Automatic diagnosis of keratitis using object localization combined with cost-sensitive deep attention convolutional neural network

• Published in: Journal of big data Vol. 10; no. 1; pp. 121 - 22
• Main Authors: Jiang, Jiewei, Liu, Wei, Pei, Mengjie, Guo, Liufei, Yang, Jingshi, Wu, Chengchao, Lu, Jiaojiao, Gao, Ruijie, Chen, Wei, Gong, Jiamin, Zhu, Mingmin, Li, Zhongwen
• Format: Journal Article
• Language: English
• Published: Cham Springer International Publishing 01.12.2023; Springer Nature B.V; SpringerOpen
• Subjects: Abnormalities; Algorithms; Artificial neural networks; Automatic diagnosis of keratitis; Automation; Big Data; Communications Engineering; Computational Science and Engineering; Computer Science; Conjunctiva; Cornea; Cost-sensitive; Data Mining and Knowledge Discovery; Database Management; Datasets; Deep attention convolutional neural network; Diagnosis; Information Storage and Retrieval; Keratitis; Localization; Localization method; Mathematical Applications in Computer Science; Medical imaging; Medical personnel; Networks; Neural networks; Object localization; Remote regions; Slit-lamp images; Object localization; Slit-lamp images; Automatic diagnosis of keratitis; Cost-sensitive; Deep attention convolutional neural network
• ISSN: 2196-1115; 2196-1115
• DOI: 10.1186/s40537-023-00800-w

Keratitis is a major cause of corneal blindness worldwide. Early identification and timely treatment of keratitis can deter the disease progression, reaching a better prognosis. The diagnosis of keratitis often requires professional ophthalmologists. However, ophthalmologists are relatively scarce and unevenly distributed, especially in underserved and remote regions, making the early diagnosis of keratitis challenging. In this study, an object localization method combined with cost-sensitive deep attention convolutional neural network (OL-CDACNN) was proposed for the automated diagnosis of keratitis. First, the single shot multibox detector (SSD) algorithm was employed to automatically locate the region of conjunctiva and cornea (Conj_Cor) on the original slit-lamp image. Then, the region of Conj_Cor was classified using a cost-sensitive deep attention convolutional network (CDACNN) to identify keratitis, other cornea abnormalities, and normal cornea. A total of 12,407 slit-lamp images collected from four clinical institutions were used to develop and evaluate the OL-CDACNN. For detecting keratitis, other cornea abnormalities, and normal cornea, the OL-CDACNN model achieved area under the receiver operating characteristic curves (AUCs) of 0.998, 0.997, and 1.000, respectively, in an internal test dataset. The comparable performance (AUCs ranged from 0.981 to 0.998) was observed in three external test datasets, further verifying its effectiveness and generalizability. Due to reliable performance, our model has a high potential to provide an accurate diagnosis and prompt referral for a patient with keratitis in an automated fashion.