Previous Issue
Volume 11, January
 
 

Tomography, Volume 11, Issue 2 (February 2025) – 4 articles

  • Issues are regarded as officially published after their release is announced to the table of contents alert mailing list.
  • You may sign up for e-mail alerts to receive table of contents of newly released issues.
  • PDF is the official format for papers published in both, html and pdf forms. To view the papers in pdf format, click on the "PDF Full-text" link, and use the free Adobe Reader to open them.
Order results
Result details
Section
Select all
Export citation of selected articles as:
19 pages, 945 KiB  
Article
Graph Neural Network Learning on the Pediatric Structural Connectome
by Anand Srinivasan, Rajikha Raja, John O. Glass, Melissa M. Hudson, Noah D. Sabin, Kevin R. Krull and Wilburn E. Reddick
Tomography 2025, 11(2), 14; https://doi.org/10.3390/tomography11020014 - 29 Jan 2025
Viewed by 378
Abstract
Purpose: Sex classification is a major benchmark of previous work in learning on the structural connectome, a naturally occurring brain graph that has proven useful for studying cognitive function and impairment. While graph neural networks (GNNs), specifically graph convolutional networks (GCNs), have gained [...] Read more.
Purpose: Sex classification is a major benchmark of previous work in learning on the structural connectome, a naturally occurring brain graph that has proven useful for studying cognitive function and impairment. While graph neural networks (GNNs), specifically graph convolutional networks (GCNs), have gained popularity lately for their effectiveness in learning on graph data, achieving strong performance in adult sex classification tasks, their application to pediatric populations remains unexplored. We seek to characterize the capacity for GNN models to learn connectomic patterns on pediatric data through an exploration of training techniques and architectural design choices. Methods: Two datasets comprising an adult BRIGHT dataset (N = 147 Hodgkin’s lymphoma survivors and N = 162 age similar controls) and a pediatric Human Connectome Project in Development (HCP-D) dataset (N = 135 healthy subjects) were utilized. Two GNN models (GCN simple and GCN residual), a deep neural network (multi-layer perceptron), and two standard machine learning models (random forest and support vector machine) were trained. Architecture exploration experiments were conducted to evaluate the impact of network depth, pooling techniques, and skip connections on the ability of GNN models to capture connectomic patterns. Models were assessed across a range of metrics including accuracy, AUC score, and adversarial robustness. Results: GNNs outperformed other models across both populations. Notably, adult GNN models achieved 85.1% accuracy in sex classification on unseen adult participants, consistent with prior studies. The extension of the adult models to the pediatric dataset and training on the smaller pediatric dataset were sub-optimal in their performance. Using adult data to augment pediatric models, the best GNN achieved comparable accuracy across unseen pediatric (83.0%) and adult (81.3%) participants. Adversarial sensitivity experiments showed that the simple GCN remained the most robust to perturbations, followed by the multi-layer perceptron and the residual GCN. Conclusions: These findings underscore the potential of GNNs in advancing our understanding of sex-specific neurological development and disorders and highlight the importance of data augmentation in overcoming challenges associated with small pediatric datasets. Further, they highlight relevant tradeoffs in the design landscape of connectomic GNNs. For example, while the simpler GNN model tested exhibits marginally worse accuracy and AUC scores in comparison to the more complex residual GNN, it demonstrates a higher degree of adversarial robustness. Full article
(This article belongs to the Section Artificial Intelligence in Medical Imaging)
12 pages, 1727 KiB  
Article
Impact of Deep Learning 3D CT Super-Resolution on AI-Based Pulmonary Nodule Characterization
by Dongok Kim, Chulkyun Ahn and Jong Hyo Kim
Tomography 2025, 11(2), 13; https://doi.org/10.3390/tomography11020013 - 27 Jan 2025
Viewed by 511
Abstract
Background/Objectives: Correct pulmonary nodule volumetry and categorization is paramount for accurate diagnosis in lung cancer screening programs. CT scanners with slice thicknesses of multiple millimetres are still common worldwide, and slice thickness has an adverse effect on the accuracy of the pulmonary nodule [...] Read more.
Background/Objectives: Correct pulmonary nodule volumetry and categorization is paramount for accurate diagnosis in lung cancer screening programs. CT scanners with slice thicknesses of multiple millimetres are still common worldwide, and slice thickness has an adverse effect on the accuracy of the pulmonary nodule volumetry. Methods: We propose a deep learning based super-resolution technique to generate thin-slice CT images from thick-slice CT images. Analysis of the lung nodule volumetry and categorization accuracy was performed using commercially available AI-based lung cancer screening software. Results: The accuracy of pulmonary nodule categorization increased from 72.7 percent to 94.5 percent when thick-slice CT images were converted to generated-thin-slice CT images. Conclusions: Applying the super-resolution-based slice generation on thick-slice CT images prior to automatic nodule evaluation significantly increases the accuracy of pulmonary nodule volumetry and corresponding pulmonary nodule category. Full article
(This article belongs to the Section Artificial Intelligence in Medical Imaging)
12 pages, 7664 KiB  
Article
Ultrashort Echo Time Magnetic Resonance Morphology of Discovertebral Junction in Chronic Low Back Pain Subjects
by Palanan Siriwananrangsun, Tim Finkenstaedt, Karen C. Chen and Won C. Bae
Tomography 2025, 11(2), 12; https://doi.org/10.3390/tomography11020012 - 23 Jan 2025
Viewed by 854
Abstract
Background: Chronic low back pain (LBP) has been associated with intervertebral disc (IVD) degeneration, but its association with abnormal morphology at the discovertebral junction (DVJ) is unclear. The goal of this study was to evaluate the DVJ morphology in asymptomatic (Asx) and symptomatic [...] Read more.
Background: Chronic low back pain (LBP) has been associated with intervertebral disc (IVD) degeneration, but its association with abnormal morphology at the discovertebral junction (DVJ) is unclear. The goal of this study was to evaluate the DVJ morphology in asymptomatic (Asx) and symptomatic (Sx) subjects for LBP using ultrashort echo time (UTE) MRI. Methods: We recruited 42 subjects (12 Asx and 32 Sx). Lumbar IVD degeneration was assessed using Pfirrmann grading (1 to 5), while the abnormality of DVJ (0 = normal; 1 = focal; 2 = broad abnormality) was assessed using UTE MRI. The effects of LBP and level on the mean IVD and DVJ grades, the correlation between IVD and DVJ grade, and the effect of LBP and age on the number of abnormal DVJs within a subject were determined. Results: IVD grade was higher in Sx subjects (p = 0.013), varying with disc level (p = 0.033), adjusted for age (p < 0.01). Similarly, DVJ grade was also significantly higher in Sx subjects (p = 0.001), but it did not vary with DVJ level (p = 0.7), adjusted for age (p = 0.5). There was a weak positive (rho = 0.344; p < 0.001) correlation between DVJ and IVD grade. The total number of abnormal DVJs within a subject was higher in Sx subjects (p < 0.001), but not with respect to age (p = 0.6) due to a large spread throughout the age range. Conclusions: These results demonstrate the feasibility of using in vivo UTE MRI of the lumbar spine to evaluate the DVJ and the correlation of DVJ with LBP. This study highlights the need for a better understanding of DVJ pathology and the inclusion of DVJ assessment in routine lumbar MRI. Full article
Show Figures

Figure 1

15 pages, 757 KiB  
Article
Correlation of Coronary Calcium Measured on Conventional Computed Tomography with Coronary Angiography Findings in Lung Transplant Patients
by Sergio Tapia Concha, Concepción Fariñas-Álvarez, Pedro Muñoz Cacho, José Manuel Cifrian Martínez, Javier Zueco Gil and José Antonio Parra Blanco
Tomography 2025, 11(2), 11; https://doi.org/10.3390/tomography11020011 - 22 Jan 2025
Viewed by 452
Abstract
Introduction and objective: The pre-transplant protocol for lung transplant candidates includes a chest CT scan to assess disease progression and often coronary angiography (CA) to rule out coronary artery disease (CAD). Coronary artery calcium is commonly observed in these pre-transplant CT scans. This [...] Read more.
Introduction and objective: The pre-transplant protocol for lung transplant candidates includes a chest CT scan to assess disease progression and often coronary angiography (CA) to rule out coronary artery disease (CAD). Coronary artery calcium is commonly observed in these pre-transplant CT scans. This study aims to evaluate the relationship between coronary calcium detected on CT and findings from CA to determine whether calcium presence could serve as an additional criterion for selecting patients for CA. Material and Methods: We included 252 consecutive lung transplant patients who had both a CT scan and CA within 365 days of each other. Coronary calcium quantification was performed using artery-based, segment artery-based, and visual assessment methods. CA findings were classified by stenosis severity: ≤20%, 21–70%, and >70%. Results: This study showed very high concordance (kappa = 0.896; 95% CI: 0.843–0.948) between the three methods, especially in distinguishing patients without and with coronary calcium (kappa = 1.000; 95% CI: 0.929–1.071). ROC analysis identified the absence of coronary calcium as the best cutoff to differentiate patients with ≤20% stenosis from those with >21%, with a sensitivity of 73.5%, specificity of 55.7%, PPV of 28.5%, and NPV of 90%. Only 11 patients (8.7%) without coronary calcium had stenosis of 21–70%, and only 2 (1.6%) had stenosis > 70%. Conclusions: The visual assessment method yielded results similar to the other two quantification methods. The absence of coronary calcium in pre-transplant CT may be a useful criterion for selecting patients for CA. Full article
(This article belongs to the Section Cardiovascular Imaging)
Show Figures

Figure 1

Previous Issue
Back to TopTop