advertisement
Abstract #5634 Published in IGR 2-1
Situation assessment of glaucoma using a hybrid fuzzy neural network
Zahlmann G; Scherf M; Wegner A; Obermaier M; Mertz MIEEE Engineering in Medicine and Biology Magazine 2000; 19: 84-91
In this study, the authors report on a differential diagnostic-decision support system to aid in early detection in primary-care environments. The paper discusses in extenso how the disc, visual field, and pressure data are entered into the system. In fact, the decision-making process is based entirely on these three considerations. The special aspect of this study is, of course, that fuzzy logic is used for this network. The system was tested in data from over 2000 patients. Fine tuning of the fuzzy rule set was realized on the basis of 58 clinical test cases. To organize the verification procedure, the authors used the Bavarian ophthalmic teleconsultation network to transmit data sets from remote private ophthalmologists to the development environment. The result as compared to the diagnosis of the expert physician was that 94% of all clinical cases and 83% of case provided by the private ophthalmologists were correctly classified.
Dr. G. Zahlmann, National Research Center for Environment and Health, Medis, Neuherberg, Germany. zahlmann@gsf.de
Classification:
6 Clinical examination methods