一种模仿皮肤科医生以准确鉴别诊断临床图像中皮肤病的新型多任务模型
A Novel Multi-Task Model Imitating Dermatologists for Accurate Differential Diagnosis of Skin Diseases in Clinical Images
- DAMO Academy, Alibaba Group（阿里巴巴集团达摩院）
- Hupan Lab（湖畔实验室）
- Beijing Normal University（北京师范大学）
- Sir Run Run Shaw Hospital（邵逸夫医院）
- The Second Affiliated Hospital Zhejiang University School of Medicine（浙江大学医学院附属第二医院）
- Xiangya Hospital Central South University（中南大学湘雅医院）
机构由 AI 辅助整理，请以论文原文为准。
AI总结：
本文提出 DermImitFormer 多任务模型，通过模拟皮肤科医生预测部位、皮损属性与疾病，并引入皮损选择和交叉交互机制，结合新建大规模数据集实现皮肤病临床图像的先进鉴别诊断。
AI中文摘要：
皮肤病是最普遍的健康问题之一，准确的计算机辅助诊断方法对皮肤科医生和患者都具有重要意义。然而，现有大多数方法忽视了皮肤病诊断所需的关键领域知识。为弥补这一差距，本文提出一种名为 DermImitFormer 的新型多任务模型，通过模仿皮肤科医生的诊断流程和策略进行建模。借助多任务学习，该模型除预测疾病本身外，还同时预测身体部位和皮损属性，从而提高诊断准确性并增强诊断可解释性。所设计的皮损选择模块模仿皮肤科医生的放大动作，可从含噪背景中有效突出局部皮损特征。此外，所提出的交叉交互模块显式建模身体部位、皮损属性与疾病之间复杂的诊断推理关系。为更稳健地评估所提方法，本文建立了一个大规模皮肤病临床图像数据集，其病例数量显著多于现有数据集。在三个不同数据集上开展的大量实验一致表明，所提方法取得了最先进的识别性能。
英文摘要：
Skin diseases are among the most prevalent health issues, and accurate computer-aided diagnosis methods are of importance for both dermatologists and patients. However, most of the existing methods overlook the essential domain knowledge required for skin disease diagnosis. A novel multi-task model, namely DermImitFormer, is proposed to fill this gap by imitating dermatologists' diagnostic procedures and strategies. Through multi-task learning, the model simultaneously predicts body parts and lesion attributes in addition to the disease itself, enhancing diagnosis accuracy and improving diagnosis interpretability. The designed lesion selection module mimics dermatologists' zoom-in action, effectively highlighting the local lesion features from noisy backgrounds. Additionally, the presented cross-interaction module explicitly models the complicated diagnostic reasoning between body parts, lesion attributes, and diseases. To provide a more robust evaluation of the proposed method, a large-scale clinical image dataset of skin diseases with significantly more cases than existing datasets has been established. Extensive experiments on three different datasets consistently demonstrate the state-of-the-art recognition performance of the proposed approach.
