LSS-VGG16: Diagnosis of Lumbar Spinal Stenosis With Deep Learning

Area of Science:

• Medical Imaging
• Artificial Intelligence in Medicine
• Neurosurgery

Background:

• Lumbar Spinal Stenosis (LSS) causes chronic low back pain and is often misdiagnosed as a herniated disk, requiring expert interpretation.
• Accurate diagnosis of LSS is crucial for determining appropriate surgical interventions.
• Nerve compression in LSS leads to functional loss, highlighting the need for precise diagnostic tools.

Purpose of the Study:

• To develop and evaluate a deep learning-based classification model for rapid and objective diagnosis of Lumbar Spinal Stenosis (LSS).
• To compare the performance of the proposed model against various deep learning and traditional machine learning techniques.

Main Methods:

• A retrospective study was conducted to assess a deep learning-based classification model for LSS diagnosis.
• The proposed model, LSS-VGG16, was trained and tested using medical imaging data.
• Performance was evaluated against other deep learning methods and traditional machine learning techniques.

Main Results:

• The VGG16 deep learning method achieved the highest classification success rate of 87.70%.
• The LSS-VGG16 model demonstrated superior performance compared to existing studies in the literature.

Conclusions:

• The developed LSS-VGG16 model can serve as a valuable computer-aided diagnosis system for spinal canal stenosis.
• This model offers a significant advancement for researchers and clinicians in the field of LSS diagnosis.
• The study underscores the potential of deep learning in improving diagnostic accuracy for complex neurological conditions.