UCL Discovery
UCL home » Library Services » Electronic resources » UCL Discovery

Robust Deep Learning–based Segmentation of Glioblastoma on Routine Clinical MRI Scans Using Sparsified Training

Eijgelaar, RS; Visser, M; Müller, DMJ; Barkhof, F; Vrenken, H; van Herk, M; Bello, L; ... Witte, MG; + view all (2020) Robust Deep Learning–based Segmentation of Glioblastoma on Routine Clinical MRI Scans Using Sparsified Training. Radiology: Artificial Intelligence , 2 (5) , Article e190103. 10.1148/ryai.2020190103. Green open access

[thumbnail of ryai.2020190103.pdf]
Preview
Text
ryai.2020190103.pdf - Published Version

Download (1MB) | Preview

Abstract

Purpose: To improve the robustness of deep learning–based glioblastoma segmentation in a clinical setting with sparsified datasets. / Materials and Methods: In this retrospective study, preoperative T1-weighted, T2-weighted, T2-weighted fluid-attenuated inversion recovery, and postcontrast T1-weighted MRI from 117 patients (median age, 64 years; interquartile range [IQR], 55–73 years; 76 men) included within the Multimodal Brain Tumor Image Segmentation (BraTS) dataset plus a clinical dataset (2012–2013) with similar imaging modalities of 634 patients (median age, 59 years; IQR, 49–69 years; 382 men) with glioblastoma from six hospitals were used. Expert tumor delineations on the postcontrast images were available, but for various clinical datasets, one or more sequences were missing. The convolutional neural network, DeepMedic, was trained on combinations of complete and incomplete data with and without site-specific data. Sparsified training was introduced, which randomly simulated missing sequences during training. The effects of sparsified training and center-specific training were tested using Wilcoxon signed rank tests for paired measurements. / Results: A model trained exclusively on BraTS data reached a median Dice score of 0.81 for segmentation on BraTS test data but only 0.49 on the clinical data. Sparsified training improved performance (adjusted P < .05), even when excluding test data with missing sequences, to median Dice score of 0.67. Inclusion of site-specific data during sparsified training led to higher model performance Dice scores greater than 0.8, on par with a model based on all complete and incomplete data. For the model using BraTS and clinical training data, inclusion of site-specific data or sparsified training was of no consequence. / Conclusion: Accurate and automatic segmentation of glioblastoma on clinical scans is feasible using a model based on large, heterogeneous, and partially incomplete datasets. Sparsified training may boost the performance of a smaller model based on public and site-specific data.

Type: Article
Title: Robust Deep Learning–based Segmentation of Glioblastoma on Routine Clinical MRI Scans Using Sparsified Training
Open access status: An open access version is available from UCL Discovery
DOI: 10.1148/ryai.2020190103
Publisher version: https://doi.org/10.1148/ryai.2020190103
Language: English
Additional information: Published under a CC BY 4.0 license (https://creativecommons.org/licenses/by/4.0/)
UCL classification: UCL
UCL > Provost and Vice Provost Offices > School of Life and Medical Sciences
UCL > Provost and Vice Provost Offices > School of Life and Medical Sciences > Faculty of Brain Sciences
UCL > Provost and Vice Provost Offices > School of Life and Medical Sciences > Faculty of Brain Sciences > UCL Queen Square Institute of Neurology
UCL > Provost and Vice Provost Offices > School of Life and Medical Sciences > Faculty of Brain Sciences > UCL Queen Square Institute of Neurology > Brain Repair and Rehabilitation
URI: https://discovery.ucl.ac.uk/id/eprint/10112240
Downloads since deposit
79Downloads
Download activity - last month
Download activity - last 12 months
Downloads by country - last 12 months

Archive Staff Only

View Item View Item