American College of Radiology, Reston, Virginia; Beth Israel Deaconess Medical Center, Harvard Medical School, Boston, Massachusetts.
American College of Radiology, Reston, Virginia.
J Am Coll Radiol. 2019 Sep;16(9 Pt B):1347-1350. doi: 10.1016/j.jacr.2019.06.003.
Our understanding of human health may be significantly enhanced in the near future because of the unprecedented volume of digitized health care data and the availability of artificial intelligence to mine these data for correlations that could drive new research hypotheses and improved patient care. Observational studies and randomized trials are traditional methods to generate and test hypotheses. Another way to generate research hypotheses is to use big data to reveal patterns and associations for further study. In 2018, the National Institutes of Health unveiled its Strategic Plan for Data Science, which includes a far-reaching plan for the use of big data to stimulate new research discoveries. Both researchers and physicians will need to learn and apply new skills in understanding the use of artificial intelligence and other tools, as well as in the direct application of data collection and mining in their own practices and patients.
由于数字化医疗保健数据的空前数量以及人工智能可用于挖掘这些数据以寻找可能推动新研究假设和改善患者护理的相关性,我们对人类健康的理解可能在不久的将来得到显著提高。观察性研究和随机试验是生成和检验假设的传统方法。生成研究假设的另一种方法是使用大数据揭示模式和关联,以进行进一步研究。2018 年,美国国立卫生研究院公布了其数据科学战略计划,其中包括一项利用大数据激发新研究发现的深远计划。研究人员和医生都将需要学习和应用新技能,以了解人工智能和其他工具的使用,以及在自己的实践和患者中直接应用数据收集和挖掘。