Predicting Regions of Local Recurrence in Glioblastomas Using Voxel-Based Radiomic Features of Multiparametric Postoperative MRI
The globally accepted surgical strategy in glioblastomas is removing the enhancing tumor. However, the peritumoral region harbors infiltration areas responsible for future tumor recurrence. This study aimed to evaluate a predictive model that identifies areas of future recurrence using a voxel-based...
Main Authors: | , , , , , , , , , , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2023-03-01
|
Series: | Cancers |
Subjects: | |
Online Access: | https://www.mdpi.com/2072-6694/15/6/1894 |
_version_ | 1797612928161021952 |
---|---|
author | Santiago Cepeda Luigi Tommaso Luppino Angel Pérez-Núñez Ole Solheim Sergio García-García María Velasco-Casares Anna Karlberg Live Eikenes Rosario Sarabia Ignacio Arrese Tomás Zamora Pedro Gonzalez Luis Jiménez-Roldán Samuel Kuttner |
author_facet | Santiago Cepeda Luigi Tommaso Luppino Angel Pérez-Núñez Ole Solheim Sergio García-García María Velasco-Casares Anna Karlberg Live Eikenes Rosario Sarabia Ignacio Arrese Tomás Zamora Pedro Gonzalez Luis Jiménez-Roldán Samuel Kuttner |
author_sort | Santiago Cepeda |
collection | DOAJ |
description | The globally accepted surgical strategy in glioblastomas is removing the enhancing tumor. However, the peritumoral region harbors infiltration areas responsible for future tumor recurrence. This study aimed to evaluate a predictive model that identifies areas of future recurrence using a voxel-based radiomics analysis of magnetic resonance imaging (MRI) data. This multi-institutional study included a retrospective analysis of patients diagnosed with glioblastoma who underwent surgery with complete resection of the enhancing tumor. Fifty-five patients met the selection criteria. The study sample was split into training (N = 40) and testing (N = 15) datasets. Follow-up MRI was used for ground truth definition, and postoperative structural multiparametric MRI was used to extract voxel-based radiomic features. Deformable coregistration was used to register the MRI sequences for each patient, followed by segmentation of the peritumoral region in the postoperative scan and the enhancing tumor in the follow-up scan. Peritumoral voxels overlapping with enhancing tumor voxels were labeled as recurrence, while non-overlapping voxels were labeled as nonrecurrence. Voxel-based radiomic features were extracted from the peritumoral region. Four machine learning-based classifiers were trained for recurrence prediction. A region-based evaluation approach was used for model evaluation. The Categorical Boosting (CatBoost) classifier obtained the best performance on the testing dataset with an average area under the curve (AUC) of 0.81 ± 0.09 and an accuracy of 0.84 ± 0.06, using region-based evaluation. There was a clear visual correspondence between predicted and actual recurrence regions. We have developed a method that accurately predicts the region of future tumor recurrence in MRI scans of glioblastoma patients. This could enable the adaptation of surgical and radiotherapy treatment to these areas to potentially prolong the survival of these patients. |
first_indexed | 2024-03-11T06:48:48Z |
format | Article |
id | doaj.art-9849897cdb174f9f9cccf331e9b21169 |
institution | Directory Open Access Journal |
issn | 2072-6694 |
language | English |
last_indexed | 2024-03-11T06:48:48Z |
publishDate | 2023-03-01 |
publisher | MDPI AG |
record_format | Article |
series | Cancers |
spelling | doaj.art-9849897cdb174f9f9cccf331e9b211692023-11-17T10:08:43ZengMDPI AGCancers2072-66942023-03-01156189410.3390/cancers15061894Predicting Regions of Local Recurrence in Glioblastomas Using Voxel-Based Radiomic Features of Multiparametric Postoperative MRISantiago Cepeda0Luigi Tommaso Luppino1Angel Pérez-Núñez2Ole Solheim3Sergio García-García4María Velasco-Casares5Anna Karlberg6Live Eikenes7Rosario Sarabia8Ignacio Arrese9Tomás Zamora10Pedro Gonzalez11Luis Jiménez-Roldán12Samuel Kuttner13Department of Neurosurgery, Río Hortega University Hospital, 47014 Valladolid, SpainDepartment of Physics and Technology, UiT The Arctic University of Norway, 9019 Tromsø, NorwayDepartment of Neurosurgery, 12 de Octubre University Hospital (i+12), 28041 Madrid, SpainDepartment of Neurosurgery, St. Olavs University Hospital, 7030 Trondheim, NorwayDepartment of Neurosurgery, Río Hortega University Hospital, 47014 Valladolid, SpainDepartment of Radiology, Río Hortega University Hospital, 47012 Valladolid, SpainDepartment of Circulation and Medical Imaging, Faculty of Medicine and Health Sciences, Norwegian University of Science and Technology (NTNU), 7034 Trondheim, NorwayDepartment of Circulation and Medical Imaging, Faculty of Medicine and Health Sciences, Norwegian University of Science and Technology (NTNU), 7034 Trondheim, NorwayDepartment of Neurosurgery, Río Hortega University Hospital, 47014 Valladolid, SpainDepartment of Neurosurgery, Río Hortega University Hospital, 47014 Valladolid, SpainDepartment of Pathology, Río Hortega University Hospital, 47014 Valladolid, SpainDepartment of Neurosurgery, 12 de Octubre University Hospital (i+12), 28041 Madrid, SpainDepartment of Neurosurgery, 12 de Octubre University Hospital (i+12), 28041 Madrid, SpainDepartment of Physics and Technology, UiT The Arctic University of Norway, 9019 Tromsø, NorwayThe globally accepted surgical strategy in glioblastomas is removing the enhancing tumor. However, the peritumoral region harbors infiltration areas responsible for future tumor recurrence. This study aimed to evaluate a predictive model that identifies areas of future recurrence using a voxel-based radiomics analysis of magnetic resonance imaging (MRI) data. This multi-institutional study included a retrospective analysis of patients diagnosed with glioblastoma who underwent surgery with complete resection of the enhancing tumor. Fifty-five patients met the selection criteria. The study sample was split into training (N = 40) and testing (N = 15) datasets. Follow-up MRI was used for ground truth definition, and postoperative structural multiparametric MRI was used to extract voxel-based radiomic features. Deformable coregistration was used to register the MRI sequences for each patient, followed by segmentation of the peritumoral region in the postoperative scan and the enhancing tumor in the follow-up scan. Peritumoral voxels overlapping with enhancing tumor voxels were labeled as recurrence, while non-overlapping voxels were labeled as nonrecurrence. Voxel-based radiomic features were extracted from the peritumoral region. Four machine learning-based classifiers were trained for recurrence prediction. A region-based evaluation approach was used for model evaluation. The Categorical Boosting (CatBoost) classifier obtained the best performance on the testing dataset with an average area under the curve (AUC) of 0.81 ± 0.09 and an accuracy of 0.84 ± 0.06, using region-based evaluation. There was a clear visual correspondence between predicted and actual recurrence regions. We have developed a method that accurately predicts the region of future tumor recurrence in MRI scans of glioblastoma patients. This could enable the adaptation of surgical and radiotherapy treatment to these areas to potentially prolong the survival of these patients.https://www.mdpi.com/2072-6694/15/6/1894glioblastomaartificial intelligenceMRIrecurrenceradiomicsmachine learning |
spellingShingle | Santiago Cepeda Luigi Tommaso Luppino Angel Pérez-Núñez Ole Solheim Sergio García-García María Velasco-Casares Anna Karlberg Live Eikenes Rosario Sarabia Ignacio Arrese Tomás Zamora Pedro Gonzalez Luis Jiménez-Roldán Samuel Kuttner Predicting Regions of Local Recurrence in Glioblastomas Using Voxel-Based Radiomic Features of Multiparametric Postoperative MRI Cancers glioblastoma artificial intelligence MRI recurrence radiomics machine learning |
title | Predicting Regions of Local Recurrence in Glioblastomas Using Voxel-Based Radiomic Features of Multiparametric Postoperative MRI |
title_full | Predicting Regions of Local Recurrence in Glioblastomas Using Voxel-Based Radiomic Features of Multiparametric Postoperative MRI |
title_fullStr | Predicting Regions of Local Recurrence in Glioblastomas Using Voxel-Based Radiomic Features of Multiparametric Postoperative MRI |
title_full_unstemmed | Predicting Regions of Local Recurrence in Glioblastomas Using Voxel-Based Radiomic Features of Multiparametric Postoperative MRI |
title_short | Predicting Regions of Local Recurrence in Glioblastomas Using Voxel-Based Radiomic Features of Multiparametric Postoperative MRI |
title_sort | predicting regions of local recurrence in glioblastomas using voxel based radiomic features of multiparametric postoperative mri |
topic | glioblastoma artificial intelligence MRI recurrence radiomics machine learning |
url | https://www.mdpi.com/2072-6694/15/6/1894 |
work_keys_str_mv | AT santiagocepeda predictingregionsoflocalrecurrenceinglioblastomasusingvoxelbasedradiomicfeaturesofmultiparametricpostoperativemri AT luigitommasoluppino predictingregionsoflocalrecurrenceinglioblastomasusingvoxelbasedradiomicfeaturesofmultiparametricpostoperativemri AT angelpereznunez predictingregionsoflocalrecurrenceinglioblastomasusingvoxelbasedradiomicfeaturesofmultiparametricpostoperativemri AT olesolheim predictingregionsoflocalrecurrenceinglioblastomasusingvoxelbasedradiomicfeaturesofmultiparametricpostoperativemri AT sergiogarciagarcia predictingregionsoflocalrecurrenceinglioblastomasusingvoxelbasedradiomicfeaturesofmultiparametricpostoperativemri AT mariavelascocasares predictingregionsoflocalrecurrenceinglioblastomasusingvoxelbasedradiomicfeaturesofmultiparametricpostoperativemri AT annakarlberg predictingregionsoflocalrecurrenceinglioblastomasusingvoxelbasedradiomicfeaturesofmultiparametricpostoperativemri AT liveeikenes predictingregionsoflocalrecurrenceinglioblastomasusingvoxelbasedradiomicfeaturesofmultiparametricpostoperativemri AT rosariosarabia predictingregionsoflocalrecurrenceinglioblastomasusingvoxelbasedradiomicfeaturesofmultiparametricpostoperativemri AT ignacioarrese predictingregionsoflocalrecurrenceinglioblastomasusingvoxelbasedradiomicfeaturesofmultiparametricpostoperativemri AT tomaszamora predictingregionsoflocalrecurrenceinglioblastomasusingvoxelbasedradiomicfeaturesofmultiparametricpostoperativemri AT pedrogonzalez predictingregionsoflocalrecurrenceinglioblastomasusingvoxelbasedradiomicfeaturesofmultiparametricpostoperativemri AT luisjimenezroldan predictingregionsoflocalrecurrenceinglioblastomasusingvoxelbasedradiomicfeaturesofmultiparametricpostoperativemri AT samuelkuttner predictingregionsoflocalrecurrenceinglioblastomasusingvoxelbasedradiomicfeaturesofmultiparametricpostoperativemri |