Automated Diffusion Analysis for Noninvasive Prediction of Isocitrate Dehydrogenase Genotype in WHO Grade 2-3 Gliomas

Glioma molecular characterization is essential for risk stratification and treatment planning. Noninvasive imaging biomarkers such as ADC values have shown potential for predicting glioma genotypes. However, manual segmentation of gliomas is time-consuming and operator-dependent. To address this limitation, we aimed to establish a single-sequence-derived automatic ADC extraction pipeline by using T2-weighted imaging to support glioma isocitrate dehydrogenase (IDH) genotyping.

Glioma volumes from a hospital data set (University College London Hospitals [UCLH]; n = 247) were manually segmented on T2-weighted MRI scans by using ITK-Snap Toolbox and coregistered to ADC map sequences by using the FMRIB Linear Image Registration Tool in FSL, followed by ADC histogram extraction (Python). Separately, a nnUNet deep learning algorithm was trained to segment glioma volumes by using T2-weighted sequences only from Brain Tumor Segmentation Challenge (BraTS) 2021 data (n = 500, 80% training, 5% validation, and 15% test split). nnUNet was then applied to the UCLH data for segmentation and ADC readouts. Univariable logistic regression was used to test the performance manual and nnUNet derived ADC metrics for IDH status prediction. Statistical equivalence was tested (paired 2-sided t test).

nnUNet segmentation achieved a median Dice of 0.85 on BraTS data, and 0.83 on UCLH data. For the best performing metric (normalized ADC_mean ) the area under the receiver operating characteristic curve (AUC) for differentiating IDH-mutant from IDH-wild-type gliomas was 0.82 (95% CI, 0.78-0.88), compared with the manual segmentation AUC 0.84 (95% CI, 0.77-0.89). For all ADC metrics, manually and nnUNet-extracted ADC were statistically equivalent (P < .01). nnUNet identified 1 area of glioma infiltration missed by human observers. In 0.8% gliomas, nnUNet missed glioma components. In 6% of cases, oversegmentation of brain remote from the tumor occurred (eg, temporal poles).

The T2-weighted trained nnUNet algorithm achieved ADC readouts for IDH genotyping with a performance statistically equivalent to human observers. This approach could support rapid ADC-based identification of glioblastoma at an early disease stage, even with limited input data. Artificial intelligence level of evidence: 5A.