Wyniki wyszukiwania - BazTech

Ograniczanie wyników

2 Polish Journal of Medical Physics and Engineering

2 2025

Znaleziono wyników: 2

Liczba wyników na stronie

Wyniki wyszukiwania

Sortuj według:

Ogranicz wyniki do:

Deep Learning-Based Synthetic-CT Generation from MRI for Enhanced Precision in MRI-Only Radiotherapy Dose Planning

Acquah Isaac Kwesi, Issahaku Shiraz, Tagoe Samuel Nii Adu, Sackey Theophilus Akumea

Polish Journal of Medical Physics and Engineering

2025

Vol. 31, Iss. 3

219--226

Introduction: Radiotherapy aims to precisely target tumors while sparing healthy tissue, traditionally relying on CT imaging for accurate dose planning. However, CT has limitations in soft tissue contrast and exposes patients to ionizing radiation. MRI offers superior soft tissue contrast without radiation but lacks electron density information, restricting its use in dose planning. This study addresses this gap by developing deep learning models to generate pseudo-CT images from MRI, enabling MRI-only workflows in radiotherapy. Methodology: Paired MRI and CT scans from 12 subjects were processed using normalization, alignment, and masking. Four deep learning architectures (U-Net, Pix2Pix, CycleGAN, and conditional GAN (cGAN)) were trained to generate synthetic CT images from MRI data. Model performance was evaluated using metrics including mean absolute error (MAE), mean squared error (MSE), peak signal-to-noise ratio (PSNR), structural similarity index (SSIM), and Pearson correlation coefficient (PCC). Results: Pix2Pix achieved the highest SSIM and PSNR, indicating strong structural preservation and reduced noise. It also had the lowest MAE and MSE, showing high accuracy in synthetic-CT generation. The cGAN model scored highest in PCC, highlighting its effective intensity alignment with real CT data. Statistical tests confirmed Pix2Pix’s superior performance, though CycleGAN and cGAN also showed notable results in alignment accuracy. Conclusion: Deep learning models, particularly Pix2Pix, can generate reliable pseudo-CT images from MRI, supporting MRI-only radiotherapy planning. This approach reduces radiation exposure and may streamline radiotherapy work-flows, offering a promising advance for patient-centered cancer care and MRI-only radiotherapy workflows.

Assessment of Radiation Doses for Paediatric Head, Chest, and Abdominopelvic Computed Tomography Examinations in a Teaching Hospital in Ghana

Amedzoame Bernard, Brobbey Isaac Frimpong, Djan Bismark, Acquah Isaac Kwesi, Tetteh Mercy Afadzi

Polish Journal of Medical Physics and Engineering

2025

Vol. 31, Iss. 3

256--262

Introduction: With the increasing number of pediatric computed tomography (CT) examinations, there is a need to optimise protocols for children by adopting examination-specific protocols customised to the patient’s age, size, imaging region, and clinical indication. This study aimed to assess the radiation doses in pediatric CT examinations and compare them to international standards. Material and methods: A cross-sectional retrospective study design was adopted to probe patient records at the radiology department of a teaching hospital in Ghana. Thus, scan parameters, volume computed tomography dose index (CTDIvol), dose length product (DLP), as well as demographic data, were recorded from 496 pediatric patients (age 0-15 years) undergoing head, chest, and abdominopelvic CT examinations. Local Diagnostic Reference Levels (LDRLs) were established using the 75th percentile of patient dose values for each protocol and age group. These local levels were then compared with DRLs from other studies. Results: Head CT was the most performed examination (35.0%) compared to chest (32.0%) and abdominopelvic (33.0%). The male group recorded the highest (59.1%) percentage of CT examinations compared to the female group. While LDRL values from this study were generally lower than data from other studies, the CTDIvol and DLP for head scans of patients between 11 and 15 years were found to be higher than the data from other studies. Conclusions: Our study has established LDRLs for standard pediatric CT examinations in the teaching hospital. The LDRLs were generally lower than those reported in other studies, except for head scans in patients aged 11 to 15 years. These findings suggest that there are opportunities for further optimisation of pediatric CT imaging protocols at this facility.