[论文解读] The Skincare project, an interactive deep learning system for differential diagnosis of malignant skin lesions. Technical Report

Daniel Sonntag, Fabrizio Nunnari+1 more|arXiv (Cornell University)|May 19, 2020

Cutaneous Melanoma Detection and Management参考文献 41被引用 4

一句话总结

Skincare项目展示了一种用于恶性皮肤病变差异诊断的交互式深度学习系统，整合皮肤镜图像与智能手机宏观图像，以支持临床医生和患者。该系统通过注意力图、分割和特征提取实现实时、可解释的AI驱动诊断，经验证的模型在20,000例匿名ISIC病例上达到最先进性能。

ABSTRACT

A shortage of dermatologists causes long wait times for patients who seek dermatologic care. In addition, the diagnostic accuracy of general practitioners has been reported to be lower than the accuracy of artificial intelligence software. This article describes the Skincare project (H2020, EIT Digital). Contributions include enabling technology for clinical decision support based on interactive machine learning (IML), a reference architecture towards a Digital European Healthcare Infrastructure (also cf. EIT MCPS), technical components for aggregating digitised patient information, and the integration of decision support technology into clinical test-bed environments. However, the main contribution is a diagnostic and decision support system in dermatology for patients and doctors, an interactive deep learning system for differential diagnosis of malignant skin lesions. In this article, we describe its functionalities and the user interfaces to facilitate machine learning from human input. The baseline deep learning system, which delivers state-of-the-art results and the potential to augment general practitioners and even dermatologists, was developed and validated using de-identified cases from a dermatology image data base (ISIC), which has about 20000 cases for development and validation, provided by board-certified dermatologists defining the reference standard for every case. ISIC allows for differential diagnosis, a ranked list of eight diagnoses, that is used to plan treatments in the common setting of diagnostic ambiguity. We give an overall description of the outcome of the Skincare project, and we focus on the steps to support communication and coordination between humans and machine in IML. This is an integral part of the development of future cognitive assistants in the medical domain, and we describe the necessary intelligent user interfaces.

研究动机与目标

• 通过开发AI辅助诊断工具，解决皮肤科医生短缺和患者等待时间过长的问题。
• 通过基于深度学习的决策支持，提高全科医生的诊断准确性。
• 创建一种交互式机器学习系统，整合人类反馈以优化模型预测。
• 开发可扩展的、可通过API访问的架构，实现临床部署并集成至数字医疗基础设施。
• 通过结合影像数据与临床记录，实现全面的患者评估，以提升诊断推理能力。

提出的方法

• 基于20,000例来自ISIC数据集的匿名皮肤镜图像训练深度学习模型，涵盖八种皮肤病变类型。
• 采用注意力图提升可解释性，通过可视化感兴趣区域和模型推理过程。
• 支持多种输入模态：皮肤镜图像（显微）和宏观智能手机图像，以提升可及性。
• 实现REST API，提供二分类（良性 vs. 恶性）、病灶分割和特定特征分割（如球状体、条纹）等端点。
• 集成与GUI无关的后端与基于Web的界面，使临床医生能够交互式标注、更正并提供反馈，以提升模型性能。
• 采用模块化软件架构，包含可配置的服务器、模型权重和静态HTML内容，以支持可扩展性与部署。

实验结果

研究问题

• RQ1 交互式深度学习系统能否在初级 care 设置中提升皮肤癌诊断的准确性？
• RQ2 可解释AI技术（如注意力图）在临床决策支持中如何增强信任度与可用性？
• RQ3 结合宏观与皮肤镜图像在多大程度上能提升差异诊断的准确性？
• RQ4 如何有效将人类在回路中的反馈整合到临床AI系统中，以提升模型泛化能力？
• RQ5 何种技术架构能够实现AI工具在真实临床工作流中的可扩展部署？

主要发现

• 该系统在基于20,000例匿名ISIC病例训练的深度学习模型支持下，在八种皮肤病变类型的差异诊断中达到最先进性能。
• 注意力图提供了可解释的可视化，突出显示相关病变特征，提升了模型预测的透明度和临床医生对预测结果的信任度。
• 该系统支持二分类（良性 vs. 恶性）以及对球状体、条纹等病变特征的细粒度分割，精度极高。
• REST API支持通过标准网络协议无缝集成至临床工作流，实现实时分类与特征提取。
• 交互式学习界面允许临床医生更正预测并提供反馈，通过人机协同学习实现模型的持续优化。
• 系统设计具备可扩展性与可扩展性，模块化架构支持在多样化的临床与非临床环境中部署。