A Hybrid Artificial Intelligence Model for Detecting Keratoconus

Machine learning models have recently provided great promise in diagnosis of several ophthalmic disorders, including keratoconus (KCN). Keratoconus, a noninflammatory ectatic corneal disorder characterized by progressive cornea thinning, is challenging to detect as signs may be subtle. Several machine been proposed KCN, however most the are supervised and thus require large well-annotated data. This paper proposes new unsupervised model based on adapted flower pollination algorithm (FPA) k-means algorithm. We will evaluate using data collected from 5430 eyes at different stages KCN severity (1520 healthy, 331 KCN1, 1319 KCN2, 1699 KCN3 579 KCN4) Department Ophthalmology Visual Sciences, Paulista Medical School, Federal University São Paulo, Paulo Brazil 1531 (Healthy = 400, KCN1 378, KCN2 285, 200, KCN4 88) Ophthalmology, Jichi University, Tochigi Japan used accuracy metrics Precision, Recall, F-Score, Purity. compared method with three other standard algorithms k-means, Kmedoids, Spectral cluster. Based two independent datasets, outperformed algorithms, could provide improved identification status patients keratoconus.