基于多模式的渐进融合变压器模型,用于预测肝细胞癌患者的免疫治疗反应
Lushan Xiao1, Jiaren Wang1, Hao Cui1
1Department of Health Management, Nanfang Hospital, Southern Medical University, Guangzhou 510515, China; Department of Infectious Diseases, Nanfang Hospital, Southern Medical University, Guangzhou 510515, China.
这项研究开发了一种基于变压器的模型 (GIFT-CIP),用于预测肝细胞癌 (HCC) 患者的免疫治疗反应. 该GIFT-CIP模型准确预测患者的结果,有助于个性化治疗策略的HCC.
科学领域:
- 在瘤学瘤学.
- 医疗成像医学成像
- 人工智能的人工智能
背景情况:
- 免疫疗法显著改善肝细胞癌 (HCC) 患者的生存率.
- 预测免疫疗法反应对于优化HCC治疗策略至关重要.
- 这项研究的重点是开发HCC免疫治疗反应的预测模型.
研究的目的:
- 开发和验证基于多模式变压器的模型,用于预测HCC的免疫治疗反应.
- 评估这些模型在不同数据模式中的性能和概括能力.
主要方法:
- 五个医疗中心的HCC患者的回顾性纳入.
- 使用最小绝对收缩和选择操作员 (LASSO) 进行临床特征选择.
- 使用临床数据和CT成像特征 (内和周贴片) 训练了基于多模渐进融合变压器 (GIFT-CIP) 的模型.
- 在内部和独立的外部测试队列上验证的模型.
主要成果:
- 结合临床,内和周成像数据的GIFT-CIP模型实现了高预测性能,AUC值在0.883到0.926之间.
- 该模型有效地将患者分为低风险和高风险组,在外部队列中显示出无进展和整体存活率 (p < 0.01) 的显著差异.
- 在不同的模式和队伍中表现出强烈的概括性.
结论:
- GIFT-CIP模型提供了一种非侵入性方法,用于预测HCC患者的免疫治疗反应.
- 这个模型可以帮助临床医生指导监测和选择最佳的免疫疗法策略.
- 突出了多模式人工智能模型在精密瘤学的潜力.
更多相关视频
04:09Predicting Treatment Response to Image-Guided Therapies Using Machine Learning: An Example for Trans-Arterial Treatment of Hepatocellular Carcinoma
Published on: October 10, 2018
05:31Transradial Access Chemoembolization for Hepatocellular Carcinoma Patients
Published on: September 20, 2020
相关概念视频
Gradually Varying Flow
Nuclear Fusion
A helium nucleus has a mass that is 0.7% less than that of four hydrogen nuclei; this lost mass is converted into energy during the fusion. This reaction produces about...
Tumor Immunotherapy
Sensory Modalities
General senses refer to the broad category of sensory information detected by receptors in the body and can be further grouped into somatic and visceral senses. Somatic sensations include touch, pressure, temperature, and pain and are essential for navigating our environment and...
Predicting Molecular Geometry
Bacterial Transformation
Griffith made an unexpected discovery when he killed the pathogenic strain and mixed its remains with the live, non-pathogenic strain. Not only did the mixture kill host mice, but it also contained living pathogenic bacteria that...
