Prediction of IDH and TERT promoter mutations in low-grade glioma from magnetic resonance images using a convolutional neural network
Identification of genotypes is crucial for treatment of glioma. Here, we developed a method to predict tumor genotypes using a pretrained convolutional neural network (CNN) from magnetic resonance (MR) images and compared the accuracy to that of a diagnosis based on conventional radiomic features and patient age. Multisite preoperative MR images of 164 patients with grade II/III glioma were grouped by IDH and TERT promoter (pTERT) mutations as follows: (1) IDH wild type, (2) IDH and pTERT co-mutations, (3) IDH mutant and pTERT wild type. We applied a CNN (AlexNet) to four types of MR sequence and obtained the CNN texture features to classify the groups with a linear support vector machine. The classification was also performed using conventional radiomic features and/or patient age. Using all features, we succeeded in classifying patients with an accuracy of 63.1%, which was significantly higher than the accuracy obtained from using either the radiomic features or patient age alone. In particular, prediction of the pTERT mutation was significantly improved by the CNN texture features. In conclusion, the pretrained CNN texture features capture the information of IDH and TERT genotypes in grade II/III gliomas better than the conventional radiomic features.
| Medienart: |
E-Artikel |
|---|
| Erscheinungsjahr: |
2019 |
|---|---|
| Erschienen: |
2019 |
| Enthalten in: |
Zur Gesamtaufnahme - volume:9 |
|---|---|
| Enthalten in: |
Scientific reports - 9(2019), 1 vom: 30. Dez., Seite 20311 |
| Sprache: |
Englisch |
|---|
| Beteiligte Personen: |
Fukuma, Ryohei [VerfasserIn] |
|---|
| Links: |
|---|
| Themen: |
Biomarkers, Tumor |
|---|
| Anmerkungen: |
Date Completed 13.11.2020 Date Revised 10.01.2021 published: Electronic Citation Status MEDLINE |
|---|
| doi: |
10.1038/s41598-019-56767-3 |
|---|
| funding: |
|
|---|---|
| Förderinstitution / Projekttitel: |
|
| PPN (Katalog-ID): |
NLM304890448 |
|---|
| LEADER | 01000caa a22002652 4500 | ||
|---|---|---|---|
| 001 | NLM304890448 | ||
| 003 | DE-627 | ||
| 005 | 20231226200947.0 | ||
| 007 | cr uuu---uuuuu | ||
| 008 | 231225s2019 xx |||||o 00| ||eng c | ||
| 024 | 7 | |a 10.1038/s41598-019-56767-3 |2 doi | |
| 028 | 5 | 2 | |a pubmed24n1016.xml |
| 035 | |a (DE-627)NLM304890448 | ||
| 035 | |a (NLM)31889117 | ||
| 040 | |a DE-627 |b ger |c DE-627 |e rakwb | ||
| 041 | |a eng | ||
| 100 | 1 | |a Fukuma, Ryohei |e verfasserin |4 aut | |
| 245 | 1 | 0 | |a Prediction of IDH and TERT promoter mutations in low-grade glioma from magnetic resonance images using a convolutional neural network |
| 264 | 1 | |c 2019 | |
| 336 | |a Text |b txt |2 rdacontent | ||
| 337 | |a ƒaComputermedien |b c |2 rdamedia | ||
| 338 | |a ƒa Online-Ressource |b cr |2 rdacarrier | ||
| 500 | |a Date Completed 13.11.2020 | ||
| 500 | |a Date Revised 10.01.2021 | ||
| 500 | |a published: Electronic | ||
| 500 | |a Citation Status MEDLINE | ||
| 520 | |a Identification of genotypes is crucial for treatment of glioma. Here, we developed a method to predict tumor genotypes using a pretrained convolutional neural network (CNN) from magnetic resonance (MR) images and compared the accuracy to that of a diagnosis based on conventional radiomic features and patient age. Multisite preoperative MR images of 164 patients with grade II/III glioma were grouped by IDH and TERT promoter (pTERT) mutations as follows: (1) IDH wild type, (2) IDH and pTERT co-mutations, (3) IDH mutant and pTERT wild type. We applied a CNN (AlexNet) to four types of MR sequence and obtained the CNN texture features to classify the groups with a linear support vector machine. The classification was also performed using conventional radiomic features and/or patient age. Using all features, we succeeded in classifying patients with an accuracy of 63.1%, which was significantly higher than the accuracy obtained from using either the radiomic features or patient age alone. In particular, prediction of the pTERT mutation was significantly improved by the CNN texture features. In conclusion, the pretrained CNN texture features capture the information of IDH and TERT genotypes in grade II/III gliomas better than the conventional radiomic features | ||
| 650 | 4 | |a Journal Article | |
| 650 | 4 | |a Research Support, Non-U.S. Gov't | |
| 650 | 7 | |a Biomarkers, Tumor |2 NLM | |
| 650 | 7 | |a Isocitrate Dehydrogenase |2 NLM | |
| 650 | 7 | |a EC 1.1.1.41 |2 NLM | |
| 650 | 7 | |a TERT protein, human |2 NLM | |
| 650 | 7 | |a EC 2.7.7.49 |2 NLM | |
| 650 | 7 | |a Telomerase |2 NLM | |
| 650 | 7 | |a EC 2.7.7.49 |2 NLM | |
| 700 | 1 | |a Yanagisawa, Takufumi |e verfasserin |4 aut | |
| 700 | 1 | |a Kinoshita, Manabu |e verfasserin |4 aut | |
| 700 | 1 | |a Shinozaki, Takashi |e verfasserin |4 aut | |
| 700 | 1 | |a Arita, Hideyuki |e verfasserin |4 aut | |
| 700 | 1 | |a Kawaguchi, Atsushi |e verfasserin |4 aut | |
| 700 | 1 | |a Takahashi, Masamichi |e verfasserin |4 aut | |
| 700 | 1 | |a Narita, Yoshitaka |e verfasserin |4 aut | |
| 700 | 1 | |a Terakawa, Yuzo |e verfasserin |4 aut | |
| 700 | 1 | |a Tsuyuguchi, Naohiro |e verfasserin |4 aut | |
| 700 | 1 | |a Okita, Yoshiko |e verfasserin |4 aut | |
| 700 | 1 | |a Nonaka, Masahiro |e verfasserin |4 aut | |
| 700 | 1 | |a Moriuchi, Shusuke |e verfasserin |4 aut | |
| 700 | 1 | |a Takagaki, Masatoshi |e verfasserin |4 aut | |
| 700 | 1 | |a Fujimoto, Yasunori |e verfasserin |4 aut | |
| 700 | 1 | |a Fukai, Junya |e verfasserin |4 aut | |
| 700 | 1 | |a Izumoto, Shuichi |e verfasserin |4 aut | |
| 700 | 1 | |a Ishibashi, Kenichi |e verfasserin |4 aut | |
| 700 | 1 | |a Nakajima, Yoshikazu |e verfasserin |4 aut | |
| 700 | 1 | |a Shofuda, Tomoko |e verfasserin |4 aut | |
| 700 | 1 | |a Kanematsu, Daisuke |e verfasserin |4 aut | |
| 700 | 1 | |a Yoshioka, Ema |e verfasserin |4 aut | |
| 700 | 1 | |a Kodama, Yoshinori |e verfasserin |4 aut | |
| 700 | 1 | |a Mano, Masayuki |e verfasserin |4 aut | |
| 700 | 1 | |a Mori, Kanji |e verfasserin |4 aut | |
| 700 | 1 | |a Ichimura, Koichi |e verfasserin |4 aut | |
| 700 | 1 | |a Kanemura, Yonehiro |e verfasserin |4 aut | |
| 700 | 1 | |a Kishima, Haruhiko |e verfasserin |4 aut | |
| 773 | 0 | 8 | |i Enthalten in |t Scientific reports |d 2011 |g 9(2019), 1 vom: 30. Dez., Seite 20311 |w (DE-627)NLM215703936 |x 2045-2322 |7 nnns |
| 773 | 1 | 8 | |g volume:9 |g year:2019 |g number:1 |g day:30 |g month:12 |g pages:20311 |
| 856 | 4 | 0 | |u http://dx.doi.org/10.1038/s41598-019-56767-3 |3 Volltext |
| 912 | |a GBV_USEFLAG_A | ||
| 912 | |a GBV_NLM | ||
| 951 | |a AR | ||
| 952 | |d 9 |j 2019 |e 1 |b 30 |c 12 |h 20311 | ||