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

Abstract Deep learning-based radiomics (DLR) was developed to extract deep information from multiple modalities of magnetic resonance (MR) images. The performance of DLR for predicting the mutation status of isocitrate dehydrogenase 1 (IDH1) was validated in a dataset of 151 patients with low-grade...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Zeju Li, Yuanyuan Wang, Jinhua Yu, Yi Guo, Wei Cao
Formato:	article
Lenguaje:	EN
Publicado:	Nature Portfolio 2017
Materias:	Medicine R Science Q
Acceso en línea:	https://doaj.org/article/1b4a7db0c4aa498c9fcfebaaadcbb792
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:1b4a7db0c4aa498c9fcfebaaadcbb792
record_format	dspace
spelling	oai:doaj.org-article:1b4a7db0c4aa498c9fcfebaaadcbb7922021-12-02T11:41:19ZDeep Learning based Radiomics (DLR) and its usage in noninvasive IDH1 prediction for low grade glioma10.1038/s41598-017-05848-22045-2322https://doaj.org/article/1b4a7db0c4aa498c9fcfebaaadcbb7922017-07-01T00:00:00Zhttps://doi.org/10.1038/s41598-017-05848-2https://doaj.org/toc/2045-2322Abstract Deep learning-based radiomics (DLR) was developed to extract deep information from multiple modalities of magnetic resonance (MR) images. The performance of DLR for predicting the mutation status of isocitrate dehydrogenase 1 (IDH1) was validated in a dataset of 151 patients with low-grade glioma. A modified convolutional neural network (CNN) structure with 6 convolutional layers and a fully connected layer with 4096 neurons was used to segment tumors. Instead of calculating image features from segmented images, as typically performed for normal radiomics approaches, image features were obtained by normalizing the information of the last convolutional layers of the CNN. Fisher vector was used to encode the CNN features from image slices of different sizes. High-throughput features with dimensionality greater than 1.6*104 were obtained from the CNN. Paired t-tests and F-scores were used to select CNN features that were able to discriminate IDH1. With the same dataset, the area under the operating characteristic curve (AUC) of the normal radiomics method was 86% for IDH1 estimation, whereas for DLR the AUC was 92%. The AUC of IDH1 estimation was further improved to 95% using DLR based on multiple-modality MR images. DLR could be a powerful way to extract deep information from medical images.Zeju LiYuanyuan WangJinhua YuYi GuoWei CaoNature PortfolioarticleMedicineRScienceQENScientific Reports, Vol 7, Iss 1, Pp 1-11 (2017)
institution	DOAJ
collection	DOAJ
language	EN
topic	Medicine R Science Q
spellingShingle	Medicine R Science Q Zeju Li Yuanyuan Wang Jinhua Yu Yi Guo Wei Cao Deep Learning based Radiomics (DLR) and its usage in noninvasive IDH1 prediction for low grade glioma
description	Abstract Deep learning-based radiomics (DLR) was developed to extract deep information from multiple modalities of magnetic resonance (MR) images. The performance of DLR for predicting the mutation status of isocitrate dehydrogenase 1 (IDH1) was validated in a dataset of 151 patients with low-grade glioma. A modified convolutional neural network (CNN) structure with 6 convolutional layers and a fully connected layer with 4096 neurons was used to segment tumors. Instead of calculating image features from segmented images, as typically performed for normal radiomics approaches, image features were obtained by normalizing the information of the last convolutional layers of the CNN. Fisher vector was used to encode the CNN features from image slices of different sizes. High-throughput features with dimensionality greater than 1.6*104 were obtained from the CNN. Paired t-tests and F-scores were used to select CNN features that were able to discriminate IDH1. With the same dataset, the area under the operating characteristic curve (AUC) of the normal radiomics method was 86% for IDH1 estimation, whereas for DLR the AUC was 92%. The AUC of IDH1 estimation was further improved to 95% using DLR based on multiple-modality MR images. DLR could be a powerful way to extract deep information from medical images.
format	article
author	Zeju Li Yuanyuan Wang Jinhua Yu Yi Guo Wei Cao
author_facet	Zeju Li Yuanyuan Wang Jinhua Yu Yi Guo Wei Cao
author_sort	Zeju Li
title	Deep Learning based Radiomics (DLR) and its usage in noninvasive IDH1 prediction for low grade glioma
title_short	Deep Learning based Radiomics (DLR) and its usage in noninvasive IDH1 prediction for low grade glioma
title_full	Deep Learning based Radiomics (DLR) and its usage in noninvasive IDH1 prediction for low grade glioma
title_fullStr	Deep Learning based Radiomics (DLR) and its usage in noninvasive IDH1 prediction for low grade glioma
title_full_unstemmed	Deep Learning based Radiomics (DLR) and its usage in noninvasive IDH1 prediction for low grade glioma
title_sort	deep learning based radiomics (dlr) and its usage in noninvasive idh1 prediction for low grade glioma
publisher	Nature Portfolio
publishDate	2017
url	https://doaj.org/article/1b4a7db0c4aa498c9fcfebaaadcbb792
work_keys_str_mv	AT zejuli deeplearningbasedradiomicsdlranditsusageinnoninvasiveidh1predictionforlowgradeglioma AT yuanyuanwang deeplearningbasedradiomicsdlranditsusageinnoninvasiveidh1predictionforlowgradeglioma AT jinhuayu deeplearningbasedradiomicsdlranditsusageinnoninvasiveidh1predictionforlowgradeglioma AT yiguo deeplearningbasedradiomicsdlranditsusageinnoninvasiveidh1predictionforlowgradeglioma AT weicao deeplearningbasedradiomicsdlranditsusageinnoninvasiveidh1predictionforlowgradeglioma
_version_	1718395387217182720

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

Ejemplares similares