Literature DB >> 34653254

Improving the noninvasive classification of glioma genetic subtype with deep learning and diffusion-weighted imaging.

Julia Cluceru1, Yannet Interian2, Joanna J Phillips3,4, Annette M Molinaro3, Tracy L Luks1, Paula Alcaide-Leon1,5, Marram P Olson1, Devika Nair1, Marisa LaFontaine1, Anny Shai3, Pranathi Chunduru3, Valentina Pedoia1, Javier E Villanueva-Meyer1, Susan M Chang3, Janine M Lupo1.   

Abstract

BACKGROUND: Diagnostic classification of diffuse gliomas now requires an assessment of molecular features, often including IDH-mutation and 1p19q-codeletion status. Because genetic testing requires an invasive process, an alternative noninvasive approach is attractive, particularly if resection is not recommended. The goal of this study was to evaluate the effects of training strategy and incorporation of biologically relevant images on predicting genetic subtypes with deep learning.
METHODS: Our dataset consisted of 384 patients with newly diagnosed gliomas who underwent preoperative MRI with standard anatomical and diffusion-weighted imaging, and 147 patients from an external cohort with anatomical imaging. Using tissue samples acquired during surgery, each glioma was classified into IDH-wildtype (IDHwt), IDH-mutant/1p19q-noncodeleted (IDHmut-intact), and IDH-mutant/1p19q-codeleted (IDHmut-codel) subgroups. After optimizing training parameters, top performing convolutional neural network (CNN) classifiers were trained, validated, and tested using combinations of anatomical and diffusion MRI with either a 3-class or tiered structure. Generalization to an external cohort was assessed using anatomical imaging models.
RESULTS: The best model used a 3-class CNN containing diffusion-weighted imaging as an input, achieving 85.7% (95% CI: [77.1, 100]) overall test accuracy and correctly classifying 95.2%, 88.9%, 60.0% of the IDHwt, IDHmut-intact, and IDHmut-codel tumors. In general, 3-class models outperformed tiered approaches by 13.5%-17.5%, and models that included diffusion-weighted imaging were 5%-8.8% more accurate than those that used only anatomical imaging.
CONCLUSION: Training a classifier to predict both IDH-mutation and 1p19q-codeletion status outperformed a tiered structure that first predicted IDH-mutation, then 1p19q-codeletion. Including apparent diffusion coefficient (ADC), a surrogate marker of cellularity, more accurately captured differences between subgroups.
© The Author(s) 2021. Published by Oxford University Press on behalf of the Society for Neuro-Oncology. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com.

Entities:  

Keywords:  ADC; convolutional neural network; deep learning; diffusion-weighted imaging; glioma subtype

Mesh:

Substances:

Year:  2022        PMID: 34653254      PMCID: PMC8972294          DOI: 10.1093/neuonc/noab238

Source DB:  PubMed          Journal:  Neuro Oncol        ISSN: 1522-8517            Impact factor:   13.029


  38 in total

1.  T2-FLAIR Mismatch, an Imaging Biomarker for IDH and 1p/19q Status in Lower-grade Gliomas: A TCGA/TCIA Project.

Authors:  Sohil H Patel; Laila M Poisson; Daniel J Brat; Yueren Zhou; Lee Cooper; Matija Snuderl; Cheddhi Thomas; Ana M Franceschi; Brent Griffith; Adam E Flanders; John G Golfinos; Andrew S Chi; Rajan Jain
Journal:  Clin Cancer Res       Date:  2017-07-27       Impact factor: 12.531

2.  Prediction of IDH1-Mutation and 1p/19q-Codeletion Status Using Preoperative MR Imaging Phenotypes in Lower Grade Gliomas.

Authors:  Y W Park; K Han; S S Ahn; S Bae; Y S Choi; J H Chang; S H Kim; S-G Kang; S-K Lee
Journal:  AJNR Am J Neuroradiol       Date:  2017-11-09       Impact factor: 3.825

3.  Advancing The Cancer Genome Atlas glioma MRI collections with expert segmentation labels and radiomic features.

Authors:  Spyridon Bakas; Hamed Akbari; Aristeidis Sotiras; Michel Bilello; Martin Rozycki; Justin S Kirby; John B Freymann; Keyvan Farahani; Christos Davatzikos
Journal:  Sci Data       Date:  2017-09-05       Impact factor: 6.444

4.  Machine Learning-Based Radiomics for Molecular Subtyping of Gliomas.

Authors:  Chia-Feng Lu; Fei-Ting Hsu; Kevin Li-Chun Hsieh; Yu-Chieh Jill Kao; Sho-Jen Cheng; Justin Bo-Kai Hsu; Ping-Huei Tsai; Ray-Jade Chen; Chao-Ching Huang; Yun Yen; Cheng-Yu Chen
Journal:  Clin Cancer Res       Date:  2018-05-22       Impact factor: 12.531

5.  A novel fully automated MRI-based deep-learning method for classification of 1p/19q co-deletion status in brain gliomas.

Authors:  Chandan Ganesh Bangalore Yogananda; Bhavya R Shah; Frank F Yu; Marco C Pinho; Sahil S Nalawade; Gowtham K Murugesan; Benjamin C Wagner; Bruce Mickey; Toral R Patel; Baowei Fei; Ananth J Madhuranthakam; Joseph A Maldjian
Journal:  Neurooncol Adv       Date:  2020-07-17

6.  IDH1 mutations as molecular signature and predictive factor of secondary glioblastomas.

Authors:  Sumihito Nobusawa; Takuya Watanabe; Paul Kleihues; Hiroko Ohgaki
Journal:  Clin Cancer Res       Date:  2009-09-15       Impact factor: 12.531

7.  Lower apparent diffusion coefficients indicate distinct prognosis in low-grade and high-grade glioma.

Authors:  Yong Cui; Li Ma; Xuzhu Chen; Zhe Zhang; Haihui Jiang; Song Lin
Journal:  J Neurooncol       Date:  2014-05-30       Impact factor: 4.130

8.  Predicting brain age with deep learning from raw imaging data results in a reliable and heritable biomarker.

Authors:  James H Cole; Rudra P K Poudel; Dimosthenis Tsagkrasoulis; Matthan W A Caan; Claire Steves; Tim D Spector; Giovanni Montana
Journal:  Neuroimage       Date:  2017-07-29       Impact factor: 6.556

9.  Deep Learning based Radiomics (DLR) and its usage in noninvasive IDH1 prediction for low grade glioma.

Authors:  Zeju Li; Yuanyuan Wang; Jinhua Yu; Yi Guo; Wei Cao
Journal:  Sci Rep       Date:  2017-07-14       Impact factor: 4.379

10.  Diffusion tensor image features predict IDH genotype in newly diagnosed WHO grade II/III gliomas.

Authors:  Paul Eichinger; Esther Alberts; Claire Delbridge; Stefano Trebeschi; Alexander Valentinitsch; Stefanie Bette; Thomas Huber; Jens Gempt; Bernhard Meyer; Juergen Schlegel; Claus Zimmer; Jan S Kirschke; Bjoern H Menze; Benedikt Wiestler
Journal:  Sci Rep       Date:  2017-10-17       Impact factor: 4.379

View more
  5 in total

1.  Deep learning approaches to non-invasively assess molecular features of gliomas.

Authors:  Rifaquat Rahman; Raymond Y Huang
Journal:  Neuro Oncol       Date:  2022-04-01       Impact factor: 12.300

2.  Improving the diagnosis of radiation necrosis after stereotactic radiosurgery to intracranial metastases with conventional MRI features: a case series.

Authors:  Arian Lasocki; Joseph Sia; Stephen L Stuckey
Journal:  Cancer Imaging       Date:  2022-07-06       Impact factor: 5.605

3.  Multiparametric MRI-Based Radiomics Model for Predicting H3 K27M Mutant Status in Diffuse Midline Glioma: A Comparative Study Across Different Sequences and Machine Learning Techniques.

Authors:  Wei Guo; Dejun She; Zhen Xing; Xiang Lin; Feng Wang; Yang Song; Dairong Cao
Journal:  Front Oncol       Date:  2022-03-03       Impact factor: 6.244

4.  Swin Transformer Improves the IDH Mutation Status Prediction of Gliomas Free of MRI-Based Tumor Segmentation.

Authors:  Jiangfen Wu; Qian Xu; Yiqing Shen; Weidao Chen; Kai Xu; Xian-Rong Qi
Journal:  J Clin Med       Date:  2022-08-08       Impact factor: 4.964

5.  Fully Automated MR Based Virtual Biopsy of Cerebral Gliomas.

Authors:  Johannes Haubold; René Hosch; Vicky Parmar; Martin Glas; Nika Guberina; Onofrio Antonio Catalano; Daniela Pierscianek; Karsten Wrede; Cornelius Deuschl; Michael Forsting; Felix Nensa; Nils Flaschel; Lale Umutlu
Journal:  Cancers (Basel)       Date:  2021-12-08       Impact factor: 6.639

  5 in total

北京卡尤迪生物科技股份有限公司 © 2022-2023.