放射学和深度学习的多式融合模型整合瘤微环境准确预测乳腺癌的病理完整反应
Deqing Hong1, Jiayi Peng2, Peng Xu1
1Institute of Computational Science and Technology, Guangzhou University, Guangzhou, China (D.H., P.X., W.L., Z.Y.).
Academic radiology
|February 8, 2026
概括
在乳腺癌中准确预测病理完整反应 (pCR) 得到了新的多式模式的改进. 这种先进的工具整合了放射学和深度学习,以更好地指导新辅助化疗 (NAC) 决策.
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:
- 预测乳腺癌中新辅助化疗 (NAC) 的病理完整反应 (pCR) 是至关重要的,但由于瘤异质性和成像限制,具有挑战性.
- 现有的放射学和深度学习 (DL) 模型往往忽略了周周微环境,这是治疗反应的关键因素.
研究的目的:
- 开发和验证一个综合内放射学,周特征和DL的多式模式,用于改善乳腺癌中的PCR预测.
- 评估模型的性能与单一模式方法相比,并评估其临床实用性.
主要方法:
- 使用NAC前的MRI数据创建了一个多式模式,结合了内放射学,周特征 (9毫米扩张) 和DL模式.
- 该模型在多中心队列 (I-SPY2试验,n=929;独立队列,n=95) 上经过训练和验证,使用拉索回归和双向选择来优化特征.
- 逻辑回归被确定为集成模型 (Intra-Peri-DL) 的最佳机器学习算法.
主要成果:
- 集成的Intra-Peri-DL模型实现了高预测性能,AUC为0.888 (内部) 和0.890 (外部验证).
- 该模型在统计学上表现出优越的性能,与单模射线或DL模型相比 (P<0.05).
- 在两个队列中都观察到高灵敏度 (>0.91),表明在治疗决策中具有很强的临床应用潜力.
结论:
- 开发的多式模式有效地协同放射学和DL捕捉瘤微环境相互作用,显著提高PCR预测的准确性.
- 这种方法为乳腺癌的个性化新辅助化疗策略提供了一个有希望的,临床可行的工具.
- 该框架将成像数据与生物见解相结合,为乳腺癌患者推进精确瘤学.
更多相关视频
07:45Orthotopic Transplantation of Breast Tumors as Preclinical Models for Breast Cancer
Published on: May 18, 2020
7.1K
07:13Initiation of Metastatic Breast Carcinoma by Targeting of the Ductal Epithelium with Adenovirus-Cre: A Novel Transgenic Mouse Model of Breast Cancer
Published on: March 26, 2014
24.0K
相关概念视频
The Tumor Microenvironment
7.9K
Every normal cell or tissue is embedded in a complex local environment called stroma, consisting of different cell types, a basal membrane, and blood vessels. As normal cells mutate and develop into cancer cells, their local environment also changes to allow cancer progression. The tumor microenvironment (TME) consists of a complex cellular matrix of stromal cells and the developing tumor. The cross-talk between cancer cells and surrounding stromal cells is critical to disrupt normal tissue...
7.9K
Nuclear Fusion
33.9K
The process of converting very light nuclei into heavier nuclei is also accompanied by the conversion of mass into large amounts of energy, a process called fusion. The principal source of energy in the sun is a net fusion reaction in which four hydrogen nuclei fuse and ultimately produce one helium nucleus and two positrons.
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...
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...
33.9K
Predicting Molecular Geometry
46.1K
VSEPR Theory for Determination of Electron Pair Geometries
46.1K
Cancer Stem Cells and Tumor Maintenance
6.0K
Early diagnosis and treatment can often cure cancer. However, even with treatment, residual cells called cancer stem cells (CSC) might remain, often causing tumor recurrence. These cancer stem cells possess the potential for self-renewal and multi-lineage differentiation and are often responsible for the therapeutic resistance displayed in most cancers.
Cancer stem cells are thought to originate from tissue-specific normal stem cells or progenitor cells. The normal stem cells usually reside in...
Cancer stem cells are thought to originate from tissue-specific normal stem cells or progenitor cells. The normal stem cells usually reside in...
6.0K
Growth Models with Integration: Problem Solving
63
In population modeling, integration provides a systematic way to determine accumulated quantities from known rates of change. One such application arises in ecology, where the total weight of a fish population in a body of water is referred to as its biomass. When the rate of growth of this biomass is known as a function of time, calculus can be used to determine the total biomass at a future date.Growth Rate and Biomass FunctionLet the growth rate of the fish population be represented by a...
63
Prediction Intervals
3.4K
The interval estimate of any variable is known as the prediction interval. It helps decide if a point estimate is dependable.
However, the point estimate is most likely not the exact value of the population parameter, but close to it. After calculating point estimates, we construct interval estimates, called confidence intervals or prediction intervals. This prediction interval comprises a range of values unlike the point estimate and is a better predictor of the observed sample value, y.
However, the point estimate is most likely not the exact value of the population parameter, but close to it. After calculating point estimates, we construct interval estimates, called confidence intervals or prediction intervals. This prediction interval comprises a range of values unlike the point estimate and is a better predictor of the observed sample value, y.
3.4K
