多通道特征融合与蒙特卡洛弃权（不执行）用于不确定性感知糖尿病视网膜病变分级
Multi-Channel Feature Fusion and Monte Carlo Dropout for Uncertainty-Aware Diabetic Retinopathy Grading
- Indian Institute of Technology Patna（印度理工学院巴特那分校）
机构由 AI 辅助整理，请以论文原文为准。
AI总结：
该研究针对医疗级糖尿病视网膜病变分级需求，提出含多模块的统一流程，在APTOS-2019测试集上取得高QWK，实现分级质量、自动化与患者安全的实用权衡。
AI中文摘要：
自动化五阶段糖尿病视网膜病变（DR）分级仅需高准确率是不够的，医疗级部署要求具备病灶感知预处理、有序预测、校准不确定性及可解释性，以支撑可靠诊断系统。本文提出统一流程，采用本·格雷厄姆绿色通道CLAHE特征表示、EfficientNetV2-L有序回归器及蒙特卡洛弃权（不执行）进行不确定性驱动转诊，Grad-CAM提供与临床相关病灶对齐的视觉解释。所提方法在APTOS-2019官方测试集上的QWK达91.31%，处于近乎完美一致区间（>80%）；在20%转诊率下，366张图像中的293张被自动分级，QWK为90.40%，更复杂病例被转诊至专科医生评估，在分级质量、自动化程度与患者安全间实现实用权衡，形成稳健、可靠且可部署的医疗诊断系统。
英文摘要：
Automated five-stage diabetic retinopathy (DR) grading requires more than high accuracy alone. Medical-grade deployment calls for lesion-aware preprocessing, ordinal predictions, calibrated uncertainty, and explainability to support reliable diagnostic systems. We present a unified pipeline that addresses these requirements using a Ben-Graham-green-channel CLAHE feature representation, an EfficientNetV2-L ordinal regressor, and Monte Carlo dropout for uncertainty-driven referral. Grad-CAM provides visual explanations aligned with clinically relevant lesions. The proposed method achieves a QWK of 91.31% on the APTOS-2019 official test split, placing it within the near-perfect agreement band (>80%). At a 20% referral rate, 293 of 366 images are automatically graded with a QWK of 90.40%. More complex cases are referred for specialist assessment, demonstrating a practical trade-off among grading quality, automation, and patient safety in robust, reliable, and deployment-ready medical diagnostic systems.
