Objective
To investigate the utility of a deep learning model in diagnosing traumatic lumbar fractures on computed tomography (CT) images.
Summary of Background
Data: CT scans are widely used as the first choice for detecting spinal fractures in patients with severe trauma. Although CT scans have high diagnostic accuracy, fractures can occasionally be missed. Recently, deep learning has been applied in various fields of medical imaging.
Methods
CT images from 480 patients (3695 vertebrae) who visited a level-one trauma center with lumbar fractures were retrospectively analyzed. The diagnostic results were confirmed by two experienced musculoskeletal radiologists and one experienced spine surgeon using magnetic resonance imaging (MRI). Deep learning networks were employed for diagnosis, with 425 cases used for training and 55 cases for testing. Sensitivity, specificity, accuracy, and the area under the receiver operating characteristic curve (AUROC) were calculated to evaluate diagnostic performance.
Results
The model successfully identified 107 out of 129 vertebrae with fractures, achieving a sensitivity of 82.95%, a specificity of 93.24%, an AUROC of 0.936, and an overall accuracy of 88.45%.
Conclusions
This study demonstrated that the deep learning model showed high accuracy in diagnosing traumatic lumbar fractures. This approach has the potential to assist spine specialists, radiologists, and trauma care experts. Further validation is needed to determine its effectiveness in clinical settings.

Numerous improvements in minimally invasive spine surgery (MISS) have been made during the past decade. Classic treatment methods have reserved surgical intervention for trauma patients with neurological compromises or instability. When used in thoracolumbar spine trauma management, MISS should achieve the similar results as classic treatment with less morbidity.(1) In the past decade, minimally invasive surgical (MIS) techniques for spine surgery have been increasingly used. The goal of minimally invasive surgery is to decrease surgical morbidity through decreased soft-tissue dissection providing similar structural stability as classic techniques. An increasing number of studies is reporting good clinical and radiographic outcomes with MIS techniques. However, the literature is lacking high-quality evidence comparing these newer techniques to classic treatments. In the future, development of techniques can expand the indications and treatment possibilities in spine trauma treatment. We reviewed the current literatures to clarify the indications of minimally invasive techniques with spinal trauma.

Purpose
A surgical treatment has been preferred in patients with unstable lumbar spine fracture-dislocation with incomplete paraplegia as it does not cause further nerve injury by regenerating and maintaining the shape of the spinal canal via the accurate reduction of fracture, and prevents additional complications by preventing neurovascular injury that secondarily occurs.¹⁾ However, the surgical treatment may be delayed or even impossible in patients with hemodynamic unstable state caused by an emergent concurrent injury. Accordingly, Stage operation was conducted on patients with unstable lumbar spine fracture-dislocation with incomplete paraplegia who had a difficulty in undergo immediate reduction and decompression due to hemodynamic unstable state caused by other concurrent injuries.
Methods
Postural reduction and minimal invasive percutaneous pedicle screw fixation L1-4 were conducted as a first stage operation, and L2-3 partial laminectomy, discectomy, and posterior fusion were then conducted as a second stage operation by applying the concept of stage operation in the department of spinal surgery.
Results
The first stage operation had a blood loss of 150cc and an operation time of 58 min. Compared to motor grade 3 shown in preoperative status, motor grade 4 was shown in L3 level or lower in a postoperative physical examination. The second stage operation was conducted two weeks later. It had an operation time of 90 min and a blood loss of 500cc. no neurologic change was further found.
Conclusion
Stage operation was conducted on patients with hemodynamic unstable state. Postural reduction and minimal invasive percutaneous pedicle screw fixation were conducted as a first stage operation to achieve the immediate reduction and stability of fracture and dislocation and the improvement of neurologic deficits. Subsequently, decompression or fusion was conducted as a second stage operation under stable systemic status for through and accurate operation.