Situation assessment of glaucoma using a hybrid fuzzy neural network

Glaucoma detection and evaluation through pattern recognition in standard automated perimetry data

“…Zahlmann et al [21] reported the first application of radial basis function networks for the classification of perimetry data. They demonstrated high sensitivity and specificity of the classification into 'normal' vs 'pathological' classes (98% and 95% respectively) but much less accurate differentiation of 'glaucomatous' versus 'pathological, not glaucomatous', and 'certain glaucomatous' versus 'uncertain glaucomatous' (sensitivity and specificity of about 70% for both cases).…”

Using soft computing to define standards of care in glaucoma monitoring

“…The capability of fuzzy systems to encode relevant information in form of fuzzy If-Then rules and process a vast amount of data, becomes very useful [11]. The already available hospital or practice patient charts can be used to learn and refine the knowledge base of fuzzy rule-based systems [12][13][14][15] Given that the transcripts of several knowledge acquisition sessions with the glaucoma specialists are characterized by a terminology of vague expressions like ''high IOP'', ''Severe kind of Glaucoma'', ''I want (to give the patient) some but not too much (sedation drops)'' and so on, fuzzy reasoning appears as an appropriate tool for emulating the expert thinking.…”