发表机构
International Institute of Information Technology Hyderabad（海得拉巴国际信息技术研究所）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
提出模块化多层次框架，整合病灶、患者和人群上下文，引入解剖部位感知掩码变换器及可学习人口统计嵌入，在 SIIM-ISIC 2020 上分别提升特异性 17.15% 和 7.14%，实现恶性黑色素瘤的循证鉴别诊断。
AI 中文摘要
我们提出了一个模块化、多层次的框架，用于恶性黑色素瘤的鉴别诊断。该框架整合了病灶、患者和人群层面的上下文信息与证据，支持在每个层面进行决策。我们引入了一个解剖部位感知的掩码变换器，通过考虑患者体内数量可变的全部病灶及其发生部位，有效建模患者上下文。此外，我们通过可学习的 demographics 嵌入整合患者元数据，以捕获人群统计信息。通过大量实验，我们探讨了特定信息对决策过程的影响，并考察了考虑不同类型信息时指标间的权衡。使用 SIIM-ISIC 2020 数据集的验证结果表明，结合病灶上下文与位置和元数据分别将特异性提高了 17.15% 和 7.14%，同时提升了平衡准确率。代码可在该 https URL 获取。
英文摘要
We present a modular and multi-level framework for the differential diagnosis of malignant melanoma. Our framework integrates contextual information and evidence at the lesion, patient, and population levels, enabling decision-making at each level. We introduce an anatomic-site aware masked transformer, which effectively models the patient context by considering all lesions in a patient, which can be variable in count, and their site of incidence. Additionally, we incorporate patient metadata via learnable demographics embeddings to capture population statistics. Through extensive experiments, we explore the influence of specific information on the decision-making process and examine the tradeoff in metrics when considering different types of information. Validation results using the SIIM-ISIC 2020 dataset indicate including the lesion context with location and metadata improves specificity by 17.15% and 7.14%, respectively, while enhancing balanced accuracy. The code is available at https://github.com/narenakash/meldd.
