Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Videos

Enhancing foundation model transfer for prostate cancer detection with patch-level contrastive learning.

Jeong Hoon Lee1, Cynthia Xinran Li2,3, Hassan Jahanandish4

  • 1Department of Radiology, Stanford University, Stanford, CA, USA. sosal@stanford.edu.

NPJ Digital Medicine
|June 5, 2026
PubMed
Summary

Related Concept Videos

Improving Translational Accuracy02:07

Improving Translational Accuracy

Base complementarity between the three base pairs of mRNA codon and the tRNA anticodon is not a failsafe mechanism. Inaccuracies can range from a single mismatch to no correct base pairing at all. The free energy difference between the correct and nearly correct base pairs can be as small as 3 kcal/ mol. With complementarity being the only proofreading step, the estimated error frequency would be one wrong amino acid in every 100 amino acids incorporated. However, error frequencies observed in...
Improving Translational Accuracy02:07

Improving Translational Accuracy

Base complementarity between the three base pairs of mRNA codon and the tRNA anticodon is not a failsafe mechanism. Inaccuracies can range from a single mismatch to no correct base pairing at all. The free energy difference between the correct and nearly correct base pairs can be as small as 3 kcal/ mol. With complementarity being the only proofreading step, the estimated error frequency would be one wrong amino acid in every 100 amino acids incorporated. However, error frequencies observed in...

You might also read

Related Articles

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

Sort by
Same author

End to end AI system for surgical gesture sequence recognition and clinical outcome prediction.

NPJ digital medicine·2026
Same author

AI model predicts patient outcomes from surgical gestures and provides insights into explainability.

npj digital surgery·2026
Same author

Prostate Cancer Detection on Micro-Ultrasound Raw Data Using a Deep Learning Neural Network.

Ultrasound in medicine & biology·2026
Same author

Frontlines of Climate Change and Global Health Inequity: How Recurring Cyclones Undermine Health, Livelihoods, and Development in the Indian Sundarbans.

Annals of global health·2026
Same author

The Current Role of Renal Mass Biopsy.

Radiologic clinics of North America·2025
Same author

Revisiting the Environment Specificity of the Protein (De)solubilization Ability of ATP.

Langmuir : the ACS journal of surfaces and colloids·2025
Same journal

A multimodal instruction dataset and benchmark for ultrasound understanding.

NPJ digital medicine·2026
Same journal

Evaluating the shift in psychiatric care: Associations between remote consultation use and clinical outcomes in a large longitudinal cohort.

NPJ digital medicine·2026
Same journal

Identifying suicide-related language in smartphone keyboard entries among high-risk adolescents.

NPJ digital medicine·2026
Same journal

Impact of a virtual nurse-led Early paLlIative Care IntervenTion (ELICIT) randomized controlled trial.

NPJ digital medicine·2026
Same journal

Towards trustworthy AI-driven cuffless blood pressure monitoring.

NPJ digital medicine·2026
Same journal

Spatially identifying regions of tumor recurrence in patients with suspected recurrent glioma using physiologic MRI and machine learning.

NPJ digital medicine·2026
See all related articles
This summary is machine-generated.

Prostate cancer detection on MRI is improved by ProViCNet, a new AI model. This vision contrastive network enhances screening and biopsy targeting, potentially reducing unnecessary biopsies.

Area of Science:

  • Medical Imaging
  • Artificial Intelligence
  • Oncology

Background:

  • Prostate cancer localization on MRI is difficult due to subtle visual cues, leading to missed diagnoses and inter-reader variability.
  • Existing vision foundation models face challenges in direct application to prostate MRI due to domain gaps and the subtle nature of prostate cancer.
  • Accurate MRI interpretation is crucial for effective screening, biopsy targeting, and focal treatment planning.

Purpose of the Study:

  • To develop and validate a weakly supervised deep learning model, ProViCNet, for enhanced prostate cancer detection and segmentation on MRI.
  • To assess ProViCNet's performance in comparison to radiologists in identifying prostate cancer lesions.
  • To evaluate the potential of integrating ProViCNet with serum PSA levels for a more specific virtual screening test.

Related Experiment Videos

Main Methods:

  • Developed ProViCNet, a prostate vision contrastive network utilizing patch-level contrastive learning on MRI data.
  • Trained and validated ProViCNet on a large dataset of 4401 patients across six cohorts, using biopsy-confirmed radiologist annotations as training labels.
  • Evaluated model performance using both biopsy and surgery-confirmed lesions, including an expert reader study comparing ProViCNet against radiologists.

Main Results:

  • ProViCNet demonstrated consistent detection and segmentation performance across multiple validation cohorts, with AUROC values ranging from 0.875 to 0.966.
  • ProViCNet outperformed expert radiologists in an MRI reader study, achieving an AUROC of 0.907 compared to the radiologists' 0.805 (p < 0.01).
  • Integration with serum PSA resulted in a virtual screening test that maintained high sensitivity for clinically significant cancers while more than doubling specificity (15% to 38%, p < 0.001) for men with PSA ≥ 4 ng/mL.

Conclusions:

  • ProViCNet shows significant potential for improving the accuracy of prostate cancer diagnosis through enhanced MRI interpretation.
  • The model's ability to outperform radiologists suggests its utility in supporting clinical decision-making for screening and treatment planning.
  • ProViCNet-based virtual screening holds promise for reducing the number of unnecessary prostate biopsies, thereby improving patient outcomes and healthcare efficiency.