Accuracy of ultrawide-field fundus ophthalmoscopy-assisted deep learning for detecting treatment-naïve proliferative diabetic retinopathy

• Published in: International Ophthalmology Vol. 39; no. 10; pp. 2153 - 2159
• Main Authors: Nagasawa, Toshihiko, Tabuchi, Hitoshi, Masumoto, Hiroki, Enno, Hiroki, Niki, Masanori, Ohara, Zaigen, Yoshizumi, Yuki, Ohsugi, Hideharu, Mitamura, Yoshinori
• Format: Journal Article
• Language: English
• Published: Dordrecht Springer Science and Business Media LLC 01.10.2019; Springer Netherlands; Springer Nature B.V
• Subjects: Adult; Aged; Area Under Curve; Artificial neural networks; Deep convolutional neural network; Deep Learning; Diabetes; Diabetes mellitus; Diabetic Retinopathy; Diabetic Retinopathy - diagnosis; Diagnosis, Computer-Assisted; Diagnosis, Computer-Assisted - methods; Female; Humans; Learning algorithms; Letter to the Editor; Machine learning; Male; Medicine; Medicine & Public Health; Middle Aged; Neural networks; Ophthalmology; Ophthalmoscopy; Ophthalmoscopy - methods; Proliferative diabetic retinopathy; Retinopathy; Sensitivity; Sensitivity and Specificity; Ultrawide-field fundus ophthalmoscopy; Deep learning; Ultrawide-field fundus ophthalmoscopy; Deep convolutional neural network; Proliferative diabetic retinopathy
• ISSN: 0165-5701; 1573-2630; 1573-2630
• DOI: 10.1007/s10792-019-01074-z

Purpose We investigated using ultrawide-field fundus images with a deep convolutional neural network (DCNN), which is a machine learning technology, to detect treatment-naïve proliferative diabetic retinopathy (PDR). Methods We conducted training with the DCNN using 378 photographic images (132 PDR and 246 non-PDR) and constructed a deep learning model. The area under the curve (AUC), sensitivity, and specificity were examined. Result The constructed deep learning model demonstrated a high sensitivity of 94.7% and a high specificity of 97.2%, with an AUC of 0.969. Conclusion Our findings suggested that PDR could be diagnosed using wide-angle camera images and deep learning.