Rong Yibiao, Chen Qifeng, Jiang Zehua, Fan Zhun, Chen Haoyu
College of Engineering, Shantou University, Shantou 515063, China.
Key Laboratory of Digital Signal and Image Processing of Guangdong Provincial, Shantou University, Shantou 515063, China.
Heliyon. 2024 Feb 23;10(5):e26872. doi: 10.1016/j.heliyon.2024.e26872. eCollection 2024 Mar 15.
This study aims to estimate the regional choroidal thickness from color fundus images from convolutional neural networks in different network structures and task learning models.
1276 color fundus photos and their corresponding choroidal thickness values from healthy subjects were obtained from the Topcon DRI Triton optical coherence tomography machine. Initially, ten commonly used convolutional neural networks were deployed to identify the most accurate model, which was subsequently selected for further training. This selected model was then employed in combination with single-, multiple-, and auxiliary-task training models to predict the average and sub-region choroidal thickness in both ETDRS (Early Treatment Diabetic Retinopathy Study) grids and 100-grid subregions. The values of mean absolute error and coefficient of determination (R) were involved to evaluate the models' performance.
Efficientnet-b0 network outperformed other networks with the lowest mean absolute error value (25.61 μm) and highest R (0.7817) in average choroidal thickness. Incorporating diopter spherical, anterior chamber depth, and lens thickness as auxiliary tasks improved predicted accuracy (p-value = , , respectively). For ETDRS regional choroidal thickness estimation, multi-task model achieved better results than single task model (lowest mean absolute error = 31.10 μm vs. 33.20 μm). The multi-task training also can simultaneously predict the choroidal thickness of 100 grids with a minimum mean absolute error of 33.86 μm.
Efficientnet-b0, in combination with multi-task and auxiliary task models, achieve high accuracy in estimating average and regional macular choroidal thickness directly from color fundus photographs.
本研究旨在通过不同网络结构和任务学习模型的卷积神经网络，从彩色眼底图像估计局部脉络膜厚度。
从拓普康DRI Triton光学相干断层扫描机获取1276张健康受试者的彩色眼底照片及其相应的脉络膜厚度值。最初，部署了十个常用的卷积神经网络以识别最准确的模型，随后选择该模型进行进一步训练。然后将该选定模型与单任务、多任务和辅助任务训练模型结合使用，以预测早期糖尿病性视网膜病变研究（ETDRS）网格和100网格子区域中的平均和子区域脉络膜厚度。采用平均绝对误差值和决定系数（R）来评估模型的性能。
Efficientnet-b0网络在平均脉络膜厚度方面表现优于其他网络，平均绝对误差值最低（25.61μm），R最高（0.7817）。将球镜度数、前房深度和晶状体厚度作为辅助任务可提高预测准确性（p值分别为 ， ， ）。对于ETDRS局部脉络膜厚度估计，多任务模型比单任务模型取得了更好的结果（最低平均绝对误差 = 31.10μm对33.20μm）。多任务训练还可以同时预测100个网格的脉络膜厚度，最小平均绝对误差为33.86μm。
Efficientnet-b0与多任务和辅助任务模型相结合，可直接从彩色眼底照片中高精度地估计平均和局部黄斑脉络膜厚度。
