Related Experiment Video
Updated: Mar 7, 2026
Automated Joint Space Detection Improves Bone Segmentation Accuracy
Published on: November 28, 2025
The feasibility of semi-automatically generated red bone marrow segmentations based on MR-only for patients with
Anna Andreychenko1, Petra S Kroon1, Matteo Maspero1
1Department of Radiotherapy, Center for Image Sciences, University Medical Center Utrecht, The Netherlands.
Purpose:
For patients with cervical cancer the delivery of chemotherapy with radiotherapy improves survival compared with radiotherapy alone. However, high rates of acute hematologic toxicity occur when combining both therapies due to the damage of the red bone marrow (RBM). This study aimed to reduce the radiation damage to the RBM. A tool has been developed for semi-automatic delineation of the red bone marrow based on MR-only. This delineation can be included into the treatment planning process to reduce the volume of RBM irradiated in patients receiving pelvic radiation therapy.
Methods:
13 patients with cervical cancer were enrolled. All the patients underwent MR, CT and FDG-PET imaging. A tool for RBM determination from water and fat MR images was developed. Our MR-based RBM tool was optimized and validated with the FDG-PET scans of the patients.
Results:
Our tool identified RBM regions in the pelvic area. The mean total volume of these regions was 34% of the pelvic bone marrow. The corresponding SUV values based on the FDG-PET scans were above the reported threshold of active/red bone marrow.
Conclusion:
This study shows that delineations of the RBM for the radiotherapy with RBM sparing can be generated semi-automatically using MR scans only.
More Related Videos
08:39Longitudinal Micro-Computed Tomography Image Analysis for User-Defined Region of Interest in Critical-Sized Bone Defects
Published on: June 24, 2025
09:31Automated Quantification of Hematopoietic Cell – Stromal Cell Interactions in Histological Images of Undecalcified Bone
Published on: April 8, 2015