AI and High-Grade Glioma for Diagnosis and Outcome Prediction: Do All Machine Learning Models Perform Equally Well?

Frontiers in Oncology

卷:11

Antonella Stoppacciaro¹  Giulio Ranazzi¹  Martina Lucignani²  Antonio Napolitano²  Emanuela Tagliente²  Veronica Villani³  Maria Camilla Rossi-Espagnet⁵  Andrea Romano⁶  Luca Pasquini⁶  Francesco Dellepiane⁶  Alessandro Bozzao⁶  Alberto Di Napoli⁷  Antonello Vidiri⁸  Matteo Ritrovato⁹ 

[1] Department of Clinical and Molecular Medicine, Surgical Pathology Units, Sant’Andrea Hospital, La Sapienza University, Rome, Italy;

[2] Medical Physics Department, Bambino Gesù Children’s Hospital, Scientific Institute for Research, Hospitalization and Healthcare (IRCCS), Rome, Italy;

[3] Neuro-Oncology Unit, Regina Elena National Cancer Institute, Scientific Institute for Research, Hospitalization and Healthcare (IRCCS), Rome, Italy;

[4] Neuroradiology Service, Department of Radiology, Memorial Sloan Kettering Cancer Center, New York, NY, United States;

[5] Neuroradiology Unit, Imaging Department, Bambino Gesù Children’s Hospital, Scientific Institute for Research, Hospitalization and Healthcare (IRCCS), Rome, Italy;

[6] Neuroradiology Unit, Neuroscience, Mental Health and Sensory Organs (NESMOS) Department, Sant’Andrea Hospital, La Sapienza University, Rome, Italy;

[7] Radiology Department, Castelli Romani Hospital, Rome, Italy;

[8] Radiology and Diagnostic Imaging Department, Regina Elena National Cancer Institute, Scientific Institute for Research, Hospitalization and Healthcare (IRCCS), Rome, Italy;

[9] Unit of Health Technology Assessment (HTA), Biomedical Technology Risk Manager, Bambino Gesù Children’s Hospital, Scientific Institute for Research, Hospitalization and Healthcare (IRCCS), Rome, Italy;

关键词: glioblastoma;    machine learning;    radiomics;    survival;    high-grade glioma (HGG);    genetics;   

DOI  :  10.3389/fonc.2021.601425

来源: DOAJ

【 摘 要 】

Radiomic models outperform clinical data for outcome prediction in high-grade gliomas (HGG). However, lack of parameter standardization limits clinical applications. Many machine learning (ML) radiomic models employ single classifiers rather than ensemble learning, which is known to boost performance, and comparative analyses are lacking in the literature. We aimed to compare ML classifiers to predict clinically relevant tasks for HGG: overall survival (OS), isocitrate dehydrogenase (IDH) mutation, O-6-methylguanine-DNA-methyltransferase (MGMT) promoter methylation, epidermal growth factor receptor vIII (EGFR) amplification, and Ki-67 expression, based on radiomic features from conventional and advanced magnetic resonance imaging (MRI). Our objective was to identify the best algorithm for each task. One hundred fifty-six adult patients with pathologic diagnosis of HGG were included. Three tumoral regions were manually segmented: contrast-enhancing tumor, necrosis, and non-enhancing tumor. Radiomic features were extracted with a custom version of Pyradiomics and selected through Boruta algorithm. A Grid Search algorithm was applied when computing ten times K-fold cross-validation (K=10) to get the highest mean and lowest spread of accuracy. Model performance was assessed as AUC-ROC curve mean values with 95% confidence intervals (CI). Extreme Gradient Boosting (xGB) obtained highest accuracy for OS (74,5%), Adaboost (AB) for IDH mutation (87.5%), MGMT methylation (70,8%), Ki-67 expression (86%), and EGFR amplification (81%). Ensemble classifiers showed the best performance across tasks. High-scoring radiomic features shed light on possible correlations between MRI and tumor histology.

【 授权许可】

Unknown