Automated quantification of brain PET in PET/CT using deep learning-based CT-to-MR translation: a feasibility study

• Published in: European journal of nuclear medicine and molecular imaging Vol. 52; no. 8; pp. 2959 - 2967
• Main Authors: Kim, Daesung, Choo, Kyobin, Lee, Sangwon, Kang, Seongjin, Yun, Mijin, Yang, Jaewon
• Format: Journal Article
• Language: English
• Published: Berlin/Heidelberg Springer Berlin Heidelberg 01.07.2025; Springer Nature B.V
• Subjects: Adult; Aged; Automation; Brain; Brain - diagnostic imaging; Cardiology; Computed tomography; Cortex (parietal); Deep Learning; Feasibility Studies; Female; Fluorine isotopes; Humans; Image acquisition; Image Processing, Computer-Assisted - methods; Imaging; Magnetic Resonance Imaging; Male; Medical imaging; Medicine; Medicine & Public Health; Middle Aged; Nuclear Medicine; Oncology; Original Article; Orthopedics; Positron emission; Positron emission tomography; Positron Emission Tomography Computed Tomography - methods; Quantitative analysis; Radiology; Rank tests; Regional analysis; Regions; Retrospective Studies; Segmentation; Deep learning; Amyloid; Quantification; Segmentation; PET/CT
• ISSN: 1619-7070; 1619-7089; 1619-7089
• DOI: 10.1007/s00259-025-07132-2

Purpose Quantitative analysis of PET images in brain PET/CT relies on MRI-derived regions of interest (ROIs). However, the pairs of PET/CT and MR images are not always available, and their alignment is challenging if their acquisition times differ considerably. To address these problems, this study proposes a deep learning framework for translating CT of PET/CT to synthetic MR images (MR SYN ) and performing automated quantitative regional analysis using MR SYN -derived segmentation. Methods In this retrospective study, 139 subjects who underwent brain [ 18 F]FBB PET/CT and T1-weighted MRI were included. A U-Net-like model was trained to translate CT images to MR SYN ; subsequently, a separate model was trained to segment MR SYN into 95 regions. Regional and composite standardised uptake value ratio (SUVr) was calculated in [ 18 F]FBB PET images using the acquired ROIs. For evaluation of MR SYN , quantitative measurements including structural similarity index measure (SSIM) were employed, while for MR SYN -based segmentation evaluation, Dice similarity coefficient (DSC) was calculated. Wilcoxon signed-rank test was performed for SUVrs computed using MR SYN and ground-truth MR (MR GT ). Results Compared to MR GT , the mean SSIM of MR SYN was 0.974 ± 0.005. The MR SYN -based segmentation achieved a mean DSC of 0.733 across 95 regions. No statistical significance ( P  > 0.05) was found for SUVr between the ROIs from MR SYN and those from MR GT , excluding the precuneus. Conclusion We demonstrated a deep learning framework for automated regional brain analysis in PET/CT with MR SYN . Our proposed framework can benefit patients who have difficulties in performing an MRI scan.