Radiomics-based neural network predicts recurrence patterns in glioblastoma using dynamic susceptibility contrast-enhanced MRI

Abstract Glioblastoma remains the most devastating brain tumor despite optimal treatment, because of the high rate of recurrence. Distant recurrence has distinct genomic alterations compared to local recurrence, which requires different treatment planning both in clinical practice and trials. To dat...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Ka Young Shim, Sung Won Chung, Jae Hak Jeong, Inpyeong Hwang, Chul-Kee Park, Tae Min Kim, Sung-Hye Park, Jae Kyung Won, Joo Ho Lee, Soon-Tae Lee, Roh-Eul Yoo, Koung Mi Kang, Tae Jin Yun, Ji-Hoon Kim, Chul-Ho Sohn, Kyu Sung Choi, Seung Hong Choi
Formato:	article
Lenguaje:	EN
Publicado:	Nature Portfolio 2021
Materias:	Medicine R Science Q
Acceso en línea:	https://doaj.org/article/980697b8918e4f6fac400772d53da387
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:980697b8918e4f6fac400772d53da387
record_format	dspace
spelling	oai:doaj.org-article:980697b8918e4f6fac400772d53da3872021-12-02T14:35:47ZRadiomics-based neural network predicts recurrence patterns in glioblastoma using dynamic susceptibility contrast-enhanced MRI10.1038/s41598-021-89218-z2045-2322https://doaj.org/article/980697b8918e4f6fac400772d53da3872021-05-01T00:00:00Zhttps://doi.org/10.1038/s41598-021-89218-zhttps://doaj.org/toc/2045-2322Abstract Glioblastoma remains the most devastating brain tumor despite optimal treatment, because of the high rate of recurrence. Distant recurrence has distinct genomic alterations compared to local recurrence, which requires different treatment planning both in clinical practice and trials. To date, perfusion-weighted MRI has revealed that perfusional characteristics of tumor are associated with prognosis. However, not much research has focused on recurrence patterns in glioblastoma: namely, local and distant recurrence. Here, we propose two different neural network models to predict the recurrence patterns in glioblastoma that utilizes high-dimensional radiomic profiles based on perfusion MRI: area under the curve (AUC) (95% confidence interval), 0.969 (0.903–1.000) for local recurrence; 0.864 (0.726–0.976) for distant recurrence for each patient in the validation set. This creates an opportunity to provide personalized medicine in contrast to studies investigating only group differences. Moreover, interpretable deep learning identified that salient radiomic features for each recurrence pattern are related to perfusional intratumoral heterogeneity. We also demonstrated that the combined salient radiomic features, or “radiomic risk score”, increased risk of recurrence/progression (hazard ratio, 1.61; p = 0.03) in multivariate Cox regression on progression-free survival.Ka Young ShimSung Won ChungJae Hak JeongInpyeong HwangChul-Kee ParkTae Min KimSung-Hye ParkJae Kyung WonJoo Ho LeeSoon-Tae LeeRoh-Eul YooKoung Mi KangTae Jin YunJi-Hoon KimChul-Ho SohnKyu Sung ChoiSeung Hong ChoiNature PortfolioarticleMedicineRScienceQENScientific Reports, Vol 11, Iss 1, Pp 1-14 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	Medicine R Science Q
spellingShingle	Medicine R Science Q Ka Young Shim Sung Won Chung Jae Hak Jeong Inpyeong Hwang Chul-Kee Park Tae Min Kim Sung-Hye Park Jae Kyung Won Joo Ho Lee Soon-Tae Lee Roh-Eul Yoo Koung Mi Kang Tae Jin Yun Ji-Hoon Kim Chul-Ho Sohn Kyu Sung Choi Seung Hong Choi Radiomics-based neural network predicts recurrence patterns in glioblastoma using dynamic susceptibility contrast-enhanced MRI
description	Abstract Glioblastoma remains the most devastating brain tumor despite optimal treatment, because of the high rate of recurrence. Distant recurrence has distinct genomic alterations compared to local recurrence, which requires different treatment planning both in clinical practice and trials. To date, perfusion-weighted MRI has revealed that perfusional characteristics of tumor are associated with prognosis. However, not much research has focused on recurrence patterns in glioblastoma: namely, local and distant recurrence. Here, we propose two different neural network models to predict the recurrence patterns in glioblastoma that utilizes high-dimensional radiomic profiles based on perfusion MRI: area under the curve (AUC) (95% confidence interval), 0.969 (0.903–1.000) for local recurrence; 0.864 (0.726–0.976) for distant recurrence for each patient in the validation set. This creates an opportunity to provide personalized medicine in contrast to studies investigating only group differences. Moreover, interpretable deep learning identified that salient radiomic features for each recurrence pattern are related to perfusional intratumoral heterogeneity. We also demonstrated that the combined salient radiomic features, or “radiomic risk score”, increased risk of recurrence/progression (hazard ratio, 1.61; p = 0.03) in multivariate Cox regression on progression-free survival.
format	article
author	Ka Young Shim Sung Won Chung Jae Hak Jeong Inpyeong Hwang Chul-Kee Park Tae Min Kim Sung-Hye Park Jae Kyung Won Joo Ho Lee Soon-Tae Lee Roh-Eul Yoo Koung Mi Kang Tae Jin Yun Ji-Hoon Kim Chul-Ho Sohn Kyu Sung Choi Seung Hong Choi
author_facet	Ka Young Shim Sung Won Chung Jae Hak Jeong Inpyeong Hwang Chul-Kee Park Tae Min Kim Sung-Hye Park Jae Kyung Won Joo Ho Lee Soon-Tae Lee Roh-Eul Yoo Koung Mi Kang Tae Jin Yun Ji-Hoon Kim Chul-Ho Sohn Kyu Sung Choi Seung Hong Choi
author_sort	Ka Young Shim
title	Radiomics-based neural network predicts recurrence patterns in glioblastoma using dynamic susceptibility contrast-enhanced MRI
title_short	Radiomics-based neural network predicts recurrence patterns in glioblastoma using dynamic susceptibility contrast-enhanced MRI
title_full	Radiomics-based neural network predicts recurrence patterns in glioblastoma using dynamic susceptibility contrast-enhanced MRI
title_fullStr	Radiomics-based neural network predicts recurrence patterns in glioblastoma using dynamic susceptibility contrast-enhanced MRI
title_full_unstemmed	Radiomics-based neural network predicts recurrence patterns in glioblastoma using dynamic susceptibility contrast-enhanced MRI
title_sort	radiomics-based neural network predicts recurrence patterns in glioblastoma using dynamic susceptibility contrast-enhanced mri
publisher	Nature Portfolio
publishDate	2021
url	https://doaj.org/article/980697b8918e4f6fac400772d53da387
work_keys_str_mv	AT kayoungshim radiomicsbasedneuralnetworkpredictsrecurrencepatternsinglioblastomausingdynamicsusceptibilitycontrastenhancedmri AT sungwonchung radiomicsbasedneuralnetworkpredictsrecurrencepatternsinglioblastomausingdynamicsusceptibilitycontrastenhancedmri AT jaehakjeong radiomicsbasedneuralnetworkpredictsrecurrencepatternsinglioblastomausingdynamicsusceptibilitycontrastenhancedmri AT inpyeonghwang radiomicsbasedneuralnetworkpredictsrecurrencepatternsinglioblastomausingdynamicsusceptibilitycontrastenhancedmri AT chulkeepark radiomicsbasedneuralnetworkpredictsrecurrencepatternsinglioblastomausingdynamicsusceptibilitycontrastenhancedmri AT taeminkim radiomicsbasedneuralnetworkpredictsrecurrencepatternsinglioblastomausingdynamicsusceptibilitycontrastenhancedmri AT sunghyepark radiomicsbasedneuralnetworkpredictsrecurrencepatternsinglioblastomausingdynamicsusceptibilitycontrastenhancedmri AT jaekyungwon radiomicsbasedneuralnetworkpredictsrecurrencepatternsinglioblastomausingdynamicsusceptibilitycontrastenhancedmri AT jooholee radiomicsbasedneuralnetworkpredictsrecurrencepatternsinglioblastomausingdynamicsusceptibilitycontrastenhancedmri AT soontaelee radiomicsbasedneuralnetworkpredictsrecurrencepatternsinglioblastomausingdynamicsusceptibilitycontrastenhancedmri AT roheulyoo radiomicsbasedneuralnetworkpredictsrecurrencepatternsinglioblastomausingdynamicsusceptibilitycontrastenhancedmri AT koungmikang radiomicsbasedneuralnetworkpredictsrecurrencepatternsinglioblastomausingdynamicsusceptibilitycontrastenhancedmri AT taejinyun radiomicsbasedneuralnetworkpredictsrecurrencepatternsinglioblastomausingdynamicsusceptibilitycontrastenhancedmri AT jihoonkim radiomicsbasedneuralnetworkpredictsrecurrencepatternsinglioblastomausingdynamicsusceptibilitycontrastenhancedmri AT chulhosohn radiomicsbasedneuralnetworkpredictsrecurrencepatternsinglioblastomausingdynamicsusceptibilitycontrastenhancedmri AT kyusungchoi radiomicsbasedneuralnetworkpredictsrecurrencepatternsinglioblastomausingdynamicsusceptibilitycontrastenhancedmri AT seunghongchoi radiomicsbasedneuralnetworkpredictsrecurrencepatternsinglioblastomausingdynamicsusceptibilitycontrastenhancedmri
_version_	1718391087024832512

Radiomics-based neural network predicts recurrence patterns in glioblastoma using dynamic susceptibility contrast-enhanced MRI

Ejemplares similares