用于无监督医学异常检测的尺度感知对比反向蒸馏
Scale-Aware Contrastive Reverse Distillation for Unsupervised Medical Anomaly Detection
AI总结：
针对无监督医学异常检测中反向蒸馏判别性不足且难适应异常尺度变化的问题，提出尺度感知对比反向蒸馏模型，通过对比师生学习与多尺度软加权损失实现最先进性能。
摘要：
使用深度学习的无监督异常检测因其广泛的适用性而获得了显著的研究关注，尤其是在标注异常数据稀缺的医学成像领域。尽管早期方法利用自编码器和生成对抗网络（GAN）等生成模型，但它们常因过度泛化而表现不足。近期方法探索了多种策略，包括记忆库、归一化流、自监督学习和知识蒸馏，以增强判别能力。其中，知识蒸馏，尤其是反向蒸馏，已展现出良好前景。遵循这一范式，我们提出了一种新颖的尺度感知对比反向蒸馏模型，以解决现有反向蒸馏方法的两个关键局限：特征判别性不足，以及无法处理异常尺度变化。具体而言，我们引入了一种对比师生学习方法，通过生成并探索正常分布之外的分布来获得更具判别性的表示。此外，我们设计了一种尺度适配机制，对不同尺度的对比蒸馏损失进行软加权，以应对尺度变化问题。在基准数据集上的大量实验证明了其最先进的性能，验证了所提方法的有效性。代码可在 https://github.com/MedAITech/SCRD4AD 获取。
英文摘要：
Unsupervised anomaly detection using deep learning has garnered significant research attention due to its broad applicability, particularly in medical imaging where labeled anomalous data are scarce. While earlier approaches leverage generative models like autoencoders and generative adversarial networks (GANs), they often fall short due to overgeneralization. Recent methods explore various strategies, including memory banks, normalizing flows, self-supervised learning, and knowledge distillation, to enhance discrimination. Among these, knowledge distillation, particularly reverse distillation, has shown promise. Following this paradigm, we propose a novel scale-aware contrastive reverse distillation model that addresses two key limitations of existing reverse distillation methods: insufficient feature discriminability and inability to handle anomaly scale variations. Specifically, we introduce a contrastive student-teacher learning approach to derive more discriminative representations by generating and exploring out-of-normal distributions. Further, we design a scale adaptation mechanism to softly weight contrastive distillation losses at different scales to account for the scale variation issue. Extensive experiments on benchmark datasets demonstrate state-of-the-art performance, validating the efficacy of the proposed method. Code is available at https://github.com/MedAITech/SCRD4AD.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
