arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2304.04579cs.CV
收藏

医学图像中基于连贯概念的解释及其在皮肤病变诊断中的应用

Coherent Concept-based Explanations in Medical Image and Its Application to Skin Lesion Diagnosis

Universidade da Beira Interior(贝拉内格罗大学);NOVA LINCS(新里斯本大学计算机科学与信息学实验室);INESC TEC(葡萄牙科技与工程计算系统研究所);Faculdade de Engenharia da Universidade do Porto(波尔图大学工程学院)

Cristiano Patrício, João C. Neves, Luís F. Teixeira

AI总结:

本文提出一种内在可解释的皮肤病变诊断框架,通过硬注意力机制和连贯性损失增强概念激活的视觉连贯性,无需额外标注即可在分类性能上超越黑盒及现有概念模型。

摘要:

黑色素瘤的早期检测对于预防严重并发症和提高成功治疗几率至关重要。现有的用于黑色素瘤皮肤病变诊断的深度学习方法被视为黑盒模型,因为它们省略了模型预测背后的依据,损害了这些诊断方法的可信度和可接受性。提供基于概念的解释的尝试依赖于事后方法,这些方法需要额外的模型来推导解释。在本文中,我们提出了一个内在可解释的框架,通过引入硬注意力机制和连贯性损失项来提高基于概念的模型的可解释性,以确保概念编码器生成的概念激活在视觉上具有连贯性,而无需额外标注的监督。所提出的框架以人类可解释的概念及其对最终预测的各自贡献来解释其决策,并以视觉方式解释概念在图像中出现的位置。在皮肤图像数据集上的实验表明,我们的方法在皮肤病变分类方面优于现有的黑盒模型和基于概念的模型。

英文摘要:

Early detection of melanoma is crucial for preventing severe complications and increasing the chances of successful treatment. Existing deep learning approaches for melanoma skin lesion diagnosis are deemed black-box models, as they omit the rationale behind the model prediction, compromising the trustworthiness and acceptability of these diagnostic methods. Attempts to provide concept-based explanations are based on post-hoc approaches, which depend on an additional model to derive interpretations. In this paper, we propose an inherently interpretable framework to improve the interpretability of concept-based models by incorporating a hard attention mechanism and a coherence loss term to assure the visual coherence of concept activations by the concept encoder, without requiring the supervision of additional annotations. The proposed framework explains its decision in terms of human-interpretable concepts and their respective contribution to the final prediction, as well as a visual interpretation of the locations where the concept is present in the image. Experiments on skin image datasets demonstrate that our method outperforms existing black-box and concept-based models for skin lesion classification.

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。

↑