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 Concept Videos

You might also read

Related Articles

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

Sort by
Same author

Ultrasound-guided versus stereotactically navigated ventriculoperitoneal shunt placement: a randomized clinical trial.

Fluids and barriers of the CNS·2026
Same author

Semantic memory navigation in mild cognitive impairment: Automated markers with neural and biofluid correlates.

NeuroImage·2026
Same author

Author Correction: Large-scale self-supervised video foundation model for intelligent surgery.

NPJ digital medicine·2026
Same author

Photoacoustic device fingerprints induce bias in deep learning models.

Scientific reports·2026
Same author

Impact of higher versus lower PEEP on mortality in mechanically ventilated patients with Sepsis - A multicenter, multi-cohort observational analysis.

Journal of critical care·2026
Same author

High vasopressor doses are associated with decreased tissue oxygenation in critically ill patients: a secondary analysis of a prospective cohort.

Critical care (London, England)·2026
Same journal

SUDA: Simultaneous unsupervised knowledge distillation and adaptation of foundation models for efficient pathological image analysis.

Medical image analysis·2026
Same journal

Beyond the LUMIR challenge: The pathway to foundational registration models.

Medical image analysis·2026
Same journal

Annotation-efficient medical image segmentation via cross-latent graphs and vector-quantized memory.

Medical image analysis·2026
Same journal

HyperCOCO: Multi-sensory HyperCOgnitive COmputing for learning population level brain connectivity.

Medical image analysis·2026
Same journal

PANTHER Challenge Report: Cross-Domain Pancreatic Tumor Segmentation in Magnetic Resonance Imaging.

Medical image analysis·2026
Same journal

DGCD-3D: Difference-guided conditional diffusion model for low-field 3D MRI enhancement to assist stroke assessment.

Medical image analysis·2026
See all related articles

Related Experiment Video

Updated: May 24, 2025

Detection of Architectural Distortion in Prior Mammograms via Analysis of Oriented Patterns
13:44

Detection of Architectural Distortion in Prior Mammograms via Analysis of Oriented Patterns

Published on: August 30, 2013

42.6K

Navigating prevalence shifts in image analysis algorithm deployment.

Patrick Godau1, Piotr Kalinowski2, Evangelia Christodoulou3

  • 1German Cancer Research Center (DKFZ) Heidelberg, Division of Intelligent Medical Systems, Germany; National Center for Tumor Diseases (NCT), NCT Heidelberg, a partnership between DKFZ and University Hospital Heidelberg, Germany; Faculty of Mathematics and Computer Science, Heidelberg University, Germany; HIDSS4Health - Helmholtz Information and Data Science School for Health, Karlsruhe/Heidelberg, Germany.

Medical Image Analysis
|February 28, 2025
PubMed
Summary
This summary is machine-generated.

Domain gaps hinder medical AI. This study shows prevalence shifts impact machine learning (ML) model calibration and performance, proposing a new workflow for prevalence-aware image classification without extra data.

Keywords:
Class imbalanceDomain gapGeneralizationMedical image classificationPrevalence shift

More Related Videos

Measuring the Shape and Size of Activated Sludge Particles Immobilized in Agar with an Open Source Software Pipeline
09:27

Measuring the Shape and Size of Activated Sludge Particles Immobilized in Agar with an Open Source Software Pipeline

Published on: January 30, 2019

7.0K
Area-based Image Analysis Algorithm for Quantification of Macrophage-fibroblast Cocultures
07:05

Area-based Image Analysis Algorithm for Quantification of Macrophage-fibroblast Cocultures

Published on: February 15, 2022

2.5K

Related Experiment Videos

Last Updated: May 24, 2025

Detection of Architectural Distortion in Prior Mammograms via Analysis of Oriented Patterns
13:44

Detection of Architectural Distortion in Prior Mammograms via Analysis of Oriented Patterns

Published on: August 30, 2013

42.6K
Measuring the Shape and Size of Activated Sludge Particles Immobilized in Agar with an Open Source Software Pipeline
09:27

Measuring the Shape and Size of Activated Sludge Particles Immobilized in Agar with an Open Source Software Pipeline

Published on: January 30, 2019

7.0K
Area-based Image Analysis Algorithm for Quantification of Macrophage-fibroblast Cocultures
07:05

Area-based Image Analysis Algorithm for Quantification of Macrophage-fibroblast Cocultures

Published on: February 15, 2022

2.5K

Area of Science:

  • Medical image analysis
  • Machine learning in healthcare
  • Artificial intelligence in medicine

Background:

  • Domain gaps, particularly prevalence shifts, pose significant challenges to the clinical deployment of machine learning (ML) for medical image analysis.
  • Current research often overlooks the impact of varying class frequencies between development and deployment datasets on ML algorithm performance.
  • Disease prevalence can differ substantially across geographical locations and time periods, necessitating robust methods for AI adoption.

Purpose of the Study:

  • To investigate the consequences of failing to account for prevalence shifts in medical image classification.
  • To demonstrate the feasibility of data-driven estimation of prevalence in deployment settings.
  • To introduce a novel workflow for prevalence-aware image classification that adapts ML models to new environments using estimated prevalences.

Main Methods:

  • Evaluation of ML model calibration, decision thresholds, and performance assessment across 30 diverse medical classification tasks under varying prevalence conditions.
  • Development and validation of a data-driven approach for accurate and reliable estimation of deployment prevalences.
  • Proposal and testing of a workflow that adjusts trained classifiers using estimated deployment prevalences without requiring additional annotated data.

Main Results:

  • Lack of prevalence shift handling significantly degrades model calibration, decision thresholds, and performance assessment.
  • Prevalence estimation can be achieved accurately and reliably through data-driven methods.
  • The proposed prevalence-aware workflow improves classifier decisions and performance estimation compared to standard practices.

Conclusions:

  • Addressing prevalence shifts is critical for the reliable clinical implementation of ML in medical image analysis.
  • Data-driven prevalence estimation and adaptive classification workflows offer a practical solution to domain gap challenges.
  • The proposed method enhances the robustness and trustworthiness of AI tools in real-world healthcare settings.