A protocol for annotation of total body photography for machine learning to analyze skin phenotype and lesion classification
Clare Primiero, Brigid Betz‐Stablein, Nathan Ascott, Brian D’Alessandro, Séraphin Gaborit, Paul Fricker, Abigail Goldsteen, Sandra González-Villà, Katie L. Lee, Sana Nazari, Hang Thanh Nguyen, Valsamis Ntouskos, Frederik Pahde, Balázs Pataki, Josep Quintana, Susana Puig, Gisele Gargantini Rezze, Rafael García, H. Peter Soyer, Josep Malvehy
A skin-scanning system can be good at judging one spot, yet still miss the person around it. This paper asks whether the whole body, medical history, and change over time can make that judgment safer. Skin-cancer AI can outperform dermatologists in carefully controlled experiments, but what happens when it sees the whole patient, unfamiliar lesions, and real clinical context? This protocol builds the data pipeline to find out.