Related Experiment Video
Updated: Jan 6, 2026
04:25
Author Spotlight: Bridging Gaps in Anatomy and Establishing a Foundation for Algorithmic Studies
Published on: December 15, 2023
3.7K
A deep learning-based MRI automatic detection model for spinal schwannoma and meningioma
1Institute of Medical Imaging and Engineering, University of Shanghai for Science and Technology, Shanghai, China.
Medical & Biological Engineering & Computing
|November 12, 2025
Summary
A new deep learning model automates the detection of spinal schwannomas (SCH) and meningiomas (MEN) using MRI scans. This AI tool aids in early diagnosis and reduces the burden on clinicians.
Area of Science:
- Medical Imaging
- Artificial Intelligence
- Oncology
Background:
- Spinal schwannomas (SCH) and meningiomas (MEN) are common primary spinal cord tumors.
- Their similar clinical and radiological features pose diagnostic challenges.
Purpose of the Study:
- To develop a deep learning-based object detection model for automated detection of spinal SCH and MEN using MRI.
- To improve early diagnosis and support clinical decision-making.
Main Methods:
- Retrospective analysis of 103 pathologically confirmed SCH and MEN MRI scans.
- Development of an optimized deep learning model (YOLOv8n-SKNeck) incorporating SKFusion and gnConv modules.
- Training and evaluation of the model for tumor detection.
Main Results:
- The YOLOv8n-SKNeck model achieved high performance metrics.
- Mean accuracy: 91.20%
- Mean recall: 90.92%
- Mean F1-score: 91.03% for SCH/MEN detection.
Conclusions:
- The optimized deep learning framework effectively automates the detection and differential diagnosis of spinal SCH and MEN via MRI.
- This novel method shows significant potential for advancing computer-aided diagnosis in clinical practice.
