在资源有限的第三级诊所中,用于乳腺癌风险分层的AI支持的POCUS
Kathryn Malherbe1, Francois Malherbe2, Liana Roodt3
1Department of Imaging, Faculty of Health Sciences, Malherbe Imaging Inc, Pretoria, South Africa.
SA journal of radiology
|November 7, 2025
概括
一个新的人工智能 (AI) 系统与护理点超声波 (POCUS) 集成,在检测乳腺癌方面显示出有前途. 这种人工智能工具有助于诊断乳腺异常,特别是在医疗资源有限的地区.
科学领域:
- 医疗成像医学成像
- 医疗保健中的人工智能
- 在瘤学瘤学.
背景情况:
- 乳腺癌在南非是一个重大的公共卫生挑战,诊断延迟使患者的结果恶化.
- 护理点超声波 (POCUS) 有助于早期发现乳腺癌,但在可访问性和操作员一致性方面面临限制.
- 将人工智能 (AI) 集成到POCUS中,为增强诊断能力提供了一个可行的解决方案.
研究的目的:
- 为了评估一种新的诊断性能,本地开发了支持人工智能的POCUS系统,命名为Breast AI.
- 评估该系统预测有明显乳腺异常的女性恶性瘤的能力.
主要方法:
- 进行了一项前性队列研究,涉及159名25岁及以上的怀疑乳腺病变的女性.
- 参与者在活检前接受了乳房AI超声波,与组织病理学相比,实时恶性瘤风险得分.
- 诊断准确度指标包括灵敏度,特异性,正预测值 (PPV),F1得分和曲线下面积 (AUC).
主要成果:
- 乳腺AI在51%的恶性瘤值下达到67.2%的灵敏度,79.4%的特异性和70.3%的PPV.
- 曲线下的面积 (AUC) 为0.76,表明中等的歧视力.
- 确定了51%的最佳切断点 (F1得分 = 65.7%),像纤维腺瘤这样的良性疾病显示出较低的AI得分. 提出了一个三级风险分层模型.
结论:
- 乳腺人工智能系统在乳腺病变分类方面表现出令人鼓舞的诊断准确性,在资源有限的环境中尤其有价值.
- 这项研究支持将AI整合到POCUS中,以提高乳腺癌检测率,并为服务不足地区的临床决策提供信息.
相关概念视频
02:59Cancer Prevention
Several factors can increase the risk of cancer in an individual. About 50% of cancer cases can be prevented by adopting a healthy lifestyle, regular exercise, eating healthy, and following a modest cancer prevention diet. Epidemiological studies have consistently shown that populations with vegetable and fruit-rich diets have reduced the incidence of cancer. On the other hand, populations who have a diet rich in animal fat, red meat, junk food, or high calories are predisposed to cancer.
Some...
Some...
01:21Cancer Survival Analysis
Cancer survival analysis focuses on quantifying and interpreting the time from a key starting point, such as diagnosis or the initiation of treatment, to a specific endpoint, such as remission or death. This analysis provides critical insights into treatment effectiveness and factors that influence patient outcomes, helping to shape clinical decisions and guide prognostic evaluations. A cornerstone of oncology research, survival analysis tackles the challenges of skewed, non-normally...