Design of an expert system for monitoring drug treatment in a psychiatric hospital

The development of a prototype system and the incorporation of a language/action model to facilitate human/computer interaction in the development of an expert system that would operate in real time and be used directly by clinicians treating psychiatric patients is described. The prototype system was implemented initially in Prolog and run on a SUN 3/50 workstation. This system was able to monitor the drug treatment of patients with any of three diagnoses: bipolar disorder, schizophrenia and major depression. The prototype was tested on subsets of patient-related data representing approximately 1200 patient-days of actual patient, laboratory, and pharmacy records, which were downloaded from the hospital's database. It was found that 100% of the clinical events of potential concern were properly identified. The language/action perspective has important implications for clinicians as end-users and the problem of acceptability of expert systems in the clinical setting.<<ETX>>

[1]  M A Overby,et al.  Psyxpert: an expert system prototype for aiding psychiatrists in the diagnosis of psychotic disorders. , 1987, Computers in biology and medicine.

[2]  D. W. Young What Makes Doctors Use Computers?: Discussion Paper , 1984, Journal of the Royal Society of Medicine.

[3]  Isaac S. Kohane,et al.  A Two-Tiered Database Cache Mechanism for a Clinician's Workstation. , 1990 .

[4]  A. Koller,et al.  Speech Acts: An Essay in the Philosophy of Language , 1969 .

[5]  Terry Winograd,et al.  Understanding computers and cognition - a new foundation for design , 1987 .

[6]  Ruven Brooks,et al.  Transferability of a Rule-Based Control Structure to a New Knowledge Domain , 1979 .

[7]  R. Duda,et al.  Expert Systems Research. , 1983, Science.

[8]  D W Young What Makes Doctors Use Computers?: Discussion Paper , 1984 .

[9]  J. F. Heiser,et al.  Design Considerations For A Clinical Psychopharmacology Advisor , 1978 .

[10]  Ralph Morelli,et al.  A Language/Action Model of Human-Computer Communication in a Psychiatric Hospital. , 1990 .

[11]  D. Servan-Schreiber,et al.  Knowledge engineering: a daily activity on a hospital ward. , 1983, Computers and biomedical research, an international journal.

[12]  D Servan-Schreiber,et al.  Artificial Intelligence and Psychiatry , 1986, The Journal of nervous and mental disease.

[13]  J.D. Bronzino,et al.  OVERSEER: a prototype expert system for monitoring drug treatment in the psychiatric clinic , 1989, IEEE Transactions on Biomedical Engineering.

[14]  P Szolovits,et al.  Artificial intelligence in medicine. Where do we stand? , 1987, The New England journal of medicine.

[15]  S J Jay,et al.  Why Doctors Don't Use Computers: Some Empirical Findings , 1986, Journal of the Royal Society of Medicine.

[16]  Ralph Morelli,et al.  Incorporating a Language/Action Design Perspective into a Computer-Based Psychiatric Alerting System. , 1989 .

[17]  M. Feinberg,et al.  Expert systems in psychiatry. , 1986, Psychopharmacology bulletin.

[18]  Isaac S. Kohane,et al.  The Children's Hospital Clinician's Workstation. , 1990 .

[19]  Peter Szolovits,et al.  Artificial Intelligence in Medicine , 1982 .

[20]  Terry Winograd A Language/Action Perspective on the Design of Cooperative Work , 1987, Hum. Comput. Interact..

[21]  E. Shortliffe,et al.  An analysis of physician attitudes regarding computer-based clinical consultation systems. , 1981, Computers and biomedical research, an international journal.