A desktop expert system for the differential diagnosis of dementia. An evaluation study. Int J Technol Assess Health Care 1990; 06: 147-56.
• 11 Gingrich CG, Kuespert DR, McAvoy TJ. Modeling human operators using neural networks. ISA-Ttrns 1992; 31: 81-90.
• 12 Hoffman RE. Computer simulations of neural information processing and the schizophrenia-mania dichotomy. Arch Gen Psychiatry 1987; 44: 178-88.
• 13 Glueck BC, Stroebel CF. Computers and clinical psychiatry. In: Kaplan HI, Freed-man AM, Sadock BJ. eds. Comprehensive Textbook of PsychiatrylUI. (3rd ed).. Baltimore: Williams & Wilkins Comp; 1980: 548-65.
• 14 Engle Jr RL. Attempts to use computers as a diagnostic aid in medical decision making, a thirty-year experience. Perspect Biol Med 1992; 35: 207-19.
• 15 McClelland JL, Rumelhart DE. Training hidden units. In: McClellan JL, Rumelhart DE. eds. Exploration in Parallel Disturbed Processing. Cambridge: MIT Press; 1988: 121-60.
• 16 American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders. (3rd revised ed).. Washington DC; 1987
• 17 Overall JE, Gorham D. The brief psychiatric rating scale. Psychol Rep 1962; 10: 799-812.
• 18 Hamilton MA. A rating scale for depression. J Neurol Neurosurg Psychiatry 1960; 23: 56-62.
• 19 Davis JM, Cole JO. Antipsychotic drugs. In: Kaplan HI, Freedman AM, Sadock BJ. eds. Comprehensive Textbook of Psychiatry/ II. (2nd ed). Baltimore: William & Wilkins Comp; 1975: 1921-41.
• 20 Davis JM, Cole JO. Antidpressant drugs. In: Kaplan HI, Freedman AM, Sadock Bj. eds. Comprehensive Textbook of Psychiatry! II. (2nd ed). Baltimore: William & Wilkins Comp; 1975: 1941-56.