Radiomics with Clinical Data and [<sup>18</sup>F]FDG-PET for Differentiating Between Infected and Non-Infected Intracavitary Vascular (Endo)Grafts: A Proof-of-Concept Study

Radiomics with Clinical Data and [<sup>18</sup>F]FDG-PET for Differentiating Between Infected and Non-Infected Intracavitary Vascular (Endo)Grafts: A Proof-of-Concept Study

<b>Objective:</b> We evaluated the feasibility of a machine-learning (ML) model based on clinical features and radiomics from [<sup>18</sup>F]FDG PET/CT images to differentiate between infected and non-infected intracavitary vascular grafts and endografts (iVGEI). <b>Me...

Full description

Saved in:
Bibliographic Details
Main Authors: Gijs D. van Praagh, Francine Vos, Stijn Legtenberg, Marjan Wouthuyzen-Bakker, Ilse J. E. Kouijzer, Erik H. J. G. Aarntzen, Jean-Paul P. M. de Vries, Riemer H. J. A. Slart, Lejla Alic, Bhanu Sinha, Ben R. Saleem
Format: Article
Language:English
Published: MDPI AG 2025-08-01
Series:Diagnostics
Subjects:
intracavitary vascular (endo)graft infections
radiomics
machine learning
[<sup>18</sup>F]FDG PET
MAGIC criteria
Online Access:https://www.mdpi.com/2075-4418/15/15/1944
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:<b>Objective:</b> We evaluated the feasibility of a machine-learning (ML) model based on clinical features and radiomics from [<sup>18</sup>F]FDG PET/CT images to differentiate between infected and non-infected intracavitary vascular grafts and endografts (iVGEI). <b>Methods:</b> Three ML models were developed: one based on pre-treatment criteria to diagnose a vascular graft infection (“<i>MAGIC-light</i> features”), another using radiomics features from diagnostic [<sup>18</sup>F]FDG-PET scans, and a third combining both datasets. The training set included 92 patients (72 iVGEI-positive, 20 iVGEI-negative), and the external test set included 20 iVGEI-positive and 12 iVGEI-negative patients. The abdominal aorta and iliac arteries in the PET/CT scans were automatically segmented using SEQUOIA and TotalSegmentator and manually adjusted, extracting 96 radiomics features. The best-performing models for the <i>MAGIC-light</i> features and <i>PET-radiomics</i> features were selected from 343 unique models. Most relevant features were combined to test three final models using ROC analysis, accuracy, sensitivity, and specificity. <b>Results:</b> The combined model achieved the highest AUC in the test set (mean ± SD: 0.91 ± 0.02) compared with the <i>MAGIC-light</i>-only model (0.85 ± 0.06) and the <i>PET-radiomics</i> model (0.73 ± 0.03). The combined model also achieved a higher accuracy (0.91 vs. 0.82) than the diagnosis based on all the MAGIC criteria and a comparable sensitivity and specificity (0.70 and 1.00 vs. 0.76 and 0.92, respectively) while providing diagnostic information at the initial presentation. The AUC for the combined model was significantly higher than the <i>PET-radiomics</i> model (<i>p</i> = 0.02 in the bootstrap test), while other comparisons were not statistically significant. <b>Conclusions:</b> This study demonstrated the potential of ML models in supporting diagnostic decision making for iVGEI. A combined model using pre-treatment clinical features and PET-radiomics features showed high diagnostic performance and specificity, potentially reducing overtreatment and enhancing patient outcomes.
ISSN:2075-4418

Similar Items