BT07 Creation of a skin lesion image pipeline: key steps in developing a mobile application with DICOM® fields and metadata

Abstract A key issue facing artificial intelligence (AI) for dermatology is the lack of standardization around capture, transfer and storage skin images clinical data. To assist systematic capture data, we developed a nationally agreed proposed minimum dataset, based on Digital Imaging Communications in Medicine (DICOM®) standards. Subsequently, collaborated with an established image management system (IMS) supplier, Waba [supported by grants from our local National Health Service (NHS) innovation department], to incorporate these fields into dermatology-specific mobile application (app), facilitating real-time data transfer. We describe steps developing this app. The app integrates national Patient Administration System (SCI Store), allowing retrieval patient demographics linked hospital or Community Index number. Each care episode captures standardized including lesion views, information consent level. On completion, datasets are securely transferred IMS via WiFi GPRS signal, following fully auditable pathway. web-based Microsoft Structured Query Language server, which allows authenticated users access trust’s centralized repository nonradiological images. Adherence industry-standard format, safety reporting specific rights ensure compliance Caldicott recommendations protection For AI research, exploring conversion DICOM format. Close collaboration between software developers end users, photography, general practice, was essential design. worked design architecture through fortnightly technical meetings, scheduled progress development months 1–8. Installation functionality testing performed 9 10, followed comprehensive user 10–12. Live test-of-change rollout 13–16, continued usability refinement during period. develop conducted consensus exercise British Association Dermatology working group. Based standards, international standard medical imaging, could improve quality, availability interoperability development. In summary, have standardize transmission associated metadata. By collaborating industry partner NHS experience, iterative user-led engagement standard-setting, believe useful step toward UK database training validating systems. Future work planned evaluate output pipeline.