A spatio-temporal Bayesian network classifier for understanding visual field deterioration

摘要: Objective: Progressive loss of the field vision is characteristic a number eye diseases such as glaucoma which leading cause irreversible blindness in world. Recently, there has been an explosion amount data being stored on patients who suffer from visual deterioration including test data, retinal image and patient demographic data. However, relatively little work modelling spatial temporal relationships common to In this paper we introduce novel method for classifying (VF) that explicitly models these relationships. Methodology: We carry out analysis our proposed spatio-temporal Bayesian classifier compare it classifiers machine learning statistical communities. These are all tested two datasets VF clinical investigate receiver operating characteristics curves, resulting network structures also make use existing anatomical knowledge order validate discovered models. Results: Results very encouraging showing comparable whilst facilitating understanding underlying within The results reveal potential using discovery ophthalmic databases, networks reflecting 'nasal step', early indicator onset glaucoma. Conclusion: outlined pave way substantial program study involving many other datasets,