Rose Christian, Díaz Mark, Díaz Tomás
Department of Emergency Medicine, School of Medicine, Stanford University, Palo Alto, CA, United States.
Ethical AI, Google, New York, NY, United States.
Interact J Med Res. 2022 Aug 17;11(2):e37584. doi: 10.2196/37584.
In the 20th century, the models used to predict the motion of heavenly bodies did not match observation. Investigating this incongruity led to the discovery of dark matter-the most abundant substance in the universe. In medicine, despite years of using a data-hungry approach, our models have been limited in their ability to predict population health outcomes-that is, our observations also do not meet our expectations. We believe this phenomenon represents medicine's "dark matter"- the features with have a tremendous effect on clinical outcomes that we cannot directly observe yet. Advancing the information science of health care systems will thus require unique solutions and a humble approach that acknowledges its limitations. Dark matter changed the way the scientific community understood the universe; what might medicine learn from what it cannot yet see?
在20世纪，用于预测天体运动的模型与观测结果不相符。对这种不一致性的研究导致了暗物质的发现——宇宙中最丰富的物质。在医学领域，尽管多年来一直采用数据驱动的方法，但我们的模型在预测人群健康结果方面的能力仍然有限——也就是说，我们的观测结果也不符合我们的预期。我们认为这种现象代表了医学的“暗物质”——那些对临床结果有巨大影响但我们尚未能直接观测到的特征。因此，推进医疗保健系统的信息科学将需要独特的解决方案和一种承认其局限性的谦逊态度。暗物质改变了科学界对宇宙的理解方式；医学能从它尚未看到的东西中学到什么呢？
