Design and Construction of the Learner-centered E-learning System for Facilitating Dermoscopy Image Analysis and Diagnosis in Medical Education

Abstract

Dermoscopy is a noninvasive imaging technology that allows visualization of deeper skin structures by reducing surface reflectance, also supports more accurate diagnosis in early stage in melanoma. Compared to clinical training, learning diagnosis and related knowledge via e-learning system could be more convenient and efficient because learning materials can be collected organized based on pedagogical rationale, and users can learn personally anytime and anywhere. Thus, in this study we aim to focus on designing and implementing an e-learning system that provides well-designed images, interactive and intuitive learning environment for medical students to actively learn melanoma dermatopathology more effectively. The development of the system focuses on learning content design and instructional design. In the learning contents, the primary image resources intend to be the numerous learning materials in two steps: to reduce noise in images, and to add noise in images. After new resources are synthesized, its type can be confirmed by deep learning system. The final AUC of classifying diseases is over 0.85 with our system. In order to get the system improved, we conducted the formative evaluation to the 10 target learners. Overall, the participants are positive to the self-directed e-learning system. Moreover, some useful comments were provided for enhancing the system. The four instructional design features applied to this system include: (1) intuitive correspondence between lesion and diagnosis; (2) segmented components with structural organization; (3) terminology with link and clear tracing document; and (4) personalized system. According to the above rational and strategies, this e-learning system is expected to be a better self-directed learning platform that can afford the target users to learn flexibly, with massive resources and intuitive instructions and learn the technique of accurate diagnosis more effectively; the formative evaluation also showed that learners may have higher learning efficiency and motivation with this self-directed e-learning system. Currently the system is still undergoing, final result will be presented in the future.