November 18, 2026
Health news, explained with evidence
LIVE
Preventive Services Task Force Resumes Meetings Under New Leadership Blood-Based Genetic Signature May Predict CAR T-Cell Therapy Success in Lymphoma Ultraprocessed Diets May Impact Brain Reward Systems in Younger Adults Addressing Body Image and the Reality of Gluteal Anatomy Migraine and Smoking Linked to Increased Early Stroke Risk in Women CBD Oil Found Ineffective for Managing Behavioral Issues in Youths with Intellectual Disabilities Expert Advice on Hydration Gear Sparks Debate Among Recreational Runners A Structured Approach to Beginning a Strength and Cardio Routine The Role of Daily Mobility Training in Managing Sedentary Tension New Theoretical Framework Explores Memory Formation and Dementia Preventive Services Task Force Resumes Meetings Under New Leadership Blood-Based Genetic Signature May Predict CAR T-Cell Therapy Success in Lymphoma Ultraprocessed Diets May Impact Brain Reward Systems in Younger Adults Addressing Body Image and the Reality of Gluteal Anatomy Migraine and Smoking Linked to Increased Early Stroke Risk in Women CBD Oil Found Ineffective for Managing Behavioral Issues in Youths with Intellectual Disabilities Expert Advice on Hydration Gear Sparks Debate Among Recreational Runners A Structured Approach to Beginning a Strength and Cardio Routine The Role of Daily Mobility Training in Managing Sedentary Tension New Theoretical Framework Explores Memory Formation and Dementia

Study Evaluates Automated AI Framework for Breast MRI Tumor Measurement Fidelity

🧠 AI Analysis 2 min read 👁 2

Researchers have developed and validated an automated 3D MedNeXt framework for breast dynamic contrast-enhanced MRI segmentation, assessing its measurement fidelity across multiple cohorts.

🧠AI Analysis

Key points

  • Automated 3D MedNeXt framework evaluated for breast DCE-MRI segmentation fidelity.
  • Tested across multiple retrospective cohorts including internal, DUKE, BC-MRI-SEG, and BreastDCEDL_AMBL data.
  • Found high measurement correlation, though downstream predictive non-inferiority was not demonstrated.

A team of researchers has published a study in npj Digital Medicine examining an automated enhancement-aware three-dimensional framework, known as MedNeXt, designed to perform breast dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) segmentation. The investigation sought to determine whether geometric mask generation translates to accurate quantitative measurements of breast tumors.

Evaluating the framework across a retrospective multi-cohort design that included internal testing, source-held-out assessments, and external validation cohorts, the authors analyzed metrics such as tumor volume and surface diameter. Across 521 paired internal-test and DUKE cases, the median errors were reported at 9.7% for tumor volume and 1.8 mm for 3D maximum surface diameter. Furthermore, in a blinded review of 80 masks by two radiologists, the majority required no major corrections. However, the study noted that non-inferiority in downstream predictive discrimination was not fully demonstrated when comparing automatic geometry to expert geometry in response-analysis cases.

What to know: The findings suggest that automated segmentation may serve as a reliable starting point for radiologist-supervised retrospective quantitative research, though authors emphasize a distinct gap between measurement fidelity and downstream predictive equivalence. Because this was a retrospective study utilizing locked and external cohorts, prospective clinical trials are necessary to further establish safety and utility. Patients should consult qualified healthcare professionals for individual medical decisions and diagnostic evaluations.

Source Nature Health Sciences — nature.com/articles/s41746-026-03446-z ↗

Contact the newsroom

Got a tip, a proposal or a partnership question? Write to us and we will reply shortly.