Current Challenges in the Use of Computed Tomography and MR Imaging in Suspected Cervical Spine Trauma

Rethinking cervical spine clearance in obtunded trauma patients: An updated systematic review and meta-analysis

BACKGROUND

Cervical spine injuries (CSI) are often challenging to diagnose in obtunded adult patients with blunt trauma and the optimal imaging modality remains uncertain. This study systematically synthesized the last decade of evidence to determine the type of imaging required to clear the c-spine in obtunded patients with blunt trauma.

METHODS

A systematic review with meta-analysis was conducted and reported using PRISMA 2020 guidelines. The protocol was registered on June 22, 2022 (PROSPERO CRD42022341386). MEDLINE (Ovid), EMBASE, and Cochrane Library were searched for studies published between January 1, 2012, and October 17, 2023. Studies comparing CT alone to CT combined with MRI for c-spine clearance were included. Two independent reviewers screened articles for eligibility in duplicate. Meta-analysis was conducted using a random-effect model. Risk of bias and quality assessment were performed using the ROBINS-I and QUADAS-2. The certainty of evidence was assessed using the GRADE methodology.

RESULTS

744 obtunded trauma patients from six included studies were included. Among the 584 that had a negative CT scan, the pooled missed rate of clinically significant CSI using CT scans alone was 6 % (95 % CI: 0.02 to 0.17), and the pooled missed rate of CSI requiring treatment was 7 % (95 % CI: 0.02 to 0.18). High heterogeneity was observed among included studies (I² > 84 %). The overall risk of bias was moderate, and the quality of evidence was low due to the retrospective nature of the included studies and high heterogeneity.

CONCLUSIONS

Limited evidence published in the last decade found that CT scans alone may not be sufficient for detecting clinically significant CSI and injuries requiring treatment in obtunded adult patients with blunt trauma.

IMPLICATIONS OF KEY FINDINGS

Clinicians should be aware of the limitations of CT scans and consider using MRI when appropriate. Future research should focus on prospective studies with standardized outcome measures and uniform reporting.

The RSNA Cervical Spine Fracture CT Dataset

This dataset is composed of cervical spine CT images with annotations related to fractures; it is available at https://www.kaggle.com/competitions/rsna-2022-cervical-spine-fracture-detection/.

Classification of Cervical Spine Fracture and Dislocation Using Refined Pre-Trained Deep Model and Saliency Map

Cervical spine (CS) fractures or dislocations are medical emergencies that may lead to more serious consequences, such as significant functional disability, permanent paralysis, or even death. Therefore, diagnosing CS injuries should be conducted urgently without any delay. This paper proposes an accurate computer-aided-diagnosis system based on deep learning (AlexNet and GoogleNet) for classifying CS injuries as fractures or dislocations. The proposed system aims to support physicians in diagnosing CS injuries, especially in emergency services. We trained the model on a dataset containing 2009 X-ray images (530 CS dislocation, 772 CS fractures, and 707 normal images). The results show 99.56%, 99.33%, 99.67%, and 99.33% for accuracy, sensitivity, specificity, and precision, respectively. Finally, the saliency map has been used to measure the spatial support of a specific class inside an image. This work targets both research and clinical purposes. The designed software could be installed on the imaging devices where the CS images are captured. Then, the captured CS image is used as an input image where the designed code makes a clinical decision in emergencies.

CT Cervical Spine Fracture Detection Using a Convolutional Neural Network

BACKGROUND AND PURPOSE

Multidetector CT has emerged as the standard of care imaging technique to evaluate cervical spine trauma. Our aim was to evaluate the performance of a convolutional neural network in the detection of cervical spine fractures on CT.

MATERIALS AND METHODS

We evaluated C-spine, an FDA-approved convolutional neural network developed by Aidoc to detect cervical spine fractures on CT. A total of 665 examinations were included in our analysis. Ground truth was established by retrospective visualization of a fracture on CT by using all available CT, MR imaging, and convolutional neural network output information. The ĸ coefficients, sensitivity, specificity, and positive and negative predictive values were calculated with 95% CIs comparing diagnostic accuracy and agreement of the convolutional neural network and radiologist ratings, respectively, compared with ground truth.

RESULTS

Convolutional neural network accuracy in cervical spine fracture detection was 92% (95% CI, 90%-94%), with 76% (95% CI, 68%-83%) sensitivity and 97% (95% CI, 95%-98%) specificity. The radiologist accuracy was 95% (95% CI, 94%-97%), with 93% (95% CI, 88%-97%) sensitivity and 96% (95% CI, 94%-98%) specificity. Fractures missed by the convolutional neural network and by radiologists were similar by level and location and included fractured anterior osteophytes, transverse processes, and spinous processes, as well as lower cervical spine fractures that are often obscured by CT beam attenuation.

CONCLUSIONS

The convolutional neural network holds promise at both worklist prioritization and assisting radiologists in cervical spine fracture detection on CT. Understanding the strengths and weaknesses of the convolutional neural network is essential before its successful incorporation into clinical practice. Further refinements in sensitivity will improve convolutional neural network diagnostic utility.