Prediction of Refractive Error Based on Ultrawide Field Images With Deep Learning Models in Myopia Patients

Danjuan Yang; Meiyan Li; Weizhen Li; Yunzhe Wang; Lingling Niu; Yang Shen; Xiaoyu Zhang; Bo Fu; Xingtao Zhou

doi:10.3389/fmed.2022.834281

Prediction of Refractive Error Based on Ultrawide Field Images With Deep Learning Models in Myopia Patients

Front Med (Lausanne). 2022 Mar 30:9:834281. doi: 10.3389/fmed.2022.834281. eCollection 2022.

Authors

Danjuan Yang^{1

2

3

4

5}, Meiyan Li^{1

2

3

4

5}, Weizhen Li⁶, Yunzhe Wang⁷, Lingling Niu^{1

2

3

4

5}, Yang Shen^{1

2

3

4

5}, Xiaoyu Zhang^{1

2

3

4

5}, Bo Fu⁶, Xingtao Zhou^{1

2

3

4

5}

Affiliations

¹ Eye Institute and Department of Ophthalmology, Eye & ENT Hospital, Fudan University, Shanghai, China.
² NHC Key Laboratory of Myopia, Fudan University, Shanghai, China.
³ Key Laboratory of Myopia, Chinese Academy of Medical Sciences, Shanghai, China.
⁴ Shanghai Research Center of Ophthalmology and Optometry, Shanghai, China.
⁵ Shanghai Engineering Research Center of Laser and Autostereoscopic 3D for Vision Care, Shanghai, China.
⁶ School of Data Science, Fudan University, Shanghai, China.
⁷ Shanghai Medical College, Fudan University, Shanghai, China.

Abstract

Summary: Ultrawide field fundus images could be applied in deep learning models to predict the refractive error of myopic patients. The predicted error was related to the older age and greater spherical power.

Purpose: To explore the possibility of predicting the refractive error of myopic patients by applying deep learning models trained with ultrawide field (UWF) images.

Methods: UWF fundus images were collected from left eyes of 987 myopia patients of Eye and ENT Hospital, Fudan University between November 2015 and January 2019. The fundus images were all captured with Optomap Daytona, a 200° UWF imaging device. Three deep learning models (ResNet-50, Inception-v3, Inception-ResNet-v2) were trained with the UWF images for predicting refractive error. 133 UWF fundus images were also collected after January 2021 as an the external validation data set. The predicted refractive error was compared with the "true value" measured by subjective refraction. Mean absolute error (MAE), mean absolute percentage error (MAPE) and coefficient (R ²) value were calculated in the test set. The Spearman rank correlation test was applied for univariate analysis and multivariate linear regression analysis on variables affecting MAE. The weighted heat map was generated by averaging the predicted weight of each pixel.

Results: ResNet-50, Inception-v3 and Inception-ResNet-v2 models were trained with the UWF images for refractive error prediction with R ² of 0.9562, 0.9555, 0.9563 and MAE of 1.72(95%CI: 1.62-1.82), 1.75(95%CI: 1.65-1.86) and 1.76(95%CI: 1.66-1.86), respectively. 29.95%, 31.47% and 29.44% of the test set were within the predictive error of 0.75D in the three models. 64.97%, 64.97%, and 64.47% was within 2.00D predictive error. The predicted MAE was related to older age (P < 0.01) and greater spherical power(P < 0.01). The optic papilla and macular region had significant predictive power in the weighted heat map.

Conclusions: It was feasible to predict refractive error in myopic patients with deep learning models trained by UWF images with the accuracy to be improved.

Keywords: Inception-ResNet-v2; Inception-V3; ResNet-50; deep learning; myopia; refractive error prediction; ultrawide field imaging.