Machine learning decision tree models for multiclass classification of common malignant brain tumors using perfusion and spectroscopy MRI data
BackgroundTo investigate the contribution of machine learning decision tree models applied to perfusion and spectroscopy MRI for multiclass classification of lymphomas, glioblastomas, and metastases, and then to bring out the underlying key pathophysiological processes involved in the hierarchizatio...
| Main Authors: | , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2023-08-01
|
| Series: | Frontiers in Oncology |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/articles/10.3389/fonc.2023.1089998/full |
| _version_ | 1827874073535840256 |
|---|---|
| author | Rodolphe Vallée Rodolphe Vallée Rodolphe Vallée Jean-Noël Vallée Jean-Noël Vallée Carole Guillevin Carole Guillevin Athéna Lallouette Clément Thomas Clément Thomas Guillaume Rittano Michel Wager Rémy Guillevin Rémy Guillevin Alexandre Vallée |
| author_facet | Rodolphe Vallée Rodolphe Vallée Rodolphe Vallée Jean-Noël Vallée Jean-Noël Vallée Carole Guillevin Carole Guillevin Athéna Lallouette Clément Thomas Clément Thomas Guillaume Rittano Michel Wager Rémy Guillevin Rémy Guillevin Alexandre Vallée |
| author_sort | Rodolphe Vallée |
| collection | DOAJ |
| description | BackgroundTo investigate the contribution of machine learning decision tree models applied to perfusion and spectroscopy MRI for multiclass classification of lymphomas, glioblastomas, and metastases, and then to bring out the underlying key pathophysiological processes involved in the hierarchization of the decision-making algorithms of the modelsMethodsFrom 2013 to 2020, 180 consecutive patients with histopathologically proved lymphomas (n = 77), glioblastomas (n = 45), and metastases (n = 58) were included in machine learning analysis after undergoing MRI. The perfusion parameters (rCBVmax, PSRmax) and spectroscopic concentration ratios (lac/Cr, Cho/NAA, Cho/Cr, and lip/Cr) were applied to construct Classification and Regression Tree (CART) models for multiclass classification of these brain tumors. A 5-fold random cross validation was performed on the dataset.ResultsThe decision tree model thus constructed successfully classified all 3 tumor types with a performance (AUC) of 0.98 for PCNSLs, 0.98 for GBM and 1.00 for METs. The model accuracy was 0.96 with a RSquare of 0.887. Five rules of classifier combinations were extracted with a predicted probability from 0.907 to 0.989 for that end nodes of the decision tree for tumor multiclass classification. In hierarchical order of importance, the root node (Cho/NAA) in the decision tree algorithm was primarily based on the proliferative, infiltrative, and neuronal destructive characteristics of the tumor, the internal node (PSRmax), on tumor tissue capillary permeability characteristics, and the end node (Lac/Cr or Cho/Cr), on tumor energy glycolytic (Warburg effect), or on membrane lipid tumor metabolism.ConclusionOur study shows potential implementation of machine learning decision tree model algorithms based on a hierarchical, convenient, and personalized use of perfusion and spectroscopy MRI data for multiclass classification of these brain tumors. |
| first_indexed | 2024-03-12T16:47:06Z |
| format | Article |
| id | doaj.art-f4e74cdcec66455ab6dd201b50ea530b |
| institution | Directory Open Access Journal |
| issn | 2234-943X |
| language | English |
| last_indexed | 2024-03-12T16:47:06Z |
| publishDate | 2023-08-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Oncology |
| spelling | doaj.art-f4e74cdcec66455ab6dd201b50ea530b2023-08-08T12:09:16ZengFrontiers Media S.A.Frontiers in Oncology2234-943X2023-08-011310.3389/fonc.2023.10899981089998Machine learning decision tree models for multiclass classification of common malignant brain tumors using perfusion and spectroscopy MRI dataRodolphe Vallée0Rodolphe Vallée1Rodolphe Vallée2Jean-Noël Vallée3Jean-Noël Vallée4Carole Guillevin5Carole Guillevin6Athéna Lallouette7Clément Thomas8Clément Thomas9Guillaume Rittano10Michel Wager11Rémy Guillevin12Rémy Guillevin13Alexandre Vallée14Interdisciplinary Laboratory in Neurosciences, Physiology and Psychology (LINP2), Université Paris Lumière (UPL), Paris Nanterre University, Nanterre, FranceLaboratory of Mathematics and Applications (LMA) Centre National de la Recherche Scientifique - Unité Mixte de Recherche (CNRS UMR)7348, i3M-DACTIM-MIH (Data Analysis and Computations Through Imaging Modeling - Mathematics, Image, Health), Poitiers University, Poitiers, FranceGlaucoma Research Center, Swiss Visio Network, Lausanne, SwitzerlandLaboratory of Mathematics and Applications (LMA) Centre National de la Recherche Scientifique - Unité Mixte de Recherche (CNRS UMR)7348, i3M-DACTIM-MIH (Data Analysis and Computations Through Imaging Modeling - Mathematics, Image, Health), Poitiers University, Poitiers, FranceDiagnostic and Functional Neuroradiology and Brain stimulation Department, 15-20 National Vision Hospital of Paris - Paris University Hospital Center, University of PARIS-SACLAY - UVSQ, Paris, FranceLaboratory of Mathematics and Applications (LMA) Centre National de la Recherche Scientifique - Unité Mixte de Recherche (CNRS UMR)7348, i3M-DACTIM-MIH (Data Analysis and Computations Through Imaging Modeling - Mathematics, Image, Health), Poitiers University, Poitiers, FranceRadiology Department, Poitiers University Hospital, Poitiers University, Poitiers, FranceCenter of Genève Ophtalmologie, Geneve, SwitzerlandLaboratory of Mathematics and Applications (LMA) Centre National de la Recherche Scientifique - Unité Mixte de Recherche (CNRS UMR)7348, i3M-DACTIM-MIH (Data Analysis and Computations Through Imaging Modeling - Mathematics, Image, Health), Poitiers University, Poitiers, FranceDiagnostic and Functional Neuroradiology and Brain stimulation Department, 15-20 National Vision Hospital of Paris - Paris University Hospital Center, University of PARIS-SACLAY - UVSQ, Paris, FranceRadiology Department, Hopital Riveira Chablais, Rennaz, SwitzerlandNeurosurgery Department, Poitiers University Hospital, Poitiers University, Poitiers, FranceLaboratory of Mathematics and Applications (LMA) Centre National de la Recherche Scientifique - Unité Mixte de Recherche (CNRS UMR)7348, i3M-DACTIM-MIH (Data Analysis and Computations Through Imaging Modeling - Mathematics, Image, Health), Poitiers University, Poitiers, FranceRadiology Department, Poitiers University Hospital, Poitiers University, Poitiers, FranceDepartment of Epidemiology and Public Health, Foch Hospital, Suresnes, FranceBackgroundTo investigate the contribution of machine learning decision tree models applied to perfusion and spectroscopy MRI for multiclass classification of lymphomas, glioblastomas, and metastases, and then to bring out the underlying key pathophysiological processes involved in the hierarchization of the decision-making algorithms of the modelsMethodsFrom 2013 to 2020, 180 consecutive patients with histopathologically proved lymphomas (n = 77), glioblastomas (n = 45), and metastases (n = 58) were included in machine learning analysis after undergoing MRI. The perfusion parameters (rCBVmax, PSRmax) and spectroscopic concentration ratios (lac/Cr, Cho/NAA, Cho/Cr, and lip/Cr) were applied to construct Classification and Regression Tree (CART) models for multiclass classification of these brain tumors. A 5-fold random cross validation was performed on the dataset.ResultsThe decision tree model thus constructed successfully classified all 3 tumor types with a performance (AUC) of 0.98 for PCNSLs, 0.98 for GBM and 1.00 for METs. The model accuracy was 0.96 with a RSquare of 0.887. Five rules of classifier combinations were extracted with a predicted probability from 0.907 to 0.989 for that end nodes of the decision tree for tumor multiclass classification. In hierarchical order of importance, the root node (Cho/NAA) in the decision tree algorithm was primarily based on the proliferative, infiltrative, and neuronal destructive characteristics of the tumor, the internal node (PSRmax), on tumor tissue capillary permeability characteristics, and the end node (Lac/Cr or Cho/Cr), on tumor energy glycolytic (Warburg effect), or on membrane lipid tumor metabolism.ConclusionOur study shows potential implementation of machine learning decision tree model algorithms based on a hierarchical, convenient, and personalized use of perfusion and spectroscopy MRI data for multiclass classification of these brain tumors.https://www.frontiersin.org/articles/10.3389/fonc.2023.1089998/fullclassification and regression tree (CART)multiclass classificationlymphomaglioblastomametastasis |
| spellingShingle | Rodolphe Vallée Rodolphe Vallée Rodolphe Vallée Jean-Noël Vallée Jean-Noël Vallée Carole Guillevin Carole Guillevin Athéna Lallouette Clément Thomas Clément Thomas Guillaume Rittano Michel Wager Rémy Guillevin Rémy Guillevin Alexandre Vallée Machine learning decision tree models for multiclass classification of common malignant brain tumors using perfusion and spectroscopy MRI data Frontiers in Oncology classification and regression tree (CART) multiclass classification lymphoma glioblastoma metastasis |
| title | Machine learning decision tree models for multiclass classification of common malignant brain tumors using perfusion and spectroscopy MRI data |
| title_full | Machine learning decision tree models for multiclass classification of common malignant brain tumors using perfusion and spectroscopy MRI data |
| title_fullStr | Machine learning decision tree models for multiclass classification of common malignant brain tumors using perfusion and spectroscopy MRI data |
| title_full_unstemmed | Machine learning decision tree models for multiclass classification of common malignant brain tumors using perfusion and spectroscopy MRI data |
| title_short | Machine learning decision tree models for multiclass classification of common malignant brain tumors using perfusion and spectroscopy MRI data |
| title_sort | machine learning decision tree models for multiclass classification of common malignant brain tumors using perfusion and spectroscopy mri data |
| topic | classification and regression tree (CART) multiclass classification lymphoma glioblastoma metastasis |
| url | https://www.frontiersin.org/articles/10.3389/fonc.2023.1089998/full |
| work_keys_str_mv | AT rodolphevallee machinelearningdecisiontreemodelsformulticlassclassificationofcommonmalignantbraintumorsusingperfusionandspectroscopymridata AT rodolphevallee machinelearningdecisiontreemodelsformulticlassclassificationofcommonmalignantbraintumorsusingperfusionandspectroscopymridata AT rodolphevallee machinelearningdecisiontreemodelsformulticlassclassificationofcommonmalignantbraintumorsusingperfusionandspectroscopymridata AT jeannoelvallee machinelearningdecisiontreemodelsformulticlassclassificationofcommonmalignantbraintumorsusingperfusionandspectroscopymridata AT jeannoelvallee machinelearningdecisiontreemodelsformulticlassclassificationofcommonmalignantbraintumorsusingperfusionandspectroscopymridata AT caroleguillevin machinelearningdecisiontreemodelsformulticlassclassificationofcommonmalignantbraintumorsusingperfusionandspectroscopymridata AT caroleguillevin machinelearningdecisiontreemodelsformulticlassclassificationofcommonmalignantbraintumorsusingperfusionandspectroscopymridata AT athenalallouette machinelearningdecisiontreemodelsformulticlassclassificationofcommonmalignantbraintumorsusingperfusionandspectroscopymridata AT clementthomas machinelearningdecisiontreemodelsformulticlassclassificationofcommonmalignantbraintumorsusingperfusionandspectroscopymridata AT clementthomas machinelearningdecisiontreemodelsformulticlassclassificationofcommonmalignantbraintumorsusingperfusionandspectroscopymridata AT guillaumerittano machinelearningdecisiontreemodelsformulticlassclassificationofcommonmalignantbraintumorsusingperfusionandspectroscopymridata AT michelwager machinelearningdecisiontreemodelsformulticlassclassificationofcommonmalignantbraintumorsusingperfusionandspectroscopymridata AT remyguillevin machinelearningdecisiontreemodelsformulticlassclassificationofcommonmalignantbraintumorsusingperfusionandspectroscopymridata AT remyguillevin machinelearningdecisiontreemodelsformulticlassclassificationofcommonmalignantbraintumorsusingperfusionandspectroscopymridata AT alexandrevallee machinelearningdecisiontreemodelsformulticlassclassificationofcommonmalignantbraintumorsusingperfusionandspectroscopymridata |