Search research articles

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by

Same author

Development of an homologous recombination deficiency scoring algorithm for unmatched ovarian tumor sample based on next-generation sequencing.

Journal of gynecologic oncology·2026

Same author

The interpretable multimodal dimension reduction framework SpaHDmap enhances resolution in spatial transcriptomics.

Nature cell biology·2026

Same author

SQLE drives bladder cancer progression by boosting mitochondrial oxidative phosphorylation.

Oncogene·2025

Same author

CD160 dictates anti-PD-1 immunotherapy resistance by regulating CD8<sup>+</sup> T cell exhaustion in colorectal cancer.

Nature cell biology·2025

Same author

A New Orthotopic Mouse Model of Laryngeal Squamous Cell Carcinoma.

Head & neck·2025

Same author

Extratumoral Signs of Malignant Nonspiculate and Noncalcified Masses on Mammography: Are They Associated With Prognostic Factors in Breast Cancer?

The breast journal·2025

ABOUT JoVE

Overview Leadership Blog JoVE Help Center

AUTHORS

Publishing Process Editorial Board Scope & Policies Peer Review FAQ Submit

LIBRARIANS

Testimonials Subscriptions Access Resources Library Advisory Board FAQ

RESEARCH

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments Archive

EDUCATION

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual Faculty Resource Center Faculty Site

Terms & Conditions of Use

Search research articles

Related Experiment Video

Updated: Apr 9, 2026

Predicting Treatment Response to Image-Guided Therapies Using Machine Learning: An Example for Trans-Arterial Treatment of Hepatocellular Carcinoma

Predicting Treatment Response to Image-Guided Therapies Using Machine Learning: An Example for Trans-Arterial Treatment of Hepatocellular Carcinoma

Published on: October 10, 2018

Radiopathomic Graph Deep Learning for Multiscale Spatial-Contextual Modeling of Intratumoral Heterogeneity to Predict

Liang-Qin Zhou¹, Xin-Yi Wang¹, Ye Xu¹

¹Imaging Center, Harbin Medical University Cancer Hospital, Haping Rd No. 150, Nangang District, Harbin 150081, China.

Radiology. Artificial Intelligence

|April 8, 2026

Summary

This summary is machine-generated.

A novel radio-pathomic graph deep-learning system accurately predicts breast cancer treatment response. This explainable AI integrates imaging and pathology data for improved prediction of pathologic complete response to neoadjuvant therapy.

Keywords:

Breast Dynamic Contrast-enhanced MRI Perception Principal Component Analysis Prognosis Radiology-Pathology Integration Reconstruction Algorithm Supervised Learning Tumor Response

More Related Videos

Tracking the Mammary Architectural Features and Detecting Breast Cancer with Magnetic Resonance Diffusion Tensor Imaging

Tracking the Mammary Architectural Features and Detecting Breast Cancer with Magnetic Resonance Diffusion Tensor Imaging

Published on: December 15, 2014

Machine Learning Algorithms for Early Detection of Bone Metastases in an Experimental Rat Model

Machine Learning Algorithms for Early Detection of Bone Metastases in an Experimental Rat Model

Published on: August 16, 2020

Related Experiment Videos

Last Updated: Apr 9, 2026

Predicting Treatment Response to Image-Guided Therapies Using Machine Learning: An Example for Trans-Arterial Treatment of Hepatocellular Carcinoma

Predicting Treatment Response to Image-Guided Therapies Using Machine Learning: An Example for Trans-Arterial Treatment of Hepatocellular Carcinoma

Published on: October 10, 2018

Tracking the Mammary Architectural Features and Detecting Breast Cancer with Magnetic Resonance Diffusion Tensor Imaging

Tracking the Mammary Architectural Features and Detecting Breast Cancer with Magnetic Resonance Diffusion Tensor Imaging

Published on: December 15, 2014

Machine Learning Algorithms for Early Detection of Bone Metastases in an Experimental Rat Model

Machine Learning Algorithms for Early Detection of Bone Metastases in an Experimental Rat Model

Published on: August 16, 2020

Area of Science:

Oncology
Radiology
Pathology
Artificial Intelligence
Deep Learning

Background:

Intratumoral heterogeneity (ITH) significantly impacts breast cancer (BC) treatment response.
Predicting pathologic complete response (pCR) to neoadjuvant therapy (NAT) is crucial for personalized BC treatment.
Current prediction methods may not fully capture the complex spatial-contextual information within tumors.

Purpose of the Study:

To develop an explainable radio-pathomic graph deep-learning (RPGDL) system for multiscale spatial-contextual modeling of ITH.
To evaluate the RPGDL system's performance in predicting pCR to NAT in breast cancer.
To enhance the interpretability of deep learning models in predicting treatment outcomes.

Main Methods:

A retrospective dual-center analysis of invasive BC patients (N=1050) using pretherapeutic MRI and H&E-stained biopsy slides.
Generation of individual radiomic and pathomic graphs from imaging and pathology data, respectively.
Development and application of three distinct graph neural networks (GNNs): radiomic, pathomic, and radio-pathomic GNNs, with performance assessed using AUC, NRI, and IDI.

Main Results:

The integrated radio-pathomic GNN achieved superior performance (AUC 0.95 training, 0.91 external test) compared to single-modality GNNs (radiomic AUC 0.89/0.84, pathomic AUC 0.87/0.83).
Pathomic graphs were key drivers for predicting pCR, while radiomic graphs influenced predictions for non-pCR.
Multifaceted analyses confirmed the explainability of the GNNs' predictions.

Conclusions:

The developed RPGDL system effectively models multiscale spatial-contextual ITH.
The RPGDL system provides high-performance and explainable prediction of pCR to NAT in breast cancer.
This approach offers a promising tool for personalized treatment strategies in breast cancer.