Evaluating the Performance of Mobile Machine-Learning Platforms for Syphilis Symptom Screening

Allan-Blitz L.-T.; Konda K.A.; Reyes-Diaz E.M.; Vargas S.; Caceres C.F.; Klausner J.D. · Sexually Transmitted Diseases (2026) · DOI: 10.1097/olq.0000000000002304

Resumen

We evaluated the performance of 3 machine-learning models for classifying 39 cases of primary and secondary syphilis using associated meta-data and clinical images. All 3 models correctly classified 33 images, with an overall precent agreement of 84.6% (95% confidence interval: 69.5%–94.1%). Machine-learning models may support patient-driven symptom screening. Copyright

Instituciones: Universidad Peruana Cayetano Heredia