Comment by arkades

I have a hard time picturing the radiologist whose reputation and malpractice rely on catching small anomalies being comfortable using a machine predicated on inferring the image contents.