Following this puts new work involving it at the top of your briefing, with a note saying why it is there. Links are taken from the source record, never inferred.
A methodological proof-of-concept for computational cell classification in histopathology using optimal transport and weak supervision, demonstrating improved performance on a single independent cohort but lacking clinical validation, peer review, and prospective validation.
Early-stage in vitro mechanistic work demonstrating a novel compound design principle and cellular effects in cancer cells, without clinical trial data or in vivo efficacy validation.