Enhancing 1p/19q Classification in Brain Gliomas Using IDH Status: A Deep Learning Study.
Authors
Affiliations (2)
Affiliations (2)
- From the Department of Radiology (J.E.B., A.S.K., B.C.W., N.C.D.T., J.M.H., D.D.R., N.S., B.F., M.C.P., C.G.B.Y., J.A.M.), Pathology (K.J.H.), Neurological Surgery (T.R.P.), UT Southwestern Medical Center, TX, USA; Department of Bioengineering (B.F.), UT Dallas, Richardson, TX, USA; Department of Radiology (M.D.L., R.J.), NYU Grossman School of Medicine, NY, USA and Department of Radiology (R.J.B.), University of Wisconsin-Madison, WI, USA. [email protected].
- From the Department of Radiology (J.E.B., A.S.K., B.C.W., N.C.D.T., J.M.H., D.D.R., N.S., B.F., M.C.P., C.G.B.Y., J.A.M.), Pathology (K.J.H.), Neurological Surgery (T.R.P.), UT Southwestern Medical Center, TX, USA; Department of Bioengineering (B.F.), UT Dallas, Richardson, TX, USA; Department of Radiology (M.D.L., R.J.), NYU Grossman School of Medicine, NY, USA and Department of Radiology (R.J.B.), University of Wisconsin-Madison, WI, USA.
Abstract
IDH mutation & 1p/19q codeletion are critical biomarkers for glioma diagnosis & therapy. 1p/19q codeletion occurs exclusively in IDH-mutated gliomas. In this study, we developed a 2-stage, non-invasive, MRI-based deep learning method that leverages IDH status to enhance 1p/19q predictions. Multi-contrast brain tumor MRI & genomic information were obtained from five publicly available (TCIA, UCSF, EGD, UPenn & LGG), and three in-house/collaborator institutions (UTSW, NYU, UWM). Subjects were screened for the availability of IDH & 1p/19q status as well as T1, T1CE, T2, FLAIR MR images. For training purposes, missing T1 and FLAIR contrasts for the LGG database were generated using an in-house multi-contrast simulator. Two separate <i>U-Nets</i> were developed for 1p/19q-classification: a multi-contrast network (<i>MC-Net)</i> and a T2w-only network (<i>T2-Net</i>). A separate <i>U-Net</i> was developed for IDH classification (<i>IDH-net</i>). A total of 2044 subjects were used in training and testing <i>IDH-N</i>et, and 1426 subjects were used in training and testing the 1p/19q models. The <i>IDH-Net</i> was trained using subjects from TCIA, UTSW, and UPenn. The 1p/19q networks were trained using subjects from TCIA, UTSW, and LGG. The trained networks were tested on true held-out cases from NYU, UWM, EGD, and UCSF. In the 2-stage approach, subjects were initially classified for IDH status using <i>IDH-Net.</i> Predicted IDH-wildtype cases default to 1p/19q non-codeleted. Then the IDH-mutated cases were further classified for 1p/19q status using the 1p/19q-networks. <i>IDH-Net</i> achieved a classification accuracy of 93.7%. 1p/19q <i>MC-Net</i> & <i>T2-Net</i> achieved classification accuracies of 86.5% & 86.0%, respectively. In the 2-stage approach, 1p/19q <i>MC-Net</i> and <i>T2-Net</i> achieved accuracies of 91.5% & 91.2% respectively, improving the classification accuracy by ∼5%. This study demonstrates the effectiveness of leveraging IDH status to enhance 1p/19q classification. A ∼5% increase in classification accuracy was achieved when using the 2-stage approach, using <i>IDH-Net</i> to gate 1p/19q predictions. The developed method offers a reliable, non-invasive approach to determine important biomarkers for glioma diagnosis.