Nicholas Nuechterlein1, Beibin Li1, Abdullah Feroze2, Eric C Holland3, Linda Shapiro1, David Haynor4, James Fink4, Patrick J Cimino3,5. 1. Paul G. Allen School of Computer Science & Engineering, University of Washington, Seattle, Washington, USA. 2. Department of Neurological Surgery, University of Washington, Seattle, Washington, USA. 3. Division of Human Biology, Fred Hutchinson Cancer Research Center, Seattle, Washington, USA. 4. Department of Radiology, University of Washington, Seattle, Washington, USA. 5. Department of Pathology, Division of Neuropathology, University of Washington, Seattle, Washington, USA.
Abstract
BACKGROUND: Combined whole-exome sequencing (WES) and somatic copy number alteration (SCNA) information can separate isocitrate dehydrogenase (IDH)1/2-wildtype glioblastoma into two prognostic molecular subtypes, which cannot be distinguished by epigenetic or clinical features. The potential for radiographic features to discriminate between these molecular subtypes has yet to be established. METHODS: Radiologic features (n = 35 340) were extracted from 46 multisequence, pre-operative magnetic resonance imaging (MRI) scans of IDH1/2-wildtype glioblastoma patients from The Cancer Imaging Archive (TCIA), all of whom have corresponding WES/SCNA data. We developed a novel feature selection method that leverages the structure of extracted MRI features to mitigate the dimensionality challenge posed by the disparity between a large number of features and the limited patients in our cohort. Six traditional machine learning classifiers were trained to distinguish molecular subtypes using our feature selection method, which was compared to least absolute shrinkage and selection operator (LASSO) feature selection, recursive feature elimination, and variance thresholding. RESULTS: We were able to classify glioblastomas into two prognostic subgroups with a cross-validated area under the curve score of 0.80 (±0.03) using ridge logistic regression on the 15-dimensional principle component analysis (PCA) embedding of the features selected by our novel feature selection method. An interrogation of the selected features suggested that features describing contours in the T2 signal abnormality region on the T2-weighted fluid-attenuated inversion recovery (FLAIR) MRI sequence may best distinguish these two groups from one another. CONCLUSIONS: We successfully trained a machine learning model that allows for relevant targeted feature extraction from standard MRI to accurately predict molecularly-defined risk-stratifying IDH1/2-wildtype glioblastoma patient groups.
BACKGROUND: Combined whole-exome sequencing (WES) and somatic copy number alteration (SCNA) information can separate isocitrate dehydrogenase (IDH)1/2-wildtype glioblastoma into two prognostic molecular subtypes, which cannot be distinguished by epigenetic or clinical features. The potential for radiographic features to discriminate between these molecular subtypes has yet to be established. METHODS: Radiologic features (n = 35 340) were extracted from 46 multisequence, pre-operative magnetic resonance imaging (MRI) scans of IDH1/2-wildtype glioblastoma patients from The Cancer Imaging Archive (TCIA), all of whom have corresponding WES/SCNA data. We developed a novel feature selection method that leverages the structure of extracted MRI features to mitigate the dimensionality challenge posed by the disparity between a large number of features and the limited patients in our cohort. Six traditional machine learning classifiers were trained to distinguish molecular subtypes using our feature selection method, which was compared to least absolute shrinkage and selection operator (LASSO) feature selection, recursive feature elimination, and variance thresholding. RESULTS: We were able to classify glioblastomas into two prognostic subgroups with a cross-validated area under the curve score of 0.80 (±0.03) using ridge logistic regression on the 15-dimensional principle component analysis (PCA) embedding of the features selected by our novel feature selection method. An interrogation of the selected features suggested that features describing contours in the T2 signal abnormality region on the T2-weighted fluid-attenuated inversion recovery (FLAIR) MRI sequence may best distinguish these two groups from one another. CONCLUSIONS: We successfully trained a machine learning model that allows for relevant targeted feature extraction from standard MRI to accurately predict molecularly-defined risk-stratifying IDH1/2-wildtype glioblastoma patient groups.
Authors: Philipp Kickingereder; David Bonekamp; Martha Nowosielski; Annekathrin Kratz; Martin Sill; Sina Burth; Antje Wick; Oliver Eidel; Heinz-Peter Schlemmer; Alexander Radbruch; Jürgen Debus; Christel Herold-Mende; Andreas Unterberg; David Jones; Stefan Pfister; Wolfgang Wick; Andreas von Deimling; Martin Bendszus; David Capper Journal: Radiology Date: 2016-09-16 Impact factor: 11.105
Authors: Panagiotis Korfiatis; Timothy L Kline; Daniel H Lachance; Ian F Parney; Jan C Buckner; Bradley J Erickson Journal: J Digit Imaging Date: 2017-10 Impact factor: 4.056
Authors: Kenneth Clark; Bruce Vendt; Kirk Smith; John Freymann; Justin Kirby; Paul Koppel; Stephen Moore; Stanley Phillips; David Maffitt; Michael Pringle; Lawrence Tarbox; Fred Prior Journal: J Digit Imaging Date: 2013-12 Impact factor: 4.056
Authors: Siyuan Zheng; Jun Fu; Rahulsimham Vegesna; Yong Mao; Lindsey E Heathcock; Wandaliz Torres-Garcia; Ravesanker Ezhilarasan; Shuzhen Wang; Aaron McKenna; Lynda Chin; Cameron W Brennan; W K Alfred Yung; John N Weinstein; Kenneth D Aldape; Erik P Sulman; Ken Chen; Dimpy Koul; Roel G W Verhaak Journal: Genes Dev Date: 2013-06-24 Impact factor: 11.361
Authors: Patrick J Cimino; Lisa McFerrin; Hans-Georg Wirsching; Sonali Arora; Hamid Bolouri; Raul Rabadan; Michael Weller; Eric C Holland Journal: Neuro Oncol Date: 2018-09-03 Impact factor: 12.300
Authors: David A Gutman; Lee A D Cooper; Scott N Hwang; Chad A Holder; Jingjing Gao; Tarun D Aurora; William D Dunn; Lisa Scarpace; Tom Mikkelsen; Rajan Jain; Max Wintermark; Manal Jilwan; Prashant Raghavan; Erich Huang; Robert J Clifford; Pattanasak Mongkolwat; Vladimir Kleper; John Freymann; Justin Kirby; Pascal O Zinn; Carlos S Moreno; Carl Jaffe; Rivka Colen; Daniel L Rubin; Joel Saltz; Adam Flanders; Daniel J Brat Journal: Radiology Date: 2013-02-07 Impact factor: 11.105
Authors: Patrick J Cimino; Michael Zager; Lisa McFerrin; Hans-Georg Wirsching; Hamid Bolouri; Bettina Hentschel; Andreas von Deimling; David Jones; Guido Reifenberger; Michael Weller; Eric C Holland Journal: Acta Neuropathol Commun Date: 2017-05-22 Impact factor: 7.801
Authors: Nicholas Nuechterlein; Linda G Shapiro; Eric C Holland; Patrick J Cimino Journal: Acta Neuropathol Commun Date: 2021-12-04 Impact factor: 7.578
Authors: Felix Corr; Dustin Grimm; Benjamin Saß; Mirza Pojskić; Jörg W Bartsch; Barbara Carl; Christopher Nimsky; Miriam H A Bopp Journal: J Pers Med Date: 2022-03-04