Feasibility of generating sagittal radiographs from coronal views using GAN-based deep learning framework in adolescent idiopathic scoliosis DOI Creative Commons
Tito Bassani, Andrea Cina, Fabio Galbusera

et al.

European Radiology Experimental, Journal Year: 2025, Volume and Issue: 9(1)

Published: Jan. 29, 2025

Abstract Background Minimizing radiation exposure is crucial in monitoring adolescent idiopathic scoliosis (AIS). Generative adversarial networks (GANs) have emerged as valuable tools being able to generate high-quality synthetic images. This study explores the use of GANs sagittal radiographs from coronal views AIS patients. Methods A dataset 3,935 patients who underwent spine and pelvis radiographic examinations using EOS system, which simultaneously acquires images, was analyzed. The divided into training-set (85%, n = 3,356) test-set (15%, 579). GAN model trained images views, with real reference standard. To assess accuracy, 100 subjects were randomly selected for manual measurement lumbar lordosis (LL), sacral slope (SS), pelvic incidence (PI), vertical axis (SVA) by two radiologists both Results Sixty-nine considered assessable. intraclass correlation coefficient ranged 0.93–0.99 measurements 0.83 0.88 Correlations between parameters 0.52 0.17 0.18 0.74 (SVA). Measurement errors showed minimal severity. Mean ± standard deviation absolute 7 7° 9 8° 1.1 0.8 cm Conclusion While generates visually consistent their quality not sufficient clinical parameter assessment, except promising results SVA. Relevance statement AI can reduce However, while these appear ones, remains insufficient accurate assessment. Key Points be exploited views. Dataset used train test AI-model; spinal compared. Synthetic but generally Graphical

Language: Английский

A scoping review of automatic and semi-automatic MRI segmentation in human brain imaging DOI Creative Commons
Minh Chau,

Han X. Vu,

Tanmoy Debnath

et al.

Radiography, Journal Year: 2025, Volume and Issue: 31(2), P. 102878 - 102878

Published: Jan. 31, 2025

Language: Английский

Citations

1
AI for image quality and patient safety in CT and MRI DOI Creative Commons
Luca Melazzini, Chandra Bortolotto, L. Brizzi

et al.

European Radiology Experimental, Journal Year: 2025, Volume and Issue: 9(1)

Published: Feb. 23, 2025

Language: Английский

Citations

1
Rethinking MRI as a measurement device through modular and portable pipelines DOI Creative Commons
Agâh Karakuzu, Nadia Blostein, Alex Valcourt Caron

et al.

Magnetic Resonance Materials in Physics Biology and Medicine, Journal Year: 2025, Volume and Issue: unknown

Published: April 24, 2025

Abstract The premise of MRI as a reliable measurement device is limited by proprietary barriers and inconsistent implementations, which prevent the establishment uncertainties. As result, biomedical studies that rely on these methods are plagued systematic variance, undermining perceived promise quantitative imaging biomarkers (QIBs) hindering their clinical translation. This review explores added value open-source pipelines in minimizing variability sources would otherwise remain unknown. First, we introduce tiered benchmarking framework (from black-box to glass-box) exposes how opacity at different workflow stages propagates uncertainty. Second, provide concise glossary promote consistent terminology for strategies enhance reproducibility before acquisition or enable valid post-hoc pooling QIBs. Building this foundation, present two illustrative workflows decouple logic from orchestration computational processes an pipeline, rooted core principles modularity portability. Designed accessible entry points implementation, examples serve practical guides, helping users adapt frameworks specific needs facilitating collaboration. Through critical evaluation existing approaches, discuss standardized can help identify outstanding challenges translating glass-box into scanner environments. Ultimately, achieving goal will require coordinated efforts QIB developers, regulators, industry partners, clinicians alike.

Language: Английский

Citations

0
Feasibility of generating sagittal radiographs from coronal views using GAN-based deep learning framework in adolescent idiopathic scoliosis DOI Creative Commons
Tito Bassani, Andrea Cina, Fabio Galbusera

et al.

European Radiology Experimental, Journal Year: 2025, Volume and Issue: 9(1)

Published: Jan. 29, 2025

Abstract Background Minimizing radiation exposure is crucial in monitoring adolescent idiopathic scoliosis (AIS). Generative adversarial networks (GANs) have emerged as valuable tools being able to generate high-quality synthetic images. This study explores the use of GANs sagittal radiographs from coronal views AIS patients. Methods A dataset 3,935 patients who underwent spine and pelvis radiographic examinations using EOS system, which simultaneously acquires images, was analyzed. The divided into training-set (85%, n = 3,356) test-set (15%, 579). GAN model trained images views, with real reference standard. To assess accuracy, 100 subjects were randomly selected for manual measurement lumbar lordosis (LL), sacral slope (SS), pelvic incidence (PI), vertical axis (SVA) by two radiologists both Results Sixty-nine considered assessable. intraclass correlation coefficient ranged 0.93–0.99 measurements 0.83 0.88 Correlations between parameters 0.52 0.17 0.18 0.74 (SVA). Measurement errors showed minimal severity. Mean ± standard deviation absolute 7 7° 9 8° 1.1 0.8 cm Conclusion While generates visually consistent their quality not sufficient clinical parameter assessment, except promising results SVA. Relevance statement AI can reduce However, while these appear ones, remains insufficient accurate assessment. Key Points be exploited views. Dataset used train test AI-model; spinal compared. Synthetic but generally Graphical

Language: Английский

Citations

0