TY - JOUR
T1 - AI-based classification of three common malignant tumors in neuro-oncology
T2 - A multi-institutional comparison of machine learning and deep learning methods
AU - Bathla, Girish
AU - Dhruba, Durjoy Deb
AU - Soni, Neetu
AU - Liu, Yanan
AU - Larson, Nicholas B.
AU - Kassmeyer, Blake A.
AU - Mohan, Suyash
AU - Roberts-Wolfe, Douglas
AU - Rathore, Saima
AU - Le, Nam H.
AU - Zhang, Honghai
AU - Sonka, Milan
AU - Priya, Sarv
N1 - Publisher Copyright:
© 2023 Elsevier Masson SAS
PY - 2023
Y1 - 2023
N2 - Purpose: To determine if machine learning (ML) or deep learning (DL) pipelines perform better in AI-based three-class classification of glioblastoma (GBM), intracranial metastatic disease (IMD) and primary CNS lymphoma (PCNSL). Methodology: Retrospective analysis included 502 cases for training (208 GBM, 67 PCNSL and 227 IMD), with external validation on 86 cases (27:27:32). Multiparametric MRI images (T1W, T2W, FLAIR, DWI and T1-CE) were co-registered, resampled, denoised and intensity normalized, followed by semiautomatic 3D segmentation of the enhancing tumor (ET) and peritumoral region (PTR). Model performance was assessed using several ML pipelines and 3D-convolutional neural networks (3D-CNN) using sequence specific masks, as well as combination of masks. All pipelines were trained and evaluated with 5-fold nested cross-validation on internal data followed by external validation using multi-class AUC. Results: Two ML models achieved similar performance on test set, one using T2-ET and T2-PTR masks (AUC: 0.885, 95% CI: [0.816, 0.935] and another using T1-CE-ET and FLAIR-PTR mask (AUC: 0.878, CI: [0.804, 0.930]). The best performing DL models achieved an AUC of 0.854, (CI [0.774, 0.914]) on external data using T1-CE-ET and T2-PTR masks, followed by model derived from T1-CE-ET, ADC-ET and FLAIR-PTR masks (AUC: 0.851, CI [0.772, 0.909]). Conclusion: Both ML and DL derived pipelines achieved similar performance. T1-CE mask was used in three of the top four overall models. Additionally, all four models had some mask derived from PTR, either T2WI or FLAIR.
AB - Purpose: To determine if machine learning (ML) or deep learning (DL) pipelines perform better in AI-based three-class classification of glioblastoma (GBM), intracranial metastatic disease (IMD) and primary CNS lymphoma (PCNSL). Methodology: Retrospective analysis included 502 cases for training (208 GBM, 67 PCNSL and 227 IMD), with external validation on 86 cases (27:27:32). Multiparametric MRI images (T1W, T2W, FLAIR, DWI and T1-CE) were co-registered, resampled, denoised and intensity normalized, followed by semiautomatic 3D segmentation of the enhancing tumor (ET) and peritumoral region (PTR). Model performance was assessed using several ML pipelines and 3D-convolutional neural networks (3D-CNN) using sequence specific masks, as well as combination of masks. All pipelines were trained and evaluated with 5-fold nested cross-validation on internal data followed by external validation using multi-class AUC. Results: Two ML models achieved similar performance on test set, one using T2-ET and T2-PTR masks (AUC: 0.885, 95% CI: [0.816, 0.935] and another using T1-CE-ET and FLAIR-PTR mask (AUC: 0.878, CI: [0.804, 0.930]). The best performing DL models achieved an AUC of 0.854, (CI [0.774, 0.914]) on external data using T1-CE-ET and T2-PTR masks, followed by model derived from T1-CE-ET, ADC-ET and FLAIR-PTR masks (AUC: 0.851, CI [0.772, 0.909]). Conclusion: Both ML and DL derived pipelines achieved similar performance. T1-CE mask was used in three of the top four overall models. Additionally, all four models had some mask derived from PTR, either T2WI or FLAIR.
KW - Artificial intelligence
KW - CNS lymphoma
KW - Glioblastoma
KW - Metastasis
KW - Radiomics
UR - http://www.scopus.com/inward/record.url?scp=85171200649&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85171200649&partnerID=8YFLogxK
U2 - 10.1016/j.neurad.2023.08.007
DO - 10.1016/j.neurad.2023.08.007
M3 - Article
AN - SCOPUS:85171200649
SN - 0150-9861
JO - Journal of Neuroradiology
JF - Journal of Neuroradiology
ER -