± HEALTH In our modern society, digital devices, social media platforms, and artificial intelligence (AI) tools have b…+ CLIMATE The growing complexity of electric vehicle charging station (EVCS) operations—driven by grid constraints, ren…± HEALTH Objectives This study aim to develop, compare and internally validate machine-learning models for predicting…+ HEALTH Diabetes management faces a number of obstacles, such as fragmented healthcare data, privacy concerns, poor e…− SCIENCE Background Craniofacial soft tissue thickness (CFSTT) is a critical parameter in forensic facial reconstructi…+ HEALTH Background Technological innovation in total joint arthroplasty (TJA) has largely focused on intraoperative p…± HEALTH Infective endocarditis (IE) continues to be an often fatal condition despite improvements in cardiac surgical…± HEALTH Purpose of review Systemic sclerosis-associated interstitial lung disease (SSc-ILD) is the leading cause of d…
TruaceTracing the truth around AIFriday, September 11, 2026
Subscribe
Health·G Space·Evidence-backed gain·Published 2026-09-07

Ultrasound Habitat Radiomics for Preoperative Prediction of Invasive Breast Cancer With a DCIS Component: A Dual-center Retrospective Study

Abstract: Purpose To develop and externally validate an ultrasound-based habitat subregional radiomics model for preoperative prediction of invasive breast cancer with concomitant ductal carcinoma in situ (IBC-DCIS). Methods A total of 1063 pathologically confirmed breast cancer patients from two centers were retrospectively enrolled and divided into a training cohort (n = 637) and an external validation cohort (n = 426). Tumor regions of interest were manually delineated on two-dimensional ultrasound images and further p…

TRV-2026-1005Peer-reviewedPermanent record — cite & verify
Ultrasound Habitat Radiomics for Preoperative Prediction of Invasive Breast Cancer With a DCIS Component: A Dual-center Retrospective Study

Siliconomas Mamarios. Dr Schenone Gustavo. Enfermedades Mamarias por la inyección de sustancias by Schenone Gustavo Emilio. CC BY-SA 4.0 · https://creativecommons.org/licenses/by-sa/4.0

The quick read

Researchers developed and externally validated an ultrasound-based habitat subregional radiomics model to predict invasive breast cancer with a DCIS component before surgery. Using 1063 patients across two centers, they delineated tumors on 2D ultrasound, clustered them into three intratumoral habitats, and built machine learning models evaluated by AUC, calibration, and decision curve analysis.

By the September 2026 publication date, the combined SVM model had demonstrated favorable external validation performance and clinical net benefit, suggesting it could supplement preoperative risk stratification. Whether this retrospective performance translates to prospective clinical impact, broader populations, and changes in surgical decision-making remains untested in the supplied text.

Main points
  • Retrospective dual-center study enrolled 1063 pathologically confirmed breast cancer patients, 637 training and 426 external validation.
  • Tumor ROIs were manually delineated on 2D ultrasound and partitioned into three intratumoral habitat subregions using unsupervised clustering.
  • Radiomics features from whole tumor and subregions were selected via Pearson correlation and LASSO, with models compared by DeLong test, calibration curves, and decision curve analysis.
Gain

An ultrasound-based habitat subregional radiomics model using SVM achieved strong external validation for preoperative prediction of invasive breast cancer with concomitant DCIS, supporting preoperative risk stratification.

The rundown

The study used two-dimensional ultrasound images from two centers. After manual delineation, tumors were divided into three habitat subregions, and radiomics features were extracted from whole tumor and each subregion for model building.

The support vector machine-based combined model had the highest external validation AUC of 0.910 with 95% CI 0.883-0.936, and DeLong testing indicated it significantly outperformed imaging-based and single-region radiomics models.

Sources

Reader signal

How should this claim be treated?

The debate