IEEE/ACM Trans Comput Biol Bioinform. 2024 Jul-Aug;21(4):1031-1041. doi: 10.1109/TCBB.2023.3234100. Epub 2024 Aug 9.
The detection of optic disc and macula is an essential step for ROP (Retinopathy of prematurity) zone segmentation and disease diagnosis. This paper aims to enhance deep learning-based object detection with domain-specific morphological rules. Based on the fundus morphology, we define five morphological rules, i.e., number restriction (maximum number of optic disc and macula is one), size restriction (e.g., optic disc width: 1.05 +/- 0.13 mm), distance restriction (distance between the optic disc and macula/fovea: 4.4 +/- 0.4 mm), angle/slope restriction (optic disc and macula should roughly be positioned in the same horizontal line), position restriction (In OD, the macula is on the left side of the optic disc; vice versa for OS). A case study on 2953 infant fundus images (with 2935 optic disc instances and 2892 macula instances) proves the effectiveness of the proposed method. Without the morphological rules, naïve object detection accuracies of optic disc and macula are 0.955 and 0.719, respectively. With the proposed method, false-positive ROIs (region of interest) are further ruled out, and the accuracy of the macula is raised to 0.811. The IoU (intersection over union) and RCE (relative center error) metrics are also improved .
视盘和黄斑的检测是早产儿视网膜病变（ROP）分区和疾病诊断的重要步骤。本文旨在通过特定于领域的形态学规则来增强基于深度学习的目标检测。基于眼底形态，我们定义了五个形态学规则，即数量限制（视盘和黄斑的最大数量为一个）、大小限制（例如，视盘宽度：1.05 +/- 0.13 毫米）、距离限制（视盘和黄斑/中心凹之间的距离：4.4 +/- 0.4 毫米）、角度/斜率限制（视盘和黄斑应大致位于同一水平线上）和位置限制（在 OD 中，黄斑位于视盘的左侧；在 OS 中则相反）。对 2953 张婴儿眼底图像（有 2935 个视盘实例和 2892 个黄斑实例）的案例研究证明了所提出方法的有效性。没有形态学规则时，视盘和黄斑的原始目标检测准确率分别为 0.955 和 0.719。通过使用所提出的方法，可以进一步排除假阳性 ROI（感兴趣区域），并将黄斑的准确率提高到 0.811。IoU（交并比）和 RCE（相对中心误差）指标也得到了改善。
