AI Article Synopsis

  • - This study modified an existing model, X2CT-GAN, to create a 2Dto3D-GAN specifically for generating 3D images of the spine from 2D X-ray images, incorporating insights from radiologists.
  • - Using data from 1012 CT scans, the model's effectiveness was tested across different sizes of training datasets and various bone signal conditions, with several evaluation metrics showing strong performance.
  • - The findings indicate that enhancing bone signal conditions and increasing training data size significantly improve the model's performance, highlighting its potential for clinical applications in automatically producing 3D spine images.

Article Abstract

In this study, we modified the previously proposed X2CT-GAN to build a 2Dto3D-GAN of the spine. This study also incorporated the radiologist's perspective in the adjustment of input signals to prove the feasibility of the automatic production of three-dimensional (3D) structures of the spine from simulated bi-planar two-dimensional (2D) X-ray images. Data from 1012 computed tomography (CT) studies of 984 patients were retrospectively collected. We tested this model under different dataset sizes (333, 666, and 1012) with different bone signal conditions to observe the training performance. A 10-fold cross-validation and five metrics-Dice similarity coefficient (DSC) value, Jaccard similarity coefficient (JSC), overlap volume (OV), and structural similarity index (SSIM)-were applied for model evaluation. The optimal mean values for DSC, JSC, OV, SSIM_anteroposterior (AP), and SSIM_Lateral (Lat) were 0.8192, 0.6984, 0.8624, 0.9261, and 0.9242, respectively. There was a significant improvement in the training performance under empirically enhanced bone signal conditions and with increasing training dataset sizes. These results demonstrate the potential of the clinical implantation of GAN for automatic production of 3D spine images from 2D images. This prototype model can serve as a foundation in future studies applying transfer learning for the development of advanced medical diagnostic techniques.

Download full-text PDF

Source
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC9139385PMC
http://dx.doi.org/10.3390/diagnostics12051121DOI Listing

Publication Analysis

Top Keywords

gan automatic
8
simulated bi-planar
8
x-ray images
8
automatic production
8
dataset sizes
8
bone signal
8
signal conditions
8
training performance
8
similarity coefficient
8
generative adversarial
4

Similar Publications

Want AI Summaries of new PubMed Abstracts delivered to your In-box?

Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!