Related Experiment Video
Updated: Jun 14, 2025
06:48
Automated Segmentation of Cortical Grey Matter from T1-Weighted MRI Images
Published on: January 7, 2019
8.9K
Spine muscle auto segmentation techniques in MRI imaging: a systematic review
Hyun-Bin Kim1, Hyeon-Su Kim1, Shin-June Kim1
1Department of Biomedical Research Institute, Inha University Hospital, 27 Inhang-ro, Jung-gu, Incheon, Republic of Korea.
BMC Musculoskeletal Disorders
|September 6, 2024
Summary
This systematic review evaluates automatic spinal muscle segmentation methods using MRI. While AI and other techniques show promise, improving accuracy and robustness is key for clinical applications in musculoskeletal disorders.
Area of Science:
- Biomedical Imaging
- Medical Image Analysis
- Musculoskeletal Research
Background:
- Accurate segmentation of spinal muscles is vital for diagnosing and treating musculoskeletal disorders.
- Magnetic Resonance Imaging (MRI) is a key technique, but muscle segmentation is complex due to anatomical variability.
- This review systematically evaluates automatic spinal muscle segmentation methods.
Purpose of the Study:
- To investigate and evaluate existing methods for automatic segmentation of spinal muscles.
- To identify current challenges and future directions in spinal muscle segmentation research.
Main Methods:
- A systematic literature search was conducted on PubMed/MEDLINE using terms like 'Segmentation spine muscle'.
- Studies were selected based on relevance to automatic spinal muscle segmentation.
- 12 studies were included from an initial pool of 369.
Main Results:
- All included studies utilized MRI for spinal muscle segmentation.
- Segmentation approaches included Artificial Intelligence (AI), Proton Density Fat Fraction (PDFF), and Region of Interest (ROI).
- Study populations included healthy volunteers, back pain patients, and individuals with Adolescent Idiopathic Scoliosis (ASD).
Conclusions:
- Current spine muscle segmentation techniques require improved accuracy and precision.
- Robustness to image quality variations, artifacts, and patient specifics is essential.
- Development of annotated datasets and more advanced algorithms is crucial for clinical applications.
