EFCM：面向医学图像分析中大型模型部署的压缩模型高效微调
EFCM: Efficient Fine-tuning on Compressed Models for deployment of large models in medical image analysis
AI总结：
提出EFCM框架，通过无监督特征蒸馏和切片级微调，解决大型医学模型部署的内存和延迟问题，在病理图像任务中提升准确率与效率。
摘要：
深度学习大模型在医学领域的最新发展在医学图像分析和诊断中表现出显著性能，但其大量参数带来了内存和推理延迟方面的挑战。知识蒸馏提供了一种解决方案，但由于高分辨率病理图像和切片级标签，切片级梯度无法反向传播以更新学生模型。本研究提出了一种高效压缩模型微调（EFCM）框架，包含两个阶段：无监督特征蒸馏和微调。在蒸馏阶段，提出了特征投影蒸馏（FPD），并带有TransScan模块用于自适应感受野调整，以增强学生模型的知识吸收能力。在切片级微调阶段，比较了三种策略（重用CLAM、重训CLAM和端到端训练CLAM（ETC））。实验在11个与三个大型医学模型相关的下游数据集上进行：用于视网膜的RETFound、用于胸部X射线的MRM和用于组织病理学的BROW。实验结果表明，EFCM框架在处理切片级病理图像问题方面显著提高了准确性和效率，有效解决了部署大型医学模型的挑战。具体而言，与大型模型BROW相比，在TCGA-NSCLC和TCGA-BRCA数据集上，ACC提高了4.33%，AUC提高了5.2%。模型推理效率分析突出了蒸馏微调方法的高效率。
英文摘要：
The recent development of deep learning large models in medicine shows remarkable performance in medical image analysis and diagnosis, but their large number of parameters causes memory and inference latency challenges. Knowledge distillation offers a solution, but the slide-level gradients cannot be backpropagated for student model updates due to high-resolution pathological images and slide-level labels. This study presents an Efficient Fine-tuning on Compressed Models (EFCM) framework with two stages: unsupervised feature distillation and fine-tuning. In the distillation stage, Feature Projection Distillation (FPD) is proposed with a TransScan module for adaptive receptive field adjustment to enhance the knowledge absorption capability of the student model. In the slide-level fine-tuning stage, three strategies (Reuse CLAM, Retrain CLAM, and End2end Train CLAM (ETC)) are compared. Experiments are conducted on 11 downstream datasets related to three large medical models: RETFound for retina, MRM for chest X-ray, and BROW for histopathology. The experimental results demonstrate that the EFCM framework significantly improves accuracy and efficiency in handling slide-level pathological image problems, effectively addressing the challenges of deploying large medical models. Specifically, it achieves a 4.33% increase in ACC and a 5.2% increase in AUC compared to the large model BROW on the TCGA-NSCLC and TCGA-BRCA datasets. The analysis of model inference efficiency highlights the high efficiency of the distillation fine-tuning method.
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。
