ExAID: A Multimodal Explanation Framework for Computer-Aided Diagnosis of Skin Lesions

Background and Objectives: One principal impediment in successful deployment of Artificial Intelligence (AI)-based Computer-Aided Diagnosis (CAD) systems in everyday clinical workflow is their lack of transparent decision making. Although commonly used eXplainable AI (XAI) methods provide some insight into these largely opaque algorithms, yet such explanations are usually convoluted and not readily comprehensible except by highly trained AI experts. The explanation of decisions regarding the malignancy of skin lesions from dermoscopic images demands particular clarity, as the underlying medical problem definition is itself ambiguous. This work presents and evaluates ExAID (Explainable AI for Dermatology), a novel XAI framework for biomedical image analysis, providing multi-modal concept-based explanations consisting of easy-to-understand textual explanations supplemented by visual maps justifying the predictions. Methods: Our framework relies on Concept Activation Vectors (CAVs) to map humanunderstandable concepts to those learnt by an arbitrary Deep Learning based algorithm in its latent space, and Concept Localisation Maps (CLMs) to highlight concepts in the input space. This identification of relevant concepts is then used to construct fine-grained textual explanations supplemented by concept-wise location information to provide comprehensive and coherent multi-modal explanations. All decision-related information is comprehensively presented in a diagnostic interface for use in clinical routines. Moreover, the framework includes an educational mode providing dataset-level explanation statistics and tools for data and model exploration to aid medical research and education processes. Results: Through rigorous quantitative and qualitative evaluation of our framework on a range of dermoscopic image datasets such as SkinL2, Derm7pt, PH2 and ISIC, we show the utility of multimodal explanations for CAD-assisted scenarios even in case of wrong disease predictions. Conclusions: We present a new multi-modal explanation framework for biomedical image analysis on the example use-case of Melanoma classification from dermoscopic images and evaluate its utility on a row of datasets. Since comprehensible explanation is one of the cornerstones of any CAD system, we believe that ExAID will provide dermatologists an effective screening tool that they both understand and trust. Moreover, ExAID will be the basis for similar applications in other biomedical imaging fields.