ABSTRACT
Vertebral fractures (VFs) cause serious problems, such as substantial functional loss and a high mortality rate, and a delayed diagnosis may further worsen the prognosis. Plain thoracolumbar radiography (PTLR) is an essential method for the evaluation of VFs. Therefore, minimizing the diagnostic errors of VFs on PTLR is crucial. Image identification based on a deep convolutional neural network (DCNN) has been recognized to be potentially effective as a diagnostic strategy; however, the accuracy for detecting VFs has not been fully investigated. A DCNN was trained with PTLR images of 300 patients (150 patients with and 150 without VFs). The accuracy, sensitivity, and specificity of diagnosis of the model were calculated and compared with those of orthopedic residents, orthopedic surgeons, and spine surgeons. The DCNN achieved accuracy, sensitivity, and specificity rates of 86.0% [95% confidence interval (CI) 82.0-90.0%], 84.7% (95% CI 78.8-90.5%), and 87.3% (95% CI 81.9-92.7%), respectively. Both the accuracy and sensitivity of the model were suggested to be noninferior to those of orthopedic surgeons. The DCNN can assist clinicians in the early identification of VFs and in managing patients, to prevent further invasive interventions and a decreased quality of life.
Subject(s)
Artificial Intelligence , Neural Networks, Computer , Osteoporotic Fractures/diagnosis , Quality of Life , Radiography/methods , Spinal Fractures/diagnosis , Absorptiometry, Photon , Aged , Case-Control Studies , Female , Follow-Up Studies , Humans , Male , Osteoporotic Fractures/diagnostic imaging , Prognosis , ROC Curve , Retrospective Studies , Spinal Fractures/diagnostic imagingABSTRACT
INTRODUCTION: Cervical ossification of the anterior longitudinal ligament (OALL) occasionally leads to dysphagia by the anterior osteophyte. A recent report explained that the dysphagia after an occipito-cervical fusion is caused by the narrowing of pharyngeal space due to the cranio-cervical malalignment. The purpose of this study was to evaluate the cranio-cervical alignment in patients with OALL complaining of the dysphagia. SUBJECTS AND METHODS: The subjects were 11 cases with complaining of dysphagia due to cervical OALL who underwent anterior cervical OALL resection and as control, age-matched 12 cases without dysphagia who have diffuse idiopathic skeletal hyperostosis in cervical spine. All subjects were male, and the mean age was 59.5 ± 9.1 years. The subjects were divided into two groups according to the symptoms of dysphagia (dysphagia, group A; control, group B). The O-C2 angle, C2-C7 angle, and the maximum thickness of OALL and the cranio-cervical alignment (pharyngeal inlet angle; PIA) and swallowing line (S-line) were measured before and after the operation on the lateral cervical radiogram at the sitting position. RESULTS: Group A showed significantly large maximum thickness of OALL, small cervical range of motion, small O-C2 angle, large C2-C7 angle, and small PIA. The S-line crossed the anterior apex of cervical osteophyte in group A. After OALL resection, dysphagia had improved, PIA had increased, and the S-line uncrossed the apex of cervical vertebrae in all cases. CONCLUSION: The prevalence of dysphagia in patients with cervical OALL was influenced by the thickness of osteophyte, cervical mobility, and cranio-cervical alignment.
