Support vector machine multiparametric MRI identification of pseudoprogression from tumor recurrence in patients with
Xintao Hu1, Kelvin K Wong, Geoffrey S Young
1Department of Radiology, Center for Bioengineering and Informatics, The Methodist Hospital Research Institute, The Methodist Hospital, Houston, Texas, USA.
Purpose:
To automatically differentiate radiation necrosis from recurrent tumor at high spatial resolution using multiparametric MRI features.
Materials And Methods:
MRI data retrieved from 31 patients (15 recurrent tumor and 16 radiation necrosis) who underwent chemoradiation therapy after surgical resection included post-gadolinium T1, T2, fluid-attenuated inversion recovery, proton density, apparent diffusion coefficient (ADC), and perfusion-weighted imaging (PWI) -derived relative cerebral blood volume (rCBV), relative cerebral blood flow (rCBF), and mean transit time maps. After alignment to post contrast T1WI, an eight-dimensional feature vector was constructed. An one-class-support vector machine classifier was trained using a radiation necrosis training set. Classifier parameters were optimized based on the area under receiver operating characteristic (ROC) curve. The classifier was then tested on the full dataset.
Results:
The sensitivity and specificity of optimized classifier for pseudoprogression was 89.91% and 93.72%, respectively. The area under ROC curve was 0.9439. The distribution of voxels classified as radiation necrosis was supported by the clinical interpretation of follow-up scans for both nonprogressing and progressing test cases. The ADC map derived from diffusion-weighted imaging and rCBV, rCBF derived from PWI were found to make a greater contribution to the discrimination than the conventional images.
Conclusion:
Machine learning using multiparametric MRI features may be a promising approach to identify the distribution of radiation necrosis tissue in resected glioblastoma multiforme patients undergoing chemoradiation.
Insights
This study developed a machine learning model using multiparametric MRI to accurately distinguish radiation necrosis from recurrent tumors in brain cancer patients. The model achieved high accuracy, aiding in treatment monitoring after chemoradiation therapy.
Area of Science:
- Neuro-oncology
- Radiology
- Machine Learning
Background:
- Distinguishing radiation necrosis from recurrent glioblastoma multiforme (GBM) after treatment is clinically challenging.
- Accurate differentiation is crucial for appropriate patient management and treatment planning.
Purpose of the Study:
- To develop and validate a machine learning model for automated differentiation of radiation necrosis from recurrent tumor.
- To leverage multiparametric magnetic resonance imaging (MRI) features for high-resolution tissue characterization.
Main Methods:
- Retrospective analysis of MRI data from 31 patients treated for GBM.
- Utilized multiparametric MRI sequences including T1, T2, FLAIR, PD, ADC, and perfusion-weighted imaging (PWI).
- Trained a one-class-support vector machine classifier on an eight-dimensional feature vector derived from MRI data.
Main Results:
- The optimized classifier demonstrated high diagnostic performance with 89.91% sensitivity and 93.72% specificity.
- The area under the receiver operating characteristic (ROC) curve was 0.9439, indicating excellent discrimination.
- Apparent diffusion coefficient (ADC) maps and PWI-derived parameters (rCBV, rCBF) were key discriminators compared to conventional MRI sequences.
Conclusions:
- Machine learning applied to multiparametric MRI is a promising tool for identifying radiation necrosis in patients with resected GBM.
- This approach can aid in the accurate assessment of treatment response and disease progression.
