BACKGROUND: Isocitrate dehydrogenase (IDH) mutation status has emerged as an important prognostic marker in gliomas. Currently, reliable IDH mutation determination requires invasive surgical procedures. The purpose of this study was to develop a highly accurate, MRI-based, voxelwise deep-learning IDH classification network using T2-weighted (T2w) MR images and compare its performance to a multicontrast network. METHODS: Multiparametric brain MRI data and corresponding genomic information were obtained for 214 subjects (94 IDH-mutated, 120 IDH wild-type) from The Cancer Imaging Archive and The Cancer Genome Atlas. Two separate networks were developed, including a T2w image-only network (T2-net) and a multicontrast (T2w, fluid attenuated inversion recovery, and T1 postcontrast) network (TS-net) to perform IDH classification and simultaneous single label tumor segmentation. The networks were trained using 3D Dense-UNets. Three-fold cross-validation was performed to generalize the networks' performance. Receiver operating characteristic analysis was also performed. Dice scores were computed to determine tumor segmentation accuracy. RESULTS: T2-net demonstrated a mean cross-validation accuracy of 97.14% ± 0.04 in predicting IDH mutation status, with a sensitivity of 0.97 ± 0.03, specificity of 0.98 ± 0.01, and an area under the curve (AUC) of 0.98 ± 0.01. TS-net achieved a mean cross-validation accuracy of 97.12% ± 0.09, with a sensitivity of 0.98 ± 0.02, specificity of 0.97 ± 0.001, and an AUC of 0.99 ± 0.01. The mean whole tumor segmentation Dice scores were 0.85 ± 0.009 for T2-net and 0.89 ± 0.006 for TS-net. CONCLUSION: We demonstrate high IDH classification accuracy using only T2-weighted MR images. This represents an important milestone toward clinical translation.
BACKGROUND:Isocitrate dehydrogenase (IDH) mutation status has emerged as an important prognostic marker in gliomas. Currently, reliable IDH mutation determination requires invasive surgical procedures. The purpose of this study was to develop a highly accurate, MRI-based, voxelwise deep-learning IDH classification network using T2-weighted (T2w) MR images and compare its performance to a multicontrast network. METHODS: Multiparametric brain MRI data and corresponding genomic information were obtained for 214 subjects (94 IDH-mutated, 120 IDH wild-type) from The Cancer Imaging Archive and The Cancer Genome Atlas. Two separate networks were developed, including a T2w image-only network (T2-net) and a multicontrast (T2w, fluid attenuated inversion recovery, and T1 postcontrast) network (TS-net) to perform IDH classification and simultaneous single label tumor segmentation. The networks were trained using 3D Dense-UNets. Three-fold cross-validation was performed to generalize the networks' performance. Receiver operating characteristic analysis was also performed. Dice scores were computed to determine tumor segmentation accuracy. RESULTS: T2-net demonstrated a mean cross-validation accuracy of 97.14% ± 0.04 in predicting IDH mutation status, with a sensitivity of 0.97 ± 0.03, specificity of 0.98 ± 0.01, and an area under the curve (AUC) of 0.98 ± 0.01. TS-net achieved a mean cross-validation accuracy of 97.12% ± 0.09, with a sensitivity of 0.98 ± 0.02, specificity of 0.97 ± 0.001, and an AUC of 0.99 ± 0.01. The mean whole tumor segmentation Dice scores were 0.85 ± 0.009 for T2-net and 0.89 ± 0.006 for TS-net. CONCLUSION: We demonstrate high IDH classification accuracy using only T2-weighted MR images. This represents an important milestone toward clinical translation.
Authors: Nicholas J Tustison; Philip A Cook; Arno Klein; Gang Song; Sandhitsu R Das; Jeffrey T Duda; Benjamin M Kandel; Niels van Strien; James R Stone; James C Gee; Brian B Avants Journal: Neuroimage Date: 2014-05-29 Impact factor: 6.556
Authors: Jason Beiko; Dima Suki; Kenneth R Hess; Benjamin D Fox; Vincent Cheung; Matthew Cabral; Nicole Shonka; Mark R Gilbert; Raymond Sawaya; Sujit S Prabhu; Jeffrey Weinberg; Frederick F Lang; Kenneth D Aldape; Erik P Sulman; Ganesh Rao; Ian E McCutcheon; Daniel P Cahill Journal: Neuro Oncol Date: 2013-12-04 Impact factor: 12.300
Authors: Michele Ceccarelli; Floris P Barthel; Tathiane M Malta; Thais S Sabedot; Sofie R Salama; Bradley A Murray; Olena Morozova; Yulia Newton; Amie Radenbaugh; Stefano M Pagnotta; Samreen Anjum; Jiguang Wang; Ganiraju Manyam; Pietro Zoppoli; Shiyun Ling; Arjun A Rao; Mia Grifford; Andrew D Cherniack; Hailei Zhang; Laila Poisson; Carlos Gilberto Carlotti; Daniela Pretti da Cunha Tirapelli; Arvind Rao; Tom Mikkelsen; Ching C Lau; W K Alfred Yung; Raul Rabadan; Jason Huse; Daniel J Brat; Norman L Lehman; Jill S Barnholtz-Sloan; Siyuan Zheng; Kenneth Hess; Ganesh Rao; Matthew Meyerson; Rameen Beroukhim; Lee Cooper; Rehan Akbani; Margaret Wrensch; David Haussler; Kenneth D Aldape; Peter W Laird; David H Gutmann; Houtan Noushmehr; Antonio Iavarone; Roel G W Verhaak Journal: Cell Date: 2016-01-28 Impact factor: 41.582
Authors: Changho Choi; Sandeep K Ganji; Ralph J DeBerardinis; Kimmo J Hatanpaa; Dinesh Rakheja; Zoltan Kovacs; Xiao-Li Yang; Tomoyuki Mashimo; Jack M Raisanen; Isaac Marin-Valencia; Juan M Pascual; Christopher J Madden; Bruce E Mickey; Craig R Malloy; Robert M Bachoo; Elizabeth A Maher Journal: Nat Med Date: 2012-01-26 Impact factor: 53.440
Authors: Panagiotis Korfiatis; Timothy L Kline; Lucie Coufalova; Daniel H Lachance; Ian F Parney; Rickey E Carter; Jan C Buckner; Bradley J Erickson Journal: Med Phys Date: 2016-06 Impact factor: 4.071
Authors: Andreas Kleppe; Ole-Johan Skrede; Sepp De Raedt; Knut Liestøl; David J Kerr; Håvard E Danielsen Journal: Nat Rev Cancer Date: 2021-01-29 Impact factor: 60.716
Authors: C G B Yogananda; B R Shah; S S Nalawade; G K Murugesan; F F Yu; M C Pinho; B C Wagner; B Mickey; T R Patel; B Fei; A J Madhuranthakam; J A Maldjian Journal: AJNR Am J Neuroradiol Date: 2021-03-04 Impact factor: 3.825
Authors: Evi J van Kempen; Max Post; Manoj Mannil; Benno Kusters; Mark Ter Laan; Frederick J A Meijer; Dylan J H A Henssen Journal: Cancers (Basel) Date: 2021-05-26 Impact factor: 6.639