Applying a Preoperative Planning Method Based on 3D Artificial Intelligence-generated Lumbar Models in Transforaminal...

Area of Science:

• Medical Imaging
• Artificial Intelligence in Medicine
• Spinal Procedures

Background:

• Transforaminal puncture is crucial for lumbar epidural steroid injections to manage radicular pain.
• Challenges in lumbar puncture include complex 3D spinal anatomy and delicate neurovascular structures.
• Existing 3D reconstruction methods have limited practical application in puncture path planning.

Purpose of the Study:

• To introduce a novel preoperative planning method using AI-generated 3D lumbar models.
• To enhance the accuracy and efficiency of the transforaminal puncture process.

Main Methods:

• A phantom study involving 12 lumbar models and 2 surgeons.
• Comparison of an AI-based 3D model planning method against conventional fluoroscopy.
• Evaluation of puncture error (mm) and procedure duration (minutes).

Main Results:

• The AI-based method showed significantly lower average puncture error (3.33 mm) compared to fluoroscopy (5.25 mm).
• The AI-based method significantly reduced average puncture time (7.29 min) versus fluoroscopy (11.48 min).
• Both improvements were statistically significant (P < 0.001).

Conclusions:

• AI-generated 3D lumbar models improve transforaminal puncture accuracy and efficiency in phantom trials.
• This AI-driven preoperative planning technique shows potential for clinical application.
• Further clinical trials are needed to validate these findings.