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

Artificial Intelligence Automatically Detects Stress Urinary Incontinence from Ultrasound Videos

🧠 AI Analysis 1 min read 👁 3

Researchers have successfully tested deep learning models capable of identifying stress urinary incontinence from transperineal ultrasound videos without requiring manual frame selection.

Artificial Intelligence Automatically Detects Stress Urinary Incontinence from Ultrasound Videos🧠AI Analysis

Photo: Nature Health Sciences

Key points

  • Deep learning models analyzed 803 transperineal ultrasound recordings.
  • The V-JEPA2 foundation model achieved AUC values between 0.89 and 0.90.
  • Automated detection eliminates the need for manual frame selection.

A team of researchers at University Hospital Aachen in Germany has demonstrated that deep learning technology can automatically detect stress urinary incontinence (SUI) from transperineal ultrasound video examinations. Published in Nature Health Sciences, the retrospective study analyzed 803 patient recordings to evaluate whether automated video classification pipelines could accurately diagnose the common bladder dysfunction without the need for manual frame or region-of-interest selection.

The study evaluated two distinct video-classification architectures: an ultrasound-pretrained frame-wise encoder paired with transformer-based temporal aggregation, and a video-native foundation model known as V-JEPA2. According to the findings, both model families demonstrated strong discriminative performance, with mean area under the receiver operating characteristic curve values ranging from 0.84 to 0.90. The video-native foundation model achieved slightly higher performance in basic configuration comparisons, while anatomically guided input variants showed comparable results.

What to know: While these findings suggest that automated ultrasound analysis may eventually serve as a reliable clinical decision-support tool, the research was based on a single-center retrospective dataset. Patients concerned about urinary health should consult a qualified healthcare provider for personal diagnosis and medical advice.

Source Nature Health Sciences — nature.com/articles/s41598-026-77927-2 ↗

Contact the newsroom

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