Abstract
Keratitis is the main cause of corneal blindness worldwide. Most vision loss caused by keratitis can be avoidable via early detection and treatment. The diagnosis of keratitis often requires skilled ophthalmologists. However, the world is short of ophthalmologists, especially in resource-limited settings, making the early diagnosis of keratitis challenging. Here, we develop a deep learning system for the automated classification of keratitis, other cornea abnormalities, and normal cornea based on 6,567 slit-lamp images. Our system exhibits remarkable performance in cornea images captured by the different types of digital slit lamp cameras and a smartphone with the super macro mode (all AUCs>0.96). The comparable sensitivity and specificity in keratitis detection are observed between the system and experienced cornea specialists. Our system has the potential to be applied to both digital slit lamp cameras and smartphones to promote the early diagnosis and treatment of keratitis, preventing the corneal blindness caused by keratitis.
Highlights
Keratitis is the main cause of corneal blindness worldwide
The best algorithm achieved an area under the curve (AUC) of 0.998 (95% confidence interval [CI], 0.996–0.999), a sensitivity of 97.7%, and a specificity of 98.2% in keratitis detection
The best algorithm discriminated cornea with other abnormalities from keratitis and normal cornea with an AUC of 0.994, a sensitivity of 94.6%, and a specificity of 98.4%
Summary
Keratitis is the main cause of corneal blindness worldwide. Most vision loss caused by keratitis can be avoidable via early detection and treatment. Our system has the potential to be applied to both digital slit lamp cameras and smartphones to promote the early diagnosis and treatment of keratitis, preventing the corneal blindness caused by keratitis. Over 200,000 ophthalmologists around the world, there is a current and expected future shortfall in the number of ophthalmologists in both developing and developed countries[11] This widening gap between need and supply can affect the detection of keratitis in a timely manner, especially in remote and underserved regions[12]. Corneal blindness caused by keratitis can be completely prevented via early detection and timely treatment[8,12] To achieve this goal, in this study, we developed a deep learning system for the automated classification of keratitis, other cornea abnormalities, and normal cornea based on slit-lamp images and externally evaluated this system in three datasets of slit-lamp images and one dataset of smartphone images. We compared the performance of this system to that of cornea specialists of different levels
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have