Novel Keratoconus Detection Method using Smartphone

Abstract

Keratoconus is a progressive cornea disease which could cause blindness in patients if it is not detected in the early stage. In this paper, we propose a portable, low-cost and robust keratoconus detection method which is based on images captured by smartphones. A newly novel gadget has been designed and manufactured using 3-d printing for this study. The specially made gadget in combination with the smartphone’s camera and processing power, provides a robust and versatile device for detecting keratoconus. The new gadget combines the functionality of a slit lamp with the accuracy of a corneal topography device combined with the processing power of a smartphone for classification and automatic disease detection. We adopted prewitt operator for edge detection and support vector machine as our classifier to detect keratoconus using smartphone camera images. We deployed SVM classifier to distinguishes normal eyes from those with keratoconus disease. Our proposed method can diagnose keratoconus with an average accuracy of 93% for all stages of the diseases.