Search research articles

Search research articles

Related Experiment Video

Updated: Feb 28, 2026

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Published on: December 15, 2023

A Two-Stage Localization and Refinement Neural Network Structure for Data-Efficient Microbleed Detection.

Lukas Rau¹, Oliver Granert², Nils G Margraf²

¹Faculty of Computer Science and Electrical Engineering, HAW Kiel, 24149 Kiel, Germany.

|February 27, 2026

Summary

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

Sex Differences in Prognostic Markers: Exploring Outcome Variability After Mechanical Thrombectomy in Large Vessel Occlusion Stroke.

Journal of clinical medicine·2025

Same author

Cerebral Microbleeds with a Venous Connection on 3 Tesla Susceptibility-Weighted Imaging in Persons with Alzheimer's Disease and Healthy Aging Controls.

Brain sciences·2025

Same author

Coexisting Subdural Hematoma in Cerebral Amyloid Angiopathy: A Case Series.

Neurology international·2025

Same author

Cerebrospinal Fluid β-Amyloid and τ Levels in Patients With Iatrogenic Cerebral Amyloid Angiopathy, Sporadic Cerebral Amyloid Angiopathy, Alzheimer Disease, and Controls.

Journal of the American Heart Association·2025

Same author

Simplified Edinburgh Criteria in Probable Iatrogenic Cerebral Amyloid Angiopathy.

Stroke·2025

Same author

Neuroradiological Findings in Cerebral Amyloid Angiopathy with a Particular Consideration of the Boston Criteria 2.0: An Imaging Review.

Biomolecules·2024

ABOUT JoVE

Overview Leadership Blog JoVE Help Center

AUTHORS

Publishing Process Editorial Board Scope & Policies Peer Review FAQ Submit

LIBRARIANS

Testimonials Subscriptions Access Resources Library Advisory Board FAQ

RESEARCH

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments Archive

EDUCATION

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual Faculty Resource Center Faculty Site

Terms & Conditions of Use

This summary is machine-generated.

This study introduces a new two-stage method for detecting cerebral microbleeds (CMBs) using artificial intelligence. The AI model achieves high accuracy with minimal training data, making it accessible for smaller clinics.

Area of Science:

Medical Imaging
Artificial Intelligence in Diagnostics
Neurology

Background:

Accurate detection of cerebral microbleeds (CMBs) is crucial but challenging in medical diagnostics.
Current semi-automatic methods for CMB detection often require large datasets, limiting their applicability.

Purpose of the Study:

To develop a novel, two-stage AI workflow for detecting cerebral microbleeds (CMBs).
To enable effective CMB detection using small, localized datasets, addressing limitations of current methods.

Main Methods:

A two-stage approach was implemented, starting with a 3D U-Net for initial CMB localization in SWI MRI.
A subsequent 3D convolutional neural network (CNN) was used for discriminating true CMBs from mimics.

Main Results:

Keywords:

U-Net cerebral microbleed convolutional neural network deep learning susceptibility-weighted imaging

Related Experiment Videos

Last Updated: Feb 28, 2026

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Published on: December 15, 2023

The proposed workflow achieved a high sensitivity of 97.5% on a small dataset (15 MRI scans, 40 CMBs).
Demonstrated effective CMB detection performance with limited training samples.

Conclusions:

A highly sensitive CMB detection workflow can be developed using minimal training data.
This approach empowers smaller radiological facilities to train AI models with their own datasets.
Further validation on larger, diverse datasets is recommended.