Related Experiment Video
Updated: Mar 28, 2026
10:48
PET and MRI Guided Irradiation of a Glioblastoma Rat Model Using a Micro-irradiator
Published on: December 28, 2017
10.2K
A Population-Based Gaussian Mixture Model Incorporating 18F-FDG PET and Diffusion-Weighted MRI Quantifies Tumor
Mathew R Divine1, Prateek Katiyar2, Ursula Kohlhofer3
1Department of Preclinical Imaging and Radiopharmacy, Werner Siemens Imaging Center, Eberhard Karls University Tuebingen, Tuebingen, Germany.
Summary
A novel Gaussian mixture modeling pipeline effectively segments tumor microenvironments using (18)F-FDG PET and DW-MRI data. This approach accurately models tumor compartments and predicts treatment response, advancing precision medicine.
Area of Science:
- Oncology
- Medical Imaging
- Computational Biology
Background:
- Tumor microenvironment characterization is crucial for understanding cancer progression and treatment response.
- Integrating multiparametric imaging data offers a more comprehensive view of tumor heterogeneity.
Purpose of the Study:
- To develop a novel Gaussian mixture modeling (GMM) pipeline for analyzing complementary (18)F-FDG PET and diffusion-weighted MRI (DW-MRI) data.
- To segment the tumor microenvironment into distinct tissue compartments and assess their longitudinal changes.
- To correlate imaging-derived tumor compartments with histological findings and predict treatment outcomes.
Main Methods:
- Coregistration of serial (18)F-FDG PET and apparent diffusion coefficient (ADC) maps from DW-MRI in NCI-H460 xenograft tumors.
- Application of a population-based GMM to segment the tumor microenvironment into three distinct regions.
- Correlation of segmented regions with histology and statistical analysis (ANCOVA) to assess predictive relationships between tumor volume, ADC, and (18)F-FDG uptake.
Main Results:
- Excellent agreement between coregistered PET/MR images and histology, validating the GMM segmentation.
- Strong correlations (r = 0.88 for necrotic, r = 0.87 for viable fractions) between GMM-derived tissue fractions and histology.
- Significant positive correlations found between total tumor volume and both ADC and (18)F-FDG uptake in whole, necrotic, and viable tissues (P < 0.05).
- ADC was identified as a significant positive predictor of (18)F-FDG uptake across all tumor compartments (P < 0.05).
Conclusions:
- The novel GMM pipeline successfully segments tumor microenvironments using longitudinal (18)F-FDG PET and ADC data.
- Multiparametric PET/MRI analysis holds potential for disease outcome assessment beyond RECIST criteria.
- This approach could significantly impact the field of precision medicine by providing more detailed tumor insights.
