Recent advances in reasoning-enhanced large language models (LLMs) show promise, yet their application in professional medicine, especially the evaluation of their reasoning process, remains ...