AI and the World's Hard Problems · Medicine & drug discovery · Nature (computational biology) · 2026-10-26
Machine learning model for predicting regional lymph node metastasis in acral melanoma
Acral melanoma is an aggressive subtype of melanoma with delayed diagnosis and poor prognosis, and non-invasively assessing regional lymph node metastasis is a clinical challenge. The authors developed an interpretable machine learning model using 189 patients split into training and internal validation cohorts, with five predictors: Breslow thickness, lymph-node short-axis diameter, contrast-enhanced ultrasound findings, vascular invasion and PET-CT findings. The selected GradientBoost model achieved an AUC of 0.868 (95% CI 0.769–0.967) and a sensitivity of 0.889 in the internal validation cohort.