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

Search research articles

Related Experiment Video

Updated: Apr 26, 2026

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

Boundary-Aware Attention with Dual-Stream Frequency Fusion for Medical Image Segmentation.

Mingjun Geng, Runmei Hu, Xiangpeng Bi

IEEE Journal of Biomedical and Health Informatics

|April 24, 2026

Summary

BDFNet enhances medical image segmentation by integrating Boundary-Aware Attention and Dual-Stream Frequency Attention within the Mamba framework, improving boundary precision and accuracy in complex cases.

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

Epithelial FOXP3 Orchestrates O-Glycosylated IL6 Secretion to Drive Pancreatic Fibrocarcinogenesis.

Gastroenterology·2026

Same author

GatedGeoGO:Multi-Modal Geometry-Aware Network with Gated Fusion and GO Semantic Attention for Protein Function Prediction.

Journal of chemical information and modeling·2026

Same author

Comparison of Body Surface Area and Optimized Multivariate Allometric Model Approaches in Correcting Echocardiographic Aortic Dimensions for Physiological Variances: An International and Multicenter Study (CITED I).

MedComm·2026

Same author

Comprehensive Review of Contrastive and Generative Self-Supervised Learning for Small Molecular Representation.

Journal of chemical information and modeling·2026

Same author

Resolvin D2 attenuates platelet activation and thrombosis via the GPR18 receptor.

Biochemical and biophysical research communications·2026

Same author

Response of riverine N<sub>2</sub>O to anthropogenic intensity and seasonal hydrological drivers in a high-altitude basin.

Journal of environmental management·2026

Area of Science:

Medical imaging analysis
Computer vision
Deep learning for healthcare

Background:

Medical image segmentation faces challenges like low contrast and blurred boundaries.
Existing Mamba-based models struggle with precise boundary delineation and feature fusion.

Purpose of the Study:

To introduce BDFNet, a novel architecture for improved medical image segmentation.
To enhance boundary awareness and local-global feature fusion in Mamba-based segmentation.

Main Methods:

Developed BDFNet integrating Boundary-Aware Attention (BAA) and Dual-Stream Frequency Attention Module (DFAM) into the Mamba framework.
BAA uses multi-branch attention for boundary enhancement and supervision.
DFAM utilizes frequency domain analysis for fusing local and global features.

Related Experiment Videos

Last Updated: Apr 26, 2026

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

Main Results:

BDFNet demonstrated significant improvements in segmentation accuracy and boundary precision.
Achieved superior performance on ACDC cardiac, Synapse multi-organ, and ISIC skin lesion datasets.
Outperformed existing methods in cardiac, abdominal organ, and skin lesion segmentation benchmarks.

Conclusions:

BDFNet effectively addresses limitations in Mamba-based medical image segmentation.
The proposed architecture offers enhanced accuracy and boundary delineation for diverse medical imaging tasks.
Code is publicly available for further research and application.