The improved faster-RCNN for spinal fracture lesions detection

Purpose: Reading spinal CT (Computed Tomography) images is very important in the diagnosis of spondylosis, which is time-consuming and prones to make biases. In this paper, we propose a framework based on Faster-RCNN to improve detection performances of three spinal fracture lesions: cfracture (cervical fracture), tfracture (thoracic fracture) and lfracture (lumbar fracture). Methods: First, we use ResNet50 to replace VGG16 in backbone network in Faster-RCNN to increase depth of training network. Second, we utilize soft-NMS (Non-Maximum Suppression) instead of NMS to avoid missed detection of overlapped lesions. Third, we simplify RPN (Region Proposal Network) to accelerate training speed and reduce missed detection. Finally, we modify the classifier layer in Faster-RCNN and choose appropriate length-width ratio by changing anchor sizes in sliding window, then adopt multi-scale strategy in training to improve efficiency and accuracy. Results: The experimental results show that the proposed scheme has a good performance, mAP (mean average precision) is 90.6%, IOU (Intersection of Union) is 88.5 and detection time is 0.053 second per CT image, which means our proposed method can accurately detect spinal fracture lesions. Conclusion: Our proposed method can provide assistance and scientific references for both doctors and patients in clinically.

The Shrank YoloV3-tiny for spinal fracture lesions detection

Purpose: at present, more and more deep learning algorithms are used to detect and segment lesions from spinal CT (Computed Tomography) images. But these algorithms usually require computers with high performance and occupy large resources, so they are not suitable for the clinical embedded and mobile devices, which only have limited computational resources and also expect a relative good performance in detecting and segmenting lesions. Methods: in this paper, we present a model based on Yolov3-tiny to detect three spinal fracture lesions, cfracture (cervical fracture), tfracture (thoracic fracture), and lfracture (lumbar fracture) with a small size model. We construct this novel model by replacing the traditional convolutional layers in YoloV3-tiny with fire modules from SqueezeNet, so as to reduce the parameters and model size, meanwhile get accurate lesions detection. Then we remove the batch normalization layers in the fire modules after the comparative experiments, though the overall performance of fire module without batch normalization layers is slightly improved, we can reduce computation complexity and low occupations of computer resources for fast lesions detection. Results: the experiments show that the shrank model only has a size of 13 MB (almost a third of Yolov3-tiny), while the mAP (mean Average Precsion) is 91.3%, and IOU (intersection over union) is 90.7. The detection time is 0.015 second per CT image, and BFLOP/s (Billion Floating Point Operations per Second) value is less than Yolov3-tiny. Conclusion: the model we presented can be deployed in clinical embedded and mobile devices, meanwhile has a relative accurate and rapid real-time lesions detection.

Treating Lumbar Fracture Using the Mixed Reality Technique

The mixed reality (MR) technique has recently been widely used in the orthopedic surgery with satisfactory results reported. However, few studies have focused on the application of MR in the Lumbar fracture (LF). In this retrospective study, our aim is to analyze some findings by investigating the feasibility of MR applied to lumbar fracture treatment. Posterior vertebrectomy has been operated on 7 patients. The MR–based intraoperative three-dimensional image-guided navigation system (MITINS) was used to assist implantation of pedicle screws. The feasibility and safety of pedicle screw implantation were assessed by postsurgery radiography. The visual analog scale (VAS) and Oswestry Disability Index (ODI) were used to assess the pain level and recovery situation before and after surgery. 57 pedicle screws were safely and precisely placed into three-dimensional lumbar models by using MITINS. No screw was found outside the pedicle of the models, and it was not necessary for the X-ray to provide extra locative information during the operation with the use of MITINS. In summary, the application of MITINS is feasible, safe, and accurate while the lumbar fracture surgery is processing, providing satisfactory assistance for spine surgeons.

UNSTABLE LUMBAR FRACTURE-DISLOCATION TREATED BY LONG SEGMENT POSTERIOR PEDICLE SCREW INSTRUMENTATION

Background: Among all the thoracolumbar fractures, 50-60% affects the thoracolumbar transitional zone, and 51% AO Type C Fractures has a neurological deficit. We experienced treating a case of unstable lumbar fracture-dislocation treated with long segment pedicle screw instrumentation.Case: A 26-year-old man came to the ER after his back hit by a canopy while working 2 hours before admission. The motoric function was diminished from the L2-S1 level and hypoesthesia at the T12 level. Plain X-Ray showed Fracture-Dislocation Lumbar Vertebral 1-2 Denis Classification Flexion Rotation (AO Type C) ASIA A. The patient underwent reduction, decompression, and long-segment posterior pedicle screw instrumentation.Discussion: The surgery’s primary purpose is to restore alignment and stability to improve the patient’s quality of life by enabling daily activity in a wheelchair without significant pain. Short segment or long segment pedicle screw instrumentation remains a debate. In this case report, we apply long segment pedicle screw instrumentation for lumbar vertebral fracture-dislocation.Conclusion: Thoracolumbar fracture and dislocation fixation aim to restore alignment and stability, to reduce kyphotic deformity, and to decompress the spinal canal. The long segment pedicle screw instrumentation can resist the deforming force of thoracolumbar fractures and dislocations that will inevitably collapse into further kyphosis, resulting in a better outcome.