Related Experiment Video
Updated: Jun 30, 2025
04:48
Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography
Published on: November 30, 2022
2.7K
Reviewing 3D convolutional neural network approaches for medical image segmentation
Ademola E Ilesanmi1, Taiwo O Ilesanmi2, Babatunde O Ajayi3
1University of Pennsylvania, 3710 Hamilton Walk, 6th Floor, Philadelphia, PA, 19104, United States.
Heliyon
|March 18, 2024
Summary
This review explores three-dimensional Convolutional Neural Networks (3D CNNs) for medical image segmentation. Findings highlight the encoder-decoder architecture
Area of Science:
- Medical Imaging and Artificial Intelligence
- Deep Learning in Healthcare
Background:
- Convolutional Neural Networks (CNNs) are crucial for clinical diagnosis and treatment.
- Three-dimensional CNNs (3D CNNs) are increasingly vital for medical image analysis, particularly for organ and anomaly segmentation.
Purpose of the Study:
- To conduct a comprehensive review of diverse 3D CNN algorithms used in medical image segmentation.
- To analyze trends, insights, and future directions in 3D CNN applications for medical imaging.
Main Methods:
- Systematic review of recent 3D CNN methodologies.
- Rigorous screening and appraisal of research papers from academic repositories.
- Analysis of network architectures and achieved accuracies for anomaly and organ segmentation.
Main Results:
- Identified prevailing trends in 3D CNN segmentation for medical images.
- The encoder-decoder network architecture is predominant in segmentation tasks.
- Detailed insights, constraints, and future research avenues were elucidated.
Conclusions:
- 3D CNN algorithms demonstrate commendable accuracy for medical image segmentation and classification.
- Findings have potential applications in clinical diagnosis and therapeutic interventions.
- The encoder-decoder framework offers a coherent methodology for medical image segmentation.
