ARTIFICIAL NEURAL-NETWORK TO ASSIST PSYCHIATRIC-DIAGNOSIS

Background. Artificial Neural Network (ANN), as a potential powerful c lassifier, was explored to assist psychiatric diagnosis of the Composi te International Diagnostic Interview (CIDI). Method. Both Back-Propag ation (BP) and Kohonen networks were developed to fit psychiatric diag nosis and programmed (using 60 cases) to classify neurosis, schizophre nia and normal people. The programmed networks were cross-tested using another 222 cases. All subjects were randomly selected from two menta l hospitals in Beijing. Results. Compared to ICD-10 diagnosis by psych iatrists, the overall kappa of BP network was 0.94 and that of Kohonen was 0.88 (both P< 0.01). In classifying patients who were difficult t o diagnose, the kappa of BP was 0.69 (P < 0.01). ANN-assisted CIDI was compared with expert system assisted CIDI (kappa = 0.72-0.76); ANN wa s more powerful than a traditional expert system. Conclusion. ANN migh t be used to improve psychiatric diagnosis.