Deep learning models have become instrumental in medical imaging, with a wide array of applications such as automated segmentation and diagnosis. However, the majority of these models are trained on adult imaging datasets, leading to underperformance when applied to the pediatric population. This discrepancy arises from anatomical differences between these populations, posing a significant challenge in fields requiring high precision like spine segmentation in Computed Tomography (CT) scans. Read More

Meeting name: SPR 2025 Annual Meeting , 2025

Authors: De Medeiros Bruno, Gatidis Sergios

Keywords: Computed Tomography, Segmentation, Image Database

Automated segmentation is an important step in automated processing of radiographs. However, manual generation of training data is tedious. To overcome this challenge, we leverage synthetic radiograph training data generated from CT volumes together with corresponding anatomy segmentation labels. We demonstrate the effectiveness of this approach on the use case of automated scoliosis Cobb angle quantification. Read More

Meeting name: SPR 2025 Annual Meeting , 2025

Authors: Vasanawala Sauram, Pauly John, Gatidis Sergios

Keywords: Segmentation, Scoliosis, Deep Learning

PET/MRI (Positron Emission Tomography–Magnetic Resonance Imaging) is a powerful hybrid imaging modality that has gained increasing clinical relevance in pediatric and neonatal imaging over the past decade. It combines the molecular imaging strengths of PET with the superior soft tissue contrast and functional capabilities of MRI, enabling both whole-body and targeted imaging in a single session. In pediatric imaging, where minimizing radiation exposure is critical and conditions often involve subtle functional or metabolic changes, PET/MRI has emerged as a powerful problem-solving modality. Additional advantage of PET/MRI in pediatric patients is reduction in the number of necessary single examinations and thus reducing the number of sedations and radiation exposure when compared with PET/CT. PET/MRI has been established as a clinical imaging modality with a wide range of applications across pediatric oncology, neurology, cardiology, infectious/ inflammatory disorders, and rheumatology. It can be utilized for presurgical planning for epilepsy, detection and staging of tumors. Beyond detection and characterization of tumor lesions, PET/MRI is highly effective in providing relevant secondary information about tumor-related or therapy-related complications. Additionally, PET/MRI is helpful diagnostic tool in unusual or challenging clinical situations such as discrepant findings from other diagnostic tests, nonspecific clinical findings, or inconclusive prior imaging findings, especially in children with fever of unknown origin. In such settings, PET/MRI can be used as a problem-solving tool to provide guidance regarding patient management when identifying disease foci or biopsy targets or when ruling out the presence of macroscopic malignancy or active inflammatory processes. This exhibit demonstrates the unique diagnostic value of hybrid PET/MRI in complex pediatric imaging and to highlight its role as a comprehensive, multiparametric tool in evaluating a wide spectrum of challenging pediatric conditions through clinical cases. The exhibit will be educational for radiologists, nuclear medicine physicians, and pediatric specialists on the strengths, clinical applications, and practical considerations of PET/MRI in children. Read More

Meeting name: SPR 2026 Annual Meeting , 2026

Authors: Antil Neha, Gatidis Sergios, Nadel Helen

Keywords: Paediatric Nuclear Medicine, Pet/MRI, Oncology