Studies on the Experimental Construction of a Neural Network-Based Decision Support System for Acute Abdominal Pain

The construction of a neural network -based decision support system for the diagnosis of acute abdominal pain was investigated. Different neural network algorithms were compared to define the optimal algorithms for this diagnostic classification problem. The problem of missing input data values was examined with various replacement techniques. Special attention was paid to the evaluation of confidence for the outputs of the networks. The results of the classification were also examined with different databases from two countries (Finland and Germany). The results were also compared with the results of statistical analysis. In our tests, the two best neural network algorithms, backpropagation (BP) and learning vector quantization (LVQ), classified patient cases with a very high degree of accuracy (90%), which is as high as that achieved in the best studies using other methods. A new method to present the results of the classification with the LVQ algorithm was developed. The use of k-nearest neighbours in deciding the correct class for the LVQ network was examined, and seemed to lead to better classification of cases. The evaluation of the network with databases from different countries revealed that the network must be trained with local patient material if one wishes to obtain good classification results. The comparison to statistical analyses showed that neural networks offer a viable alternative to those methods, though with this database, no clear advantage was detected in classification. INSPEC Classification: C 12.30D, C 12.90L, C 71.02, C 73.30 ACM Computing Reviews Classification: H.4.2: Decision support, I.2.1: Medicine and Science, I.2.6: Connectionism and Neural nets, I.5.1: Neural nets, J.3: Medical Information Systems National Library of Medicine Classification: W 26.55.A7, W 26.55.D2, WI 900 INSPEC Thesaurus: neural nets; artificial intelligence; decision support systems; medical diagnostic computing; patient diagnosis Medical Subject Headings: neural networks (computer); decision making, computer-assisted; diagnosis, computer-assisted; abdomen, acute; appendicitis