面向眼底图像分类的训练后特征剪枝
Post-training Feature Pruning for Fundus Images Classification
提出一种训练后贪心特征剪枝（GFP）方法，可去除展平特征中的冗余维度，提升跨设备泛化性和可解释性，适用于任何视觉骨干网络的特征压缩。
原文依据 post-training feature pruning framework, termed greedy feature pruning (GFP), which removes weak or redundant dimensions from the flattened features of trained backbones
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论深度神经网络在眼底图像分类中取得了强劲性能，但其展平后的特征表示往往高度冗余。这种冗余可能导致跨成像设备的泛化能力差、可解释性降低以及模型容量利用效率低下。为解决该问题，本研究提出了一种训练后特征剪枝框架，称为贪心特征剪枝（GFP），它从已训练骨干网络的展平特征中移除弱或冗余的维度。GFP采用由训练集上的性能指标引导的贪心构建过程，并受最小特征保留比例约束，以识别紧凑且具有判别性的特征子集。实验在五个公开眼底数据集上进行，涵盖多种任务，包括糖尿病视网膜病变检测（DDR、Messidor-2）、青光眼检测（PAPILA）、多标签分类（ODIR）和多类视网膜疾病分类（RETINA），使用EfficientNetV2、ViT和CoAtNet作为骨干网络。结果表明，GFP在减少高达96%的展平特征数量的同时，持续提升了各数据集上的AUROC和AUPRC。特征可视化和定量分析证实，GFP增强了潜变量的紧凑性和可分性。此外，跨数据集评估表明，GFP提高了数据集间的可迁移性，表明具有更好的域鲁棒性。总体而言，所提出的GFP框架为压缩特征表示并提升眼底图像分类中的判别性和泛化性提供了一种简单而有效的方法。
英文原文摘要 1630 字符
Deep neural networks have achieved strong performance in fundus image classification, yet their flattened feature representations are often highly redundant. Such redundancy can lead to poor generalization across imaging devices, reduced interpretability, and inefficient use of model capacity. To address this issue, this study proposes a post-training feature pruning framework, termed greedy feature pruning (GFP) , which removes weak or redundant dimensions from the flattened features of trained backbones. GFP employs a greedy build-up process guided by performance metrics on the training set, constrained by a minimum feature keeping ratio, to identify compact yet discriminative subsets of features. Experiments are conducted on five public fundus datasets covering multiple tasks, including diabetic retinopathy detection (DDR, Messidor-2), glaucoma detection (PAPILA), multi-label classification (ODIR) and multi-class retinal disease classification (RETINA), using EfficientNetV2, ViT, and CoAtNet as backbones. Results show that GFP consistently improves AUROC and AUPRC across datasets while reducing the number of flattened features by up to 96%. Feature visualizations and quantitative analyses confirm that GFP enhances the compactness and separability of latent features. Moreover, cross-dataset evaluation demonstrates that GFP improves transferability between datasets, indicating better domain robustness. Overall, the proposed GFP framework provides a simple yet effective approach for compressing feature representations and improving both discriminability and generalization in fundus image classification.