Fully automated hybrid approach to predict theIDHmutation status of gliomas via deep learning and radiomics

Abstract

Background Glioma prognosis depends on the isocitrate dehydrogenase (IDH) mutation status. We aimed to predict the IDH status of gliomas from preoperative MR images using a fully automated hybrid approach with convolutional neural networks (CNNs) and radiomics. Methods We reviewed 1,166 preoperative MR images of gliomas (grades II-IV) from Severance Hospital (n=856, Severance Set), Seoul National University Hospital (n=107, SNUH set), and The Cancer Imaging Archive (n=203, TCIA set). The Severance set was subdivided into the development (n=727) and internal test (n=129) sets. Based on T1 postcontrast, T2, and fluid-attenuated inversion-recovery images, a fully automated model was developed that comprised a CNN for tumor segmentation (Model 1) and CNN-based classifier for IDH status prediction (Model 2) that uses a hybrid approach based on 2-dimensional tumor images and radiomic features from 3-dimensional tumor shape and loci guided by Model 1. The trained model was tested on internal (a subset of the Severance set) and external (SNUH and TCIA) test sets. Results The CNN for tumor segmentation (Model 1) achieved a dice coefficient of 0.86–0.92 across datasets. Our hybrid model achieved accuracies of 93.8%, 87.9%, and 78.8%; with areas under the receiver operating characteristic curves of 0.96, 0.94, and 0.86; and areas under the precision-recall curves of 0.88, 0.82, and 0.81 in the internal test, SNUH, and TCIA sets, respectively. Conclusions Our fully automated hybrid model demonstrated the potential to be a highly reproducible and generalizable tool across different datasets for the noninvasive prediction of the IDH status of gliomas.

Probing individual-level structural atrophy in frontal glioma patients

“…Several studies(Bangalore Yogananda et al, 2020; K. S. Choi, Choi, & Jeong, 2019; Y. S. Choi et al, 2020; B.…”

“…Medical imaging technologies such as magnetic resonance imaging (MRI) and computed tomography (CT) are routine ways to obtain macroscopic information about glioma for the advantages of low cost, low or no damage and convenience(Jin et al, 2020; Lasocki, Anjari, Ӧrs Kokurcan, & Thust, 2020). Besides, the microscopic characteristics concealed in medical images have also been discovered, such as Radiomics, which has been widely used to predict histological and molecular biomarkers for glioma patients(Chen et al, 2018; Y. S. Choi et al, 2020; Smits & van den Bent, 2017).…”

Individual structural MRI abnormality indexes are associated with molecular, histological and cognitive indicators for frontal glioma patients

“…However, it is still di cult to timely know genetic examination results during the surgery, therefore the analysis of preoperative images has become the most possible way to predict these indicators. Several studies [18][19][20]7] using different modality images have been adopted for IDH prediction with high accuracy, but these models (e.g. deep learning) are hard to interpret intuitionally.…”

Application of radiomics and machine learning in head and neck cancers

“…Convolutional analysis is performed on the image through the CNN, and the data in the fully connected layer is used as the obtained depth feature. These features can continue to be used in the CNN or in other classifiers [50][51]. In the stage of radiomics feature extraction, a large amount of data will be obtained.…”