Prediction of causative genes in inherited retinal disorder from fundus photography and autofluorescence imaging using deep learning techniques
Yu Fujinami-Yokokawa,Hideki Ninomiya,Xiao Liu,Lizhu Yang,Nikolas Pontikos,Kazutoshi Yoshitake,Takeshi Iwata,Yasunori Sato,Takeshi Hashimoto,Kazushige Tsunoda,Hiroaki Miyata,Kaoru Fujinami,Takeshi Iwata,Kazushige Tsunoda,Kaoru Fujinami,Shinji Ueno,Kazuki Kuniyoshi,Takaaki Hayashi,Mineo Kondo,Atsushi Mizota,Nobuhisa Naoi,Kei Shinoda,Shuhei Kameya,Hiroyuki Kondo,Taro Kominami,Hiroko Terasaki,Hiroyuki Sakuramoto,Satoshi Katagiri,Kei Mizobuchi,Natsuko Nakamura,Go Mawatari,Toshihide Kurihara,Kazuo Tsubota,Yozo Miyake,Kazutoshi Yoshitake,Toshihide Nishimura,Yoshihide Hayashizaki,Nobuhiro Shimozawa,Masayuki Horiguchi,Shuichi Yamamoto,Manami Kuze,Shigeki Machida,Yoshiaki Shimada,Makoto Nakamura,Takashi Fujikado,Yoshihiro Hotta,Masayo Takahashi,Kiyofumi Mochizuki,Akira Murakami,Hiroyuki Kondo,Susumu Ishida,Mitsuru Nakazawa,Tetsuhisa Hatase,Tatsuo Matsunaga,Akiko Maeda,Kosuke Noda,Atsuhiro Tanikawa,Syuji Yamamoto,Hiroyuki Yamamoto,Makoto Araie,Makoto Aihara,Toru Nakazawa,Tetsuju Sekiryu,Kenji Kashiwagi,Kenjiro Kosaki,Carninci Piero,Takeo Fukuchi,Atsushi Hayashi,Katsuhiro Hosono,Keisuke Mori,Kouji Tanaka,Koichi Furuya,Keiichirou Suzuki,Ryo Kohata,Yasuo Yanagi,Yuriko Minegishi,Daisuke Iejima,Akiko Suga,Brian P Rossmiller,Yang Pan,Tomoko Oshima,Mao Nakayama,Megumi Yamamoto,Naoko Minematsu,Daisuke Mori,Yusuke Kijima,Kentaro Kurata,Norihiro Yamada,Masayoshi Itoh,Hideya Kawaji,Yasuhiro Murakawa,Ryo Ando,Wataru Saito,Yusuke Murakami,Hiroaki Miyata,Lizhu Yang,Yu Fujinami-Yokokawa,Xiao Liu,Gavin Arno,Nikolas Pontikos,Mihori Kita,Hiroshi Hirose,Katsuyuki Sakai,Yasumasa Otori,Kazuki Yamazawa,Satomi Inoue,Takayuki Kinoshita
DOI: https://doi.org/10.1136/bjophthalmol-2020-318544
2021-04-20
British Journal of Ophthalmology
Abstract:Background/Aims To investigate the utility of a data-driven deep learning approach in patients with inherited retinal disorder (IRD) and to predict the causative genes based on fundus photography and fundus autofluorescence (FAF) imaging. Methods Clinical and genetic data from 1302 subjects from 729 genetically confirmed families with IRD registered with the Japan Eye Genetics Consortium were reviewed. Three categories of genetic diagnosis were selected, based on the high prevalence of their causative genes: Stargardt disease ( ABCA4 ), retinitis pigmentosa ( EYS ) and occult macular dystrophy ( RP1L1 ). Fundus photographs and FAF images were cropped in a standardised manner with a macro algorithm. Images for training/testing were selected using a randomised, fourfold cross-validation method. The application program interface was established to reach the learning accuracy of concordance (target: >80%) between the genetic diagnosis and the machine diagnosis ( ABCA4 , EYS , RP1L1 and normal). Results A total of 417 images from 156 Japanese subjects were examined, including 115 genetically confirmed patients caused by the three prevalent causative genes and 41 normal subjects. The mean overall test accuracy for fundus photographs and FAF images was 88.2% and 81.3%, respectively. The mean overall sensitivity/specificity values for fundus photographs and FAF images were 88.3%/97.4% and 81.8%/95.5%, respectively. Conclusion A novel application of deep neural networks in the prediction of the causative IRD genes from fundus photographs and FAF, with a high prediction accuracy of over 80%, was highlighted. These achievements will extensively promote the quality of medical care by facilitating early diagnosis, especially by non-specialists, access to care, reducing the cost of referrals, and preventing unnecessary clinical and genetic testing.
ophthalmology