Related Experiment Video
Updated: Jun 27, 2026
10:09
Transtubular Endoscopic Posterolateral Decompression for L5-S1 Lumbar Lateral Disc Herniation
Published on: October 14, 2022
4.1K
Automatically Diagnosing Disk Bulge and Disk Herniation With Lumbar Magnetic Resonance Images by Using Deep
Qiong Pan1,2, Kai Zhang3,4, Lin He3
1School of Telecommunications Engineering, Xidian University, Xi'an, China.
JMIR Medical Informatics
|May 21, 2021
Summary
An automated system accurately diagnoses lumbar disk herniation and bulge using deep learning on MR images, reducing radiologist workload. This tool aids in efficient and standardized diagnosis of spinal conditions.
Area of Science:
- Medical Imaging Analysis
- Artificial Intelligence in Radiology
- Spinal Diagnostics
Background:
- Lumbar intervertebral disk (IVD) disorders like herniation and bulge cause significant pain and neurological symptoms.
- Magnetic resonance (MR) imaging is crucial for diagnosing these conditions, but interpreting large volumes of images is challenging for radiologists.
Purpose of the Study:
- To develop an automated system for diagnosing lumbar disk bulge and herniation.
- To reduce radiologist workload and improve diagnostic efficiency through automated MR image interpretation.
Main Methods:
- Utilized deep convolutional neural networks to classify axial lumbar disk MR images.
- Employed a four-step process involving automatic localization of vertebral bodies and intervertebral disk regions.
- Analyzed geometrical relationships between sagittal and axial MR images for precise identification.
Main Results:
- Achieved 100% accuracy in locating vertebral bodies and identifying corresponding disks in MR images.
- Demonstrated high classification accuracies for normal disks, disk bulges, and disk herniations across multiple lumbar levels (84.2%–92.7%).
Conclusions:
- Successfully developed an automated system for classifying lumbar disk abnormalities from MR images.
- The web-based system enhances diagnostic efficiency, standardizes reports, and shows potential for detecting other spinal conditions.
