Related Experiment Video
Updated: Jun 19, 2025
Role of Diffusion MRI Tractography in Endoscopic Endonasal Skull Base Surgery
Published on: July 5, 2021
Machine Learning Analysis in Diffusion Kurtosis Imaging for Discriminating Pediatric Posterior Fossa Tumors: A
Ioan Paul Voicu1, Francesco Dotta1,2, Antonio Napolitano3
1Oncological Neuroradiology and Advanced Diagnostics Unit, Bambino Gesù Children's Hospital, IRCCS, 00165 Rome, Italy.
Insights
Diffusion kurtosis imaging (DKI) with whole-tumor-based (VOI) segmentations accurately distinguishes pediatric posterior fossa tumors. Machine learning analysis of DKI metrics achieved 92.8% accuracy, improving upon traditional region-of-interest (ROI) methods.
Area of Science:
- Radiology and Imaging
- Oncology
- Pediatric Neuro-oncology
Background:
- Pediatric posterior fossa tumors (medulloblastoma, ependymoma, pilocytic astrocytoma) require accurate differentiation for treatment and prognosis.
- Diffusion kurtosis imaging (DKI) has not been previously explored for discriminating these pediatric tumors.
- Whole-tumor-based (VOI) segmentations may offer improved repeatability for diffusion measurements compared to region-of-interest (ROI) approaches.
Purpose of the Study:
- To compare the repeatability of DKI-derived diffusion measurements between ROI and VOI segmentation methods.
- To assess the accuracy of DKI in discriminating between pediatric posterior fossa tumor subtypes.
- To evaluate the utility of machine learning algorithms for tumor classification using DKI metrics.
Main Methods:
- Retrospective analysis of 34 pediatric patients with posterior fossa tumors who underwent preoperative DKI.
- Independent segmentation of tumors by two neuroradiologists using ROI and VOI methods.
- Statistical analysis (MANOVA, Bland-Altman plots) and Random Forest machine learning classification on DKI metrics, with SMOTE for dataset balancing.
Main Results:
- VOI-based DKI measurements demonstrated lower inter-observer variability compared to ROI-based measurements.
- DKI metrics significantly discriminated between tumor subtypes (Pillai's trace: p < 0.001).
- Machine learning classification achieved a 0.928 accuracy in predicting tumor histology on a balanced dataset.
Conclusions:
- VOI-based DKI measurements offer improved repeatability for analyzing pediatric posterior fossa tumors.
- Machine learning algorithms utilizing DKI metrics are effective for accurate discrimination of these tumor types.
- DKI combined with ML represents a promising non-invasive tool for pediatric PF tumor classification.
Abstract:
Background and purpose: Differentiating pediatric posterior fossa (PF) tumors such as medulloblastoma (MB), ependymoma (EP), and pilocytic astrocytoma (PA) remains relevant, because of important treatment and prognostic implications. Diffusion kurtosis imaging (DKI) has not yet been investigated for discrimination of pediatric PF tumors. Estimating diffusion values from whole-tumor-based (VOI) segmentations may improve diffusion measurement repeatability compared to conventional region-of-interest (ROI) approaches. Our purpose was to compare repeatability between ROI and VOI DKI-derived diffusion measurements and assess DKI accuracy in discriminating among pediatric PF tumors. Materials and methods: We retrospectively analyzed 34 children (M, F, mean age 7.48 years) with PF tumors who underwent preoperative examination on a 3 Tesla magnet, including DKI. For each patient, two neuroradiologists independently segmented the whole solid tumor, the ROI of the area of maximum tumor diameter, and a small 5 mm ROI. The automated analysis pipeline included inter-observer variability, statistical, and machine learning (ML) analyses. We evaluated inter-observer variability with coefficient of variation (COV) and Bland-Altman plots. We estimated DKI metrics accuracy in discriminating among tumor histology with MANOVA analysis. In order to account for class imbalances, we applied SMOTE to balance the dataset. Finally, we performed a Random Forest (RF) machine learning classification analysis based on all DKI metrics from the SMOTE dataset by partitioning 70/30 the training and testing cohort. Results: Tumor histology included medulloblastoma (15), pilocytic astrocytoma (14), and ependymoma (5). VOI-based measurements presented lower variability than ROI-based measurements across all DKI metrics and were used for the analysis. DKI-derived metrics could accurately discriminate between tumor subtypes (Pillai's trace: p < 0.001). SMOTE generated 11 synthetic observations (10 EP and 1 PA), resulting in a balanced dataset with 45 instances (34 original and 11 synthetic). ML analysis yielded an accuracy of 0.928, which correctly predicted all but one lesion in the testing set. Conclusions: VOI-based measurements presented improved repeatability compared to ROI-based measurements across all diffusion metrics. An ML classification algorithm resulted accurate in discriminating PF tumors on a SMOTE-generated dataset. ML techniques based on DKI-derived metrics are useful for the discrimination of pediatric PF tumors.
