Researchers used machine learning to predict chemoresistance in muscle-invasive bladder cancer, according to a study published in Experimental & Molecular Medicine. The approach integrates transcriptomic data with digital histology features, aiming to identify resistance risk prior to therapy response. If validated in prospective cohorts, models that can forecast chemoresistance could influence treatment sequencing—such as steering patients toward alternative regimens earlier rather than after failure. The work adds to a growing set of oncology efforts using multimodal data to translate molecular profiles into actionable clinical decision support, with bladder cancer serving as a high-need setting given aggressive disease behavior and response variability.