Radiophysiomics: Brain Tumors Classification by Machine Learning and Physiological MRI Data
The precise initial characterization of contrast-enhancing brain tumors has significant consequences for clinical outcomes. Various novel neuroimaging methods have been developed to increase the specificity of conventional magnetic resonance imaging (cMRI) but also the increased complexity of data a...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-05-01
|
| Series: | Cancers |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2072-6694/14/10/2363 |
| _version_ | 1797501077698904064 |
|---|---|
| author | Andreas Stadlbauer Franz Marhold Stefan Oberndorfer Gertraud Heinz Michael Buchfelder Thomas M. Kinfe Anke Meyer-Bäse |
| author_facet | Andreas Stadlbauer Franz Marhold Stefan Oberndorfer Gertraud Heinz Michael Buchfelder Thomas M. Kinfe Anke Meyer-Bäse |
| author_sort | Andreas Stadlbauer |
| collection | DOAJ |
| description | The precise initial characterization of contrast-enhancing brain tumors has significant consequences for clinical outcomes. Various novel neuroimaging methods have been developed to increase the specificity of conventional magnetic resonance imaging (cMRI) but also the increased complexity of data analysis. Artificial intelligence offers new options to manage this challenge in clinical settings. Here, we investigated whether multiclass machine learning (ML) algorithms applied to a high-dimensional panel of radiomic features from advanced MRI (advMRI) and physiological MRI (phyMRI; thus, radiophysiomics) could reliably classify contrast-enhancing brain tumors. The recently developed phyMRI technique enables the quantitative assessment of microvascular architecture, neovascularization, oxygen metabolism, and tissue hypoxia. A training cohort of 167 patients suffering from one of the five most common brain tumor entities (glioblastoma, anaplastic glioma, meningioma, primary CNS lymphoma, or brain metastasis), combined with nine common ML algorithms, was used to develop overall 135 classifiers. Multiclass classification performance was investigated using tenfold cross-validation and an independent test cohort. Adaptive boosting and random forest in combination with advMRI and phyMRI data were superior to human reading in accuracy (0.875 vs. 0.850), precision (0.862 vs. 0.798), F-score (0.774 vs. 0.740), AUROC (0.886 vs. 0.813), and classification error (5 vs. 6). The radiologists, however, showed a higher sensitivity (0.767 vs. 0.750) and specificity (0.925 vs. 0.902). We demonstrated that ML-based radiophysiomics could be helpful in the clinical routine diagnosis of contrast-enhancing brain tumors; however, a high expenditure of time and work for data preprocessing requires the inclusion of deep neural networks. |
| first_indexed | 2024-03-10T03:13:09Z |
| format | Article |
| id | doaj.art-ef8d58aa53b74b1c826fe4d5a627eda9 |
| institution | Directory Open Access Journal |
| issn | 2072-6694 |
| language | English |
| last_indexed | 2024-03-10T03:13:09Z |
| publishDate | 2022-05-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Cancers |
| spelling | doaj.art-ef8d58aa53b74b1c826fe4d5a627eda92023-11-23T10:22:06ZengMDPI AGCancers2072-66942022-05-011410236310.3390/cancers14102363Radiophysiomics: Brain Tumors Classification by Machine Learning and Physiological MRI DataAndreas Stadlbauer0Franz Marhold1Stefan Oberndorfer2Gertraud Heinz3Michael Buchfelder4Thomas M. Kinfe5Anke Meyer-Bäse6Institute of Medical Radiology, University Clinic St. Pölten, Karl Landsteiner University of Health Sciences, A-3100 St. Pölten, AustriaDepartment of Neurosurgery, University Clinic of St. Pölten, Karl Landsteiner University of Health Sciences, A-3100 St. Pölten, AustriaDepartment of Neurology, University Clinic of St. Pölten, Karl Landsteiner University of Health Sciences, A-3100 St. Pölten, AustriaInstitute of Medical Radiology, University Clinic St. Pölten, Karl Landsteiner University of Health Sciences, A-3100 St. Pölten, AustriaDepartment of Neurosurgery, Friedrich-Alexander University (FAU) Erlangen-Nürnberg, D-91054 Erlangen, GermanyDepartment of Neurosurgery, Friedrich-Alexander University (FAU) Erlangen-Nürnberg, D-91054 Erlangen, GermanyDepartment of Scientific Computing, Florida State University, 400 Dirac Science Library, Tallahassee, FL 32306-4120, USAThe precise initial characterization of contrast-enhancing brain tumors has significant consequences for clinical outcomes. Various novel neuroimaging methods have been developed to increase the specificity of conventional magnetic resonance imaging (cMRI) but also the increased complexity of data analysis. Artificial intelligence offers new options to manage this challenge in clinical settings. Here, we investigated whether multiclass machine learning (ML) algorithms applied to a high-dimensional panel of radiomic features from advanced MRI (advMRI) and physiological MRI (phyMRI; thus, radiophysiomics) could reliably classify contrast-enhancing brain tumors. The recently developed phyMRI technique enables the quantitative assessment of microvascular architecture, neovascularization, oxygen metabolism, and tissue hypoxia. A training cohort of 167 patients suffering from one of the five most common brain tumor entities (glioblastoma, anaplastic glioma, meningioma, primary CNS lymphoma, or brain metastasis), combined with nine common ML algorithms, was used to develop overall 135 classifiers. Multiclass classification performance was investigated using tenfold cross-validation and an independent test cohort. Adaptive boosting and random forest in combination with advMRI and phyMRI data were superior to human reading in accuracy (0.875 vs. 0.850), precision (0.862 vs. 0.798), F-score (0.774 vs. 0.740), AUROC (0.886 vs. 0.813), and classification error (5 vs. 6). The radiologists, however, showed a higher sensitivity (0.767 vs. 0.750) and specificity (0.925 vs. 0.902). We demonstrated that ML-based radiophysiomics could be helpful in the clinical routine diagnosis of contrast-enhancing brain tumors; however, a high expenditure of time and work for data preprocessing requires the inclusion of deep neural networks.https://www.mdpi.com/2072-6694/14/10/2363brain tumorspretreatment classificationartificial intelligencemachine learningphysiological MRIneuro-oncology |
| spellingShingle | Andreas Stadlbauer Franz Marhold Stefan Oberndorfer Gertraud Heinz Michael Buchfelder Thomas M. Kinfe Anke Meyer-Bäse Radiophysiomics: Brain Tumors Classification by Machine Learning and Physiological MRI Data Cancers brain tumors pretreatment classification artificial intelligence machine learning physiological MRI neuro-oncology |
| title | Radiophysiomics: Brain Tumors Classification by Machine Learning and Physiological MRI Data |
| title_full | Radiophysiomics: Brain Tumors Classification by Machine Learning and Physiological MRI Data |
| title_fullStr | Radiophysiomics: Brain Tumors Classification by Machine Learning and Physiological MRI Data |
| title_full_unstemmed | Radiophysiomics: Brain Tumors Classification by Machine Learning and Physiological MRI Data |
| title_short | Radiophysiomics: Brain Tumors Classification by Machine Learning and Physiological MRI Data |
| title_sort | radiophysiomics brain tumors classification by machine learning and physiological mri data |
| topic | brain tumors pretreatment classification artificial intelligence machine learning physiological MRI neuro-oncology |
| url | https://www.mdpi.com/2072-6694/14/10/2363 |
| work_keys_str_mv | AT andreasstadlbauer radiophysiomicsbraintumorsclassificationbymachinelearningandphysiologicalmridata AT franzmarhold radiophysiomicsbraintumorsclassificationbymachinelearningandphysiologicalmridata AT stefanoberndorfer radiophysiomicsbraintumorsclassificationbymachinelearningandphysiologicalmridata AT gertraudheinz radiophysiomicsbraintumorsclassificationbymachinelearningandphysiologicalmridata AT michaelbuchfelder radiophysiomicsbraintumorsclassificationbymachinelearningandphysiologicalmridata AT thomasmkinfe radiophysiomicsbraintumorsclassificationbymachinelearningandphysiologicalmridata AT ankemeyerbase radiophysiomicsbraintumorsclassificationbymachinelearningandphysiologicalmridata |