Search research articles

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

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

Intracranial hemorrhage segmentation and classification framework in computer tomography images using deep learning techniques.

Scientific reports·2025

Same author

A systematic review on intracranial aneurysm and hemorrhage detection using machine learning and deep learning techniques.

Progress in biophysics and molecular biology·2023

Same author

Tibia Fracture Healing Prediction Using First-Order Mathematical Model.

Computational and mathematical methods in medicine·2015

Same journal

Diagnosing breast masses in BI-RADS 4: construction and interpretation of an intralesional and perilesional radiomics nomogram based on digital breast tomosynthesis.

BMC medical imaging·2026

Same journal

Comparative evaluation of automated MRI-based brain volumetry software for estimating cognitive domains in early Alzheimer's disease.

BMC medical imaging·2026

Same journal

A multimodal deep learning classifier for prediction of HER2-low expression in triple-negative breast cancer.

BMC medical imaging·2026

Same journal

Multisequence MRI and clinical data-based deep learning radiomics model for predicting adjacent segment degeneration post-lumbar fusion: a retrospective multicenter study.

BMC medical imaging·2026

Same journal

Development of multiphasic CT-based delta-radiomics model for predicting postoperative recurrence risk in bladder cancer.

BMC medical imaging·2026

Same journal

Construction of a prognostic model for colorectal cancer based on clinical parameters and quantitative body composition using CT imaging.

BMC medical imaging·2026

See all related articles

Search research articles

Related Experiment Video

Updated: May 23, 2025

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Published on: November 30, 2022

Convolutional block attention gate-based Unet framework for microaneurysm segmentation using retinal fundus images.

C B Vanaja¹, P Prakasam²

¹School of Electronics Engineering, Vellore Institute of Technology, Vellore, India.

BMC Medical Imaging

|March 11, 2025

Summary

This summary is machine-generated.

A new deep learning model accurately segments microaneurysms, an early sign of diabetic retinopathy. This automated approach aids in timely diagnosis and treatment to prevent vision loss.

Keywords:

Attention gate Attention mechanism Deep learning Diabetic retinopathy Microaneurysm

More Related Videos

An Acute Retinal Model for Evaluating Blood Retinal Barrier Breach and Potential Drugs for Treatment

An Acute Retinal Model for Evaluating Blood Retinal Barrier Breach and Potential Drugs for Treatment

Published on: September 13, 2016

Using Retinal Imaging to Study Dementia

Using Retinal Imaging to Study Dementia

Published on: November 6, 2017

Related Experiment Videos

Last Updated: May 23, 2025

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Published on: November 30, 2022

An Acute Retinal Model for Evaluating Blood Retinal Barrier Breach and Potential Drugs for Treatment

An Acute Retinal Model for Evaluating Blood Retinal Barrier Breach and Potential Drugs for Treatment

Published on: September 13, 2016

Using Retinal Imaging to Study Dementia

Using Retinal Imaging to Study Dementia

Published on: November 6, 2017

Area of Science:

Medical Imaging
Artificial Intelligence
Ophthalmology

Background:

Diabetic retinopathy is a leading cause of global vision loss.
Early detection of microaneurysms (MAs) in retinal images is crucial for preventing blindness.
Manual identification of MAs is challenging due to their small size and image complexities.

Purpose of the Study:

To develop an automated method for accurate microaneurysm segmentation in retinal fundus images.
To improve the early diagnosis of diabetic retinopathy through enhanced image analysis.

Main Methods:

A novel CBAM-AG U-Net model was developed, integrating attention mechanisms (CBAM, AG) into the U-Net architecture.
The model leverages U-Net's encoder-decoder structure for comprehensive feature extraction.
Attention mechanisms focus on relevant features and significant locations for improved segmentation.

Main Results:

The CBAM-AG U-Net model achieved high segmentation accuracy on the IDRiD dataset.
Key performance metrics include an Intersection over Union (IoU) of 0.758, Dice Coefficient of 0.865, and AUC-ROC of 0.996.
The model outperformed existing methods in microaneurysm segmentation accuracy.

Conclusions:

The proposed deep learning technique shows significant promise for automated microaneurysm segmentation in diabetic retinopathy diagnosis.
This method can streamline the diagnostic process, enabling faster and more precise detection of diabetic retinopathy.
Accurate MA segmentation is vital for timely intervention and management of diabetic retinopathy.