Back to all available

A transformer-based deep learning algorithm for diagnosing spinal infections on axial non-contrast computed tomography images

infectious-diseases
neurology
radiology-and-medical-imaging
computational-science
data-mining-and-machine-learning

Abstract

Objectives: To explore a deep learning approach for the diagnosis of spinal infections based on non-contrast CT images.

Methods: In this retrospective study, non-contrast CT data from 127 patients diagnosed with spinal infections between 2020 and 2023 were used. A deep learning model based on the Swin-Transformer was developed. Patients were split into training (70%) and test (30%) sets. Two radiologists evaluated the test images. Performance was measured by sensitivity, specificity, accuracy, and AUC.

Results: For the test set, the deep learning model got an accuracy of 90.5%, sensitivity of 94.5%, specificity of 89.9% and AUC of 0.975. The deep learning -based approach showed higher AUC and sensitivity than the radiologists (P<0.001 for AUC and P<0.01 for sensitivity). Furthermore, radiologists using the deep learning -based approach had significantly higher AUC, accuracy, sensitivity, and specificity compared to those without assistance (P<0.05). Deep learning models have shorter diagnosis times. Radiologists show shorter diagnosis times with the help of deep learning models(P<0.001). In addition, we found spinal epidural abscess and pathogen can influence the diagnostic outcome(P<0.001).

Conclusions: The transformer-based deep learning model effectively diagnosed spinal infections using non-contrast CT images, surpassing radiologists in performance. It enhanced radiologists' diagnostic efficiency and accuracy.

Ask to review this manuscript

Notes for potential reviewers

  • Volunteering is not a guarantee that you will be asked to review. There are many reasons: reviewers must be qualified, there should be no conflicts of interest, a minimum of two reviewers have already accepted an invitation, etc.
  • This is NOT OPEN peer review. The review is single-blind, and all recommendations are sent privately to the Academic Editor handling the manuscript. All reviews are published and reviewers can choose to sign their reviews.
  • What happens after volunteering? It may be a few days before you receive an invitation to review with further instructions. You will need to accept the invitation to then become an official referee for the manuscript. If you do not receive an invitation it is for one of many possible reasons as noted above.
  • PeerJ does not judge submissions based on subjective measures such as novelty, impact or degree of advance. Effectively, reviewers are asked to comment on whether or not the submission is scientifically and technically sound and therefore deserves to join the scientific literature. Our Peer Review criteria can be found on the "Editorial Criteria" page - reviewers are specifically asked to comment on 3 broad areas: "Basic Reporting", "Experimental Design" and "Validity of the Findings".
  • Reviewers are expected to comment in a timely, professional, and constructive manner.
  • Until the article is published, reviewers must regard all information relating to the submission as strictly confidential.
  • When submitting a review, reviewers are given the option to "sign" their review (i.e. to associate their name with their comments). Otherwise, all review comments remain anonymous.
  • All reviews of published articles are published. This includes manuscript files, peer review comments, author rebuttals and revised materials.
  • Each time a decision is made by the Academic Editor, each reviewer will receive a copy of the Decision Letter (which will include the comments of all reviewers).

If you have any questions about submitting your review, please email us at [email protected].