一个新的人工智能工具预测了高精度结肠结肠炎患者的直肠癌风险,从而可以进行更安全的结肠镜检查。 A new AI tool predicts colorectal cancer risk in ulcerative colitis patients with high accuracy, enabling safer colonoscopy scheduling.

圣地亚哥大学研究人员开发的新型人工智能工具可以准确预测性结肠炎患者患有癌前变化的结肠直肠癌风险, 采用了来自VA医疗系统的55,000名患者的数据. A new AI tool developed by UC San Diego researchers can accurately predict colorectal cancer risk in ulcerative colitis patients with precancerous changes, using data from 55,000 patients in the VA health system. 该系统评估了将病人分为五种危险等级的四个关键因素,正确确定近一半为低危险等级,在两年内几乎99%避免癌症。 The system assesses four key factors to classify patients into five risk levels, correctly identifying nearly half as low-risk—with almost 99% avoiding cancer within two years. 这样可以更安全地延长低风险个人的结肠镜检查间隔,减少不必要的程序和病人的负担。 This could allow safer extension of colonoscopy intervals for low-risk individuals, reducing unnecessary procedures and patient burden. 该工具为目前主观风险评估提供了一种标准化的、以数据驱动的替代方法,可能有助于防止推迟采取后续行动。 The tool offers a standardized, data-driven alternative to current subjective risk assessments and may help prevent delayed follow-ups. 研究人员计划在更广泛的人群中验证并纳入遗传数据。 Researchers plan to validate it in broader populations and incorporate genetic data. 该研究于2026年2月17日发表。 The study was published on February 17, 2026.