Related Experiment Video
Updated: Sep 29, 2025
07:56
Scanning Skeletal Remains for Bone Mineral Density in Forensic Contexts
Published on: January 29, 2018
17.7K
Multi-scanner and multi-modal lumbar vertebral body and intervertebral disc segmentation database.
Yasmina Al Khalil1, Edoardo A Becherucci2, Jan S Kirschke2,3
1Biomedical Engineering Department, Eindhoven University of Technology, Eindhoven, The Netherlands.
Scientific Data
|March 24, 2022
Summary
This study introduces a new lumbar spine MRI dataset for developing automated segmentation algorithms. This database aims to improve the accuracy and efficiency of spinal disorder diagnosis and monitoring.
Area of Science:
- Medical Imaging
- Radiology
- Biomedical Engineering
Background:
- Magnetic resonance imaging (MRI) is crucial for diagnosing spinal disorders.
- Manual segmentation of spinal structures in MRI is time-consuming and error-prone.
- Automating segmentation is needed for efficient and objective quantitative MRI analysis.
Purpose of the Study:
- To address the limitations of manual segmentation in spinal MRI.
- To create a diverse, manually segmented lumbar spine MRI database.
- To facilitate the development and testing of automated segmentation algorithms across multi-domain scenarios.
Main Methods:
- Compilation of a manually segmented lumbar spine MRI database.
- Inclusion of data from multiple scanners and pulse sequences.
- Segmentation of lumbar vertebral bodies and intervertebral discs.
Main Results:
- A comprehensive database of segmented lumbar spine MRIs is now available.
- The database encompasses data from various acquisition characteristics.
- Provides a resource for validating automated segmentation techniques.
Conclusions:
- The presented database supports the development of robust automated lumbar spine segmentation.
- It aims to overcome challenges posed by heterogeneous MRI data.
- Enables more reliable and objective quantitative MRI measurements for spinal disorders.
