从数字化乳腺癌组织学幻灯片对转移性复发风险的深度学习评估
I Garberis1, V Gaury2, C Saillard2
1INSERM U981, Gustave Roussy, Paris-Saclay University, Villejuif, France. ingrid-judith.GARBERIS@gustaveroussy.fr.
Nature communications
|July 2, 2025
概括
一个人工智能工具RlapsRisk BC准确地预测早期乳腺癌患者的无转移生存率. 这种人工智能模型改善了风险分层,超出了传统方法,以更好地做出治疗决策.
科学领域:
- 在瘤学瘤学.
- 计算病理学计算病理学
- 人工智能在医学中的应用
背景情况:
- 准确的风险分层对于优化早期乳腺癌 (EBC) 治疗策略至关重要.
- 现有的临床病理变量在预测雌激素受体阳性,HER2-阴性 (ER+/HER2-) EBC 的长期结果方面存在局限性.
- 需要新的方法来提高EBC管理中的预后准确性.
研究的目的:
- 开发和验证基于人工智能 (AI) 的工具,用于预测ER+/HER2-EBC的5年无转移存活率 (MFS).
- 与传统的临床病理因素相比,评估人工智能工具的预后性能.
- 评估人工智能工具在将患者分成不同的风险群体方面的临床实用性.
主要方法:
- 开发一个深度学习模型,RlapsRisk BC,利用数字化瘤幻灯片图像.
- 独立验证RlapsRisk BC模型来预测5年的MFS.
- 对AI模型的预测性能 (C指数) 与已确定的临床病理变量进行比较.
- 根据5%的MFS事件概率值,将患者分为低风险和高风险组.
- 评估RlapsRisk BC和临床病理因素对敏感性和特异性的联合性能.
主要成果:
- 与单独的临床病理变量 (临床病理变量0.76,p <0.05) 相比,RlapsRisk BC深度学习模型显示了MFS显著的独立预后价值 (C指数0.81).
- 将RlapsRisk BC与临床病理因素相结合,提高了累积灵敏度 (0.69对0.63) 和动态特异性 (0.80对0.76).
- 该模型确定的高影响区域与已知的形态特征相对应,支持其生物相关性和可解释性.
结论:
- 基于AI的RlapsRisk BC工具提供了ER+/HER2-EBC中MFS的准确和独立预测.
- RlapsRisk BC增强了风险分层超越传统方法,有助于个性化治疗决策.
- 该模型通过形态特征分析的可解释性支持其潜在的临床整合.
更多相关视频
相关概念视频
Mouse Models of Cancer Study
Mice have long served as models for studying human biology and pathology because of their phylogenetic and physiological similarity with humans. They are also easy to maintain and breed in the laboratory, and hence, many inbred strains are now available for research. Studies on mice have contributed immeasurably to our understanding of cancer biology.
The development of transgenic, knockout, and knock-in mice has led to an exponential increase in their use as model organisms in research,...
The development of transgenic, knockout, and knock-in mice has led to an exponential increase in their use as model organisms in research,...
Mouse Models of Cancer Study
Mice have long served as models for studying human biology and pathology because of their phylogenetic and physiological similarity with humans. They are also easy to maintain and breed in the laboratory, and hence, many inbred strains are now available for research. Studies on mice have contributed immeasurably to our understanding of cancer biology.
The development of transgenic, knockout, and knock-in mice has led to an exponential increase in their use as model organisms in research,...
The development of transgenic, knockout, and knock-in mice has led to an exponential increase in their use as model organisms in research,...


