Development of a Deep-Learning Model for Diagnosing Lumbar Spinal Stenosis Based on CT Images.

STUDY DESIGN: Retrospective study. OBJECTIVES: This study aimed to develop an initial deep-learning (DL) model based on computerized tomography (CT) scans for diagnosing lumbar spinal stenosis. SUMMARY OF BACKGROUND DATA: Magnetic resonance imaging is commonly used for diagnosing lumbar spinal stenosis due to its high soft tissue resolution, but CT is more portable, cost-effective, and has wider regional coverage. Using DL models to improve the accuracy of CT diagnosis can effectively reduce missed diagnoses and misdiagnoses in clinical practice. MATERIALS AND METHODS: Axial lumbar spine CT scans obtained between March 2022 and September 2023 were included. The data set was divided into a training set (62.3%), a validation set (22.9%), and a control set (14.8%). All data were labeled by two spine surgeons using the widely accepted grading system for lumbar spinal stenosis. The training and validation sets were used to annotate the regions of interest by the two spine surgeons. First, a region of interest detection model and a convolutional neural network classifier were trained using the training set. After training, the model was preliminarily evaluated using a validation set. Finally, the performance of the DL model was evaluated on the control set, and a comparison was made between the model and the classification performance of specialists with varying levels of experience. RESULTS: The central stenosis grading accuracies of DL Model Version 1 and DL Model Version 2 were 88% and 83%, respectively. The lateral recess grading accuracies of DL Model Version 1 and DL Model Version 2 were 75% and 71%, respectively. CONCLUSIONS: Our preliminarily developed DL system for assessing the degree of lumbar spinal stenosis in CT, including the central canal and lateral recess, has shown similar accuracy to experienced specialist physicians. This holds great value for further development and clinical application.

Osteoperiosteal decortication and bone grafting combined with wave plating for femoral shaft aseptic atrophic nonunion after intramedullary nailing.

OBJECTIVE: Femoral shaft nonunion after intramedullary nailing fixation remains a challenge for orthopedic surgeons. This study evaluated osteoperiosteal decortication and iliac bone grafting combined with wave plating as a treatment for aseptic atrophic nonunion of the femoral shaft after intramedullary nailing. METHODS: The study included 22 patients (two female, 20 male; mean age 40.8 years [range, 19-61]) with aseptic atrophic nonunion of the femoral shaft who underwent osteoperiosteal decortication and iliac bone grafting combined with wave plating between January 2016 and January 2020. Cases with infected nonunion, hypertrophic nonunion, and nonunion after plate osteosynthesis were excluded. Clinical outcomes were assessed retrospectively using the Samantha X-ray and Paley scale scores. RESULTS: The mean follow-up period was 18.8 months (range, 12-35). Bone union was achieved in all patients. The average clinical healing time was 5.7 months (range, 3-14). The mean Samantha X-ray score was 5.7 ± 0.45 and the Paley scores indicated excellent or good clinical results in all cases. One patient developed sterile wound leakage that resolved after a dressing change. CONCLUSIONS: Osteoperiosteal decortication and bone grafting combined with wave plating is an excellent treatment for aseptic atrophic nonunion of the femoral shaft after intramedullary nailing.

[Effectiveness of tunnel osteogenesis technique in treatment of aseptic non-hypertrophic nonunion of femoral shaft].

Objective: To investigate the effectiveness of tunnel osteogenesis technique combined with locking plate in the treatment of aseptic non-hypertrophic nonunion of femoral shaft. Methods: The clinical data of 23 cases of aseptic non-hypertrophic nonunion of femoral shaft treated with tunnel osteogenesis technique combined with locking plate between January 2017 and December 2020 were retrospectively analysed. There were 17 males and 6 females with an average age of 41.4 years (range, 22-72 years). There were 22 cases of closed fracture and 1 case of open fracture. The types of internal fixation at admission included intramedullary nail in 14 cases and steel plate in 9 cases. The number of nonunion operations received in the past was 0 to 1; the duration of nonunion was 6-60 months, with an average of 20.1 months. Among them, there were 17 cases of aseptic atrophic nonunion of the femoral shaft and 6 cases of dystrophic nonunion. Twenty-two cases were fixed with 90° double plates and 1 case with lateral single plate. The operation time, theoretical blood loss, hospitalization stay, nonunion healing, and postoperative complications were recorded. Harris hip function score, Lysholm knee function score, lower extremity function scale (LEFS), and short-form 36 health survey scale (SF-36) were used at last follow-up to evaluate hip and knee functions. Visual analogue scale (VAS) score was used to evaluate the relief of pain at 1 day after operation and at last follow-up. Results: The average operation time was 190.4 minutes, the average theoretical blood loss was 1 458.4 mL, and the average hospitalization stay was 8.2 days. All the 23 patients were followed up 9-26 months, with an average of 18.2 months. The healing time of nonunion in 22 patients was 3-12 months, with an average of 5.6 months. There were 8 cases of limb pain, 8 cases of claudication, 6 cases of limitation of knee joint movement, and 2 cases of limitation of hip joint movement. At last follow-up, the imaging of 1 patient showed that the nonunion did not heal, accompanied by pain of the affected limb, lameness, and limitation of knee joint movement. At 1 day after operation, the VAS score of 23 patients was 6.5±1.8, the pain degree was good in 7 cases, moderate in 12 cases, and poor in 4 cases; at last follow-up, the VAS score was 0.9±1.3, the pain degree was excellent in 21 cases and good in 2 cases, which were significantly better than those at 1 day after operation ( t=12.234, P<0.001; Z=-5.802, P<0.001). At last follow-up, the average Harris hip function score of 22 patients with nonunion healing was 94.8, and the good rate was 100%; Lysholm knee function score averaged 94.0, and the excellent and good rate was 90.9%; LEFS score averaged 74.6; SF-36 physical functioning score averaged 85.0 and the mental health score averaged 83.6. Conclusion: Tunnel osteogenesis technique combined with locking plate in the treatment of aseptic non-hypertrophic nonunion of femoral shaft has a high healing rate and fewer complications, which can effectively relieve pain and improve lower limb function and quality of patients' life.