Prediction of glioma‑subtypes: comparison of performance on a DL classifier using bounding box areas versus annotated tumors

Journal article, 2022

Background: For brain tumors, identifying the molecular subtypes from magnetic resonance imaging (MRI) isdesirable, but remains a challenging task. Recent machine learning and deep learning (DL) approaches may help theclassification/prediction of tumor subtypes through MRIs. However, most of these methods require annotated datawith ground truth (GT) tumor areas manually drawn by medical experts. The manual annotation is a time consumingprocess with high demand on medical personnel. As an alternative automatic segmentation is often used. However, itdoes not guarantee the quality and could lead to improper or failed segmented boundaries due to differences in MRIacquisition parameters across imaging centers, as segmentation is an ill‑defined problem. Analogous to visual objecttracking and classification, this paper shifts the paradigm by training a classifier using tumor bounding box areas inMR images. The aim of our study is to see whether it is possible to replace GT tumor areas by tumor bounding boxareas (e.g. ellipse shaped boxes) for classification without a significant drop in performance.Method: In patients with diffuse gliomas, training a deep learning classifier for subtype prediction by employ‑ing tumor regions of interest (ROIs) using ellipse bounding box versus manual annotated data. Experiments wereconducted on two datasets (US and TCGA) consisting of multi‑modality MRI scans where the US dataset containedpatients with diffuse low‑grade gliomas (dLGG) exclusively.Results: Prediction rates were obtained on 2 test datasets: 69.86% for 1p/19q codeletion status on US dataset and79.50% for IDH mutation/wild‑type on TCGA dataset. Comparisons with that of using annotated GT tumor data fortraining showed an average of 3.0% degradation (2.92% for 1p/19q codeletion status and 3.23% for IDH genotype).Conclusion: Using tumor ROIs, i.e., ellipse bounding box tumor areas to replace annotated GT tumor areas for train‑ing a deep learning scheme, cause only a modest decline in performance in terms of subtype prediction. With moredata that can be made available, this may be a reasonable trade‑off where decline in performance may be counter‑acted with more data.