by Bermejo-Peláez D, Darias O, Pastor L, Vallés R, Diez N, Lin L, García-Villena J, Cuadrado D, Vladimirov A, Álamo E, Postigo M, Rodriguez-Dominguez M, Cantón R, Rodriguez-Tudela JL, Alastruey Izquierdo A, Marchiol A, Segura Alba ML, Cortés Cortés LJ, Bohorquez LC, Rubio JM, Dacal E, Luengo-Oroz M. PLOS Global Public Health 2026, 6(11):e0005817. doi:10.1371/journal.pgph.0005817
Summary: Lateral flow assays (LFAs) are indispensable rapid diagnostic tools, however, misinterpretation errors and unreported cases are a concern. The authors of this manuscript describe a universal AI model to digitize and interpret results from multiple LFA types through a smartphone application. The algorithm was evaluated in 17 LFA types for different diseases and manufacturers, and trained on 22,576 images captured under diverse lighting conditions with different smartphone models. The model achieved an overall AUC of 95.3% for second band detection of LFA types not previously analyzed, which was enhanced to 99.4% after training with 50 images of each LFA type. This demonstrates the feasibility of an AI-powered universal reader for interpreting LFA results from diverse test types using smartphone-captured images