What could we make of AI in plastic surgery education
Virve Koljonen
Abstract
To explore the possibilities of artificial intelligence (AI) text-to-picture system, DALL·E 2 was used to generated clinical photographs for medical and plastic surgery education. Generic English text was used to guide AI in three categories: subcutaneous tumor, wound and skin tumor. The most clinically accurate images were chosen for the article or for further editing. AI-generated images with variating clinical accuracy in different categories. The most accurate images were the soft-tissue tumors and the least accurate wounds. This study showed that AI text-to-picture system might be worthy tool for medical education.
Topics & Concepts
Computer scienceArtificial intelligenceMedical physicsNatural language processingSurgeryMedicineRadiologyRadiomics and Machine Learning in Medical ImagingAI in cancer detectionSocial Media in Health Education