[论文解读] State-of-the-Art Periorbital Distance Prediction and Disease Classification Using Periorbital Features

George R. Nahass, Sasha Hubschman+10 more|arXiv (Cornell University)|Sep 27, 2024

Medical Imaging and Analysis被引用 5

一句话总结

本论文开发了一个眼眶周围分割管道，达到最先进的准确性，并在 ID 与 OOD 设置下研究将眼眶周围距离预测作为疾病分类特征，在 OOD 任务上优于 CNN。

ABSTRACT

Periorbital distances are critical markers for diagnosing and monitoring a range of oculoplastic and craniofacial conditions. Manual measurement, however, is subjective and prone to intergrader variability. Automated methods have been developed but remain limited by standardized imaging requirements, small datasets, and a narrow focus on individual measurements. We developed a segmentation pipeline trained on a domain-specific dataset of healthy eyes and compared its performance against the Segment Anything Model (SAM) and the prior benchmark, PeriorbitAI. Segmentation accuracy was evaluated across multiple disease classes and imaging conditions. We further investigated the use of predicted periorbital distances as features for disease classification under in-distribution (ID) and out-of-distribution (OOD) settings, comparing shallow classifiers, CNNs, and fusion models. Our segmentation model achieved state-of-the-art accuracy across all datasets, with error rates within intergrader variability and superior performance relative to SAM and PeriorbitAI. In classification tasks, models trained on periorbital distances matched CNN performance on ID data (77--78\% accuracy) and substantially outperformed CNNs under OOD conditions (63--68\% accuracy vs. 14\%). Fusion models achieved the highest ID accuracy (80\%) but were sensitive to degraded CNN features under OOD shifts. Segmentation-derived periorbital distances provide robust, explainable features for disease classification and generalize better under domain shift than CNN image classifiers. These results establish a new benchmark for periorbital distance prediction and highlight the potential of anatomy-based AI pipelines for real-world deployment in oculoplastic and craniofacial care.

研究动机与目标

• 将眼眶周围距离作为眼眶整形及颅面疾病的鲁棒生物标志物进行动机阐述。
• 为健康眼睛开发一个领域特定的分割管道，并与现有模型进行比较。
• 评估预测的眼眶周围距离在 ID 与 OOD 条件下作为疾病分类特征的作用。

提出的方法

• 在健康眼数据集上训练一个领域特定的分割模型。
• 将分割性能与 Segment Anything Model (SAM) 和 PeriorbitAI 进行比较。
• 从分割输出中提取眼眶周围距离并用作浅层、CNN 和融合模型的特征进行分类。
• 评估分布内（ID）和分布外（OOD）的性能。
• 评估基于解剖结构的特征相对于 CNN 特征的可解释性和鲁棒性。

实验结果

研究问题

• RQ1 一个领域特定的分割模型是否能在多样的疾病类别和成像条件下实现最先进的眼眶周围分割精度？
• RQ2 眼眶周围距离特征在 ID 条件下是否能实现与 CNN 相竞争的疾病分类？
• RQ3 在领域转移（OOD）下，基于距离的特征是否比 CNN 特征更鲁棒？
• RQ4 融合模型对 ID 性能的影响以及在 OOD 转移下的敏感性如何？

主要发现

• 分割模型在误差处于评审之间变异范围内的情况下实现最先进的准确性，并且优于 SAM 和 PeriorbitAI。
• 基于距离的特征在 ID 疾病分类中达到 77–78% 的准确性，达到了 CNN 的性能。
• 在 OOD 条件下，基于距离的方法比 CNN 领先很大幅度（63–68% 对 14%）。
• 融合模型达到最高的 ID 准确性（约 80%），但当 OOD 转移下 CNN 特征变弱时会退化。
• 总体而言，分割模型得到的距离提供鲁棒且可解释的特征，在领域转移下比 CNN 图像分类器具有更好的泛化。