Observing deep radiomics for the classification of glioma grades

Abstract Deep learning is a promising method for medical image analysis because it can automatically acquire meaningful representations from raw data. However, a technical challenge lies in the difficulty of determining which types of internal representation are associated with a specific task, beca...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Kazuma Kobayashi, Mototaka Miyake, Masamichi Takahashi, Ryuji Hamamoto
Formato:	article
Lenguaje:	EN
Publicado:	Nature Portfolio 2021
Materias:	Medicine R Science Q
Acceso en línea:	https://doaj.org/article/7d9ab30c52ab416ebfdf238913536662
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:7d9ab30c52ab416ebfdf238913536662
record_format	dspace
spelling	oai:doaj.org-article:7d9ab30c52ab416ebfdf2389135366622021-12-02T14:49:10ZObserving deep radiomics for the classification of glioma grades10.1038/s41598-021-90555-22045-2322https://doaj.org/article/7d9ab30c52ab416ebfdf2389135366622021-05-01T00:00:00Zhttps://doi.org/10.1038/s41598-021-90555-2https://doaj.org/toc/2045-2322Abstract Deep learning is a promising method for medical image analysis because it can automatically acquire meaningful representations from raw data. However, a technical challenge lies in the difficulty of determining which types of internal representation are associated with a specific task, because feature vectors can vary dynamically according to individual inputs. Here, based on the magnetic resonance imaging (MRI) of gliomas, we propose a novel method to extract a shareable set of feature vectors that encode various parts in tumor imaging phenotypes. By applying vector quantization to latent representations, features extracted by an encoder are replaced with a fixed set of feature vectors. Hence, the set of feature vectors can be used in downstream tasks as imaging markers, which we call deep radiomics. Using deep radiomics, a classifier is established using logistic regression to predict the glioma grade with 90% accuracy. We also devise an algorithm to visualize the image region encoded by each feature vector, and demonstrate that the classification model preferentially relies on feature vectors associated with the presence or absence of contrast enhancement in tumor regions. Our proposal provides a data-driven approach to enhance the understanding of the imaging appearance of gliomas.Kazuma KobayashiMototaka MiyakeMasamichi TakahashiRyuji HamamotoNature PortfolioarticleMedicineRScienceQENScientific Reports, Vol 11, Iss 1, Pp 1-13 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	Medicine R Science Q
spellingShingle	Medicine R Science Q Kazuma Kobayashi Mototaka Miyake Masamichi Takahashi Ryuji Hamamoto Observing deep radiomics for the classification of glioma grades
description	Abstract Deep learning is a promising method for medical image analysis because it can automatically acquire meaningful representations from raw data. However, a technical challenge lies in the difficulty of determining which types of internal representation are associated with a specific task, because feature vectors can vary dynamically according to individual inputs. Here, based on the magnetic resonance imaging (MRI) of gliomas, we propose a novel method to extract a shareable set of feature vectors that encode various parts in tumor imaging phenotypes. By applying vector quantization to latent representations, features extracted by an encoder are replaced with a fixed set of feature vectors. Hence, the set of feature vectors can be used in downstream tasks as imaging markers, which we call deep radiomics. Using deep radiomics, a classifier is established using logistic regression to predict the glioma grade with 90% accuracy. We also devise an algorithm to visualize the image region encoded by each feature vector, and demonstrate that the classification model preferentially relies on feature vectors associated with the presence or absence of contrast enhancement in tumor regions. Our proposal provides a data-driven approach to enhance the understanding of the imaging appearance of gliomas.
format	article
author	Kazuma Kobayashi Mototaka Miyake Masamichi Takahashi Ryuji Hamamoto
author_facet	Kazuma Kobayashi Mototaka Miyake Masamichi Takahashi Ryuji Hamamoto
author_sort	Kazuma Kobayashi
title	Observing deep radiomics for the classification of glioma grades
title_short	Observing deep radiomics for the classification of glioma grades
title_full	Observing deep radiomics for the classification of glioma grades
title_fullStr	Observing deep radiomics for the classification of glioma grades
title_full_unstemmed	Observing deep radiomics for the classification of glioma grades
title_sort	observing deep radiomics for the classification of glioma grades
publisher	Nature Portfolio
publishDate	2021
url	https://doaj.org/article/7d9ab30c52ab416ebfdf238913536662
work_keys_str_mv	AT kazumakobayashi observingdeepradiomicsfortheclassificationofgliomagrades AT mototakamiyake observingdeepradiomicsfortheclassificationofgliomagrades AT masamichitakahashi observingdeepradiomicsfortheclassificationofgliomagrades AT ryujihamamoto observingdeepradiomicsfortheclassificationofgliomagrades
_version_	1718389519974137856

Observing deep radiomics for the classification of glioma grades

Ejemplares similares