Related Experiment Video
Updated: Sep 7, 2025
Patient-Specific Polyvinyl Alcohol Phantom Fabrication with Ultrasound and X-Ray Contrast for Brain Tumor Surgery Planning
Published on: July 14, 2020
Deep-learning synthesized pseudo-CT for MR high-resolution pediatric cranial bone imaging (MR-HiPCB)
Parna Eshraghi Boroojeni1, Yasheng Chen2, Paul K Commean3
1Department of Biomedical Engineering, Washington University in St. Louis, St. Louis, Missouri.
Insights
This study developed a deep learning method to create pseudo-CT (pCT) images from MRI, reducing radiation exposure for pediatric patients. The synthesized pCT images accurately depicted cranial bone structures, comparable to traditional CT scans.
Area of Science:
- Medical Imaging
- Artificial Intelligence in Medicine
- Pediatric Radiology
Background:
- Computed Tomography (CT) is standard for detecting pediatric cranial abnormalities, but involves ionizing radiation.
- Magnetic Resonance Imaging (MRI) offers high-resolution bone imaging without radiation, but lacks CT's bone visualization capabilities.
- Developing radiation-free imaging methods is crucial for pediatric patient care.
Purpose of the Study:
- To develop a deep learning (DL) method for synthesizing pseudo-CT (pCT) images from high-resolution pediatric MRI.
- To enable accurate cranial bone imaging in children without using ionizing radiation from CT scans.
- To evaluate the performance of DL-synthesized pCT images against traditional CT and manual MRI processing.
Main Methods:
- 44 pediatric participants underwent 3D golden-angle stack-of-stars MRI.
- Two patch-based residual UNets (NetWH and NetBA) were trained to synthesize pCT from MR and CT patches.
- A combined approach (pCTCom) merged NetWH for brain areas and NetBA for non-brain areas, with a third UNet for brain masking.
- Manual processing using inverted MR images served as a comparison.
Main Results:
- The combined pCTCom method showed significantly lower mean absolute errors (MAEs) compared to NetWH and NetBA across the whole head.
- Within cranial bone, pCTCom had a significantly lower MAE than pCTNetWH and comparable results to pCTNetBA.
- pCTCom achieved a significantly higher Dice similarity coefficient for segmented bone than other methods, including inverted MR, with reduced age dependence.
- pCTCom demonstrated excellent visibility of sutures and fractures, comparable to CT.
Conclusions:
- The developed deep learning method successfully synthesizes pseudo-CT images from pediatric MRI.
- This radiation-free approach offers high-resolution cranial bone imaging comparable to CT.
- The findings support the clinical translation of MR-based cranial bone imaging for pediatric patients, reducing radiation risks.
Purpose:
CT is routinely used to detect cranial abnormalities in pediatric patients with head trauma or craniosynostosis. This study aimed to develop a deep learning method to synthesize pseudo-CT (pCT) images for MR high-resolution pediatric cranial bone imaging to eliminating ionizing radiation from CT.
Methods:
3D golden-angle stack-of-stars MRI were obtained from 44 pediatric participants. Two patch-based residual UNets were trained using paired MR and CT patches randomly selected from the whole head (NetWH) or in the vicinity of bone, fractures/sutures, or air (NetBA) to synthesize pCT. A third residual UNet was trained to generate a binary brain mask using only MRI. The pCT images from NetWH (pCTNetWH ) in the brain area and NetBA (pCTNetBA ) in the nonbrain area were combined to generate pCTCom . A manual processing method using inverted MR images was also employed for comparison.
Results:
pCTCom (68.01 ± 14.83 HU) had significantly smaller mean absolute errors (MAEs) than pCTNetWH (82.58 ± 16.98 HU, P < 0.0001) and pCTNetBA (91.32 ± 17.2 HU, P < 0.0001) in the whole head. Within cranial bone, the MAE of pCTCom (227.92 ± 46.88 HU) was significantly lower than pCTNetWH (287.85 ± 59.46 HU, P < 0.0001) but similar to pCTNetBA (230.20 ± 46.17 HU). Dice similarity coefficient of the segmented bone was significantly higher in pCTCom (0.90 ± 0.02) than in pCTNetWH (0.86 ± 0.04, P < 0.0001), pCTNetBA (0.88 ± 0.03, P < 0.0001), and inverted MR (0.71 ± 0.09, P < 0.0001). Dice similarity coefficient from pCTCom demonstrated significantly reduced age dependence than inverted MRI. Furthermore, pCTCom provided excellent suture and fracture visibility comparable to CT.
Conclusion:
MR high-resolution pediatric cranial bone imaging may facilitate the clinical translation of a radiation-free MR cranial bone imaging method for pediatric patients.
Related Concept Videos
01:10Computed Tomography
The technique was invented in the 1970s and is based on the principle that as X-rays pass through the body, they are absorbed or reflected at different levels. In the technique, a patient lies on a motorized platform while a computerized axial tomography (CAT) scanner rotates...
01:14Imaging Studies I: CT and MRI
Description of the Procedures
Computed Tomography (CT) scan:
Computed Tomography (CT) scans use X-ray technology to generate detailed images of bones, organs, and tissues. During the scan, the patient lies on a moving table...
01:27Imaging Studies III: Computed Tomography