Table 1.
Summary of studies.
| Author, Journal, Year | Dataset (No.) | Gliomas (No.) | PCNSLs (No.) | Imaging protocol | Method | Classifiers |
|---|---|---|---|---|---|---|
| Bathla, European Radiology, 2021 (12) | 94 | 60 | 34 | T1W, CE-T1W, T2W, T2-FLAIR, DWI | ML | Linear Regression, LR, RR, ENR, LASSO, NN, SVM with a polynomial kernel, SVM with a radial kernel, MLP, RF, GBRM, AdaBoost |
| Chen, The International Journal of Neuroscience, 2018 (13) | 96 | 66 | 30 | CE-T1W | DL | SVM |
| Kang, Neuro-Oncology 2018 (14) | 196 | 119 | 77 | T1W, CE-T1W, T2W, T2-FLAIR, DWI, PWI | ML | K-NN, NB, DT, LDA, RF, AdaBoost, Linear SVM, RBF kernel SVM |
| Kim, Neuroradiology, 2018 (15) | 143 | 78 | 65 | T1-FFE, T2W, DWI, T2-FLAIR | ML | LR, SVM, RF |
| Kong, Neuroimage Clinical, 2019 (16) | 77 | 53 | 24 | 18F-FDG-PET/CT | ML | DT |
| Lu, Frontiers in neurology, 2022 (17) | 101 | 51 | 50 | CT scans | ML | LR, RF, DT, K-NN, SVM, NB |
| Lv, Journal of Neurosurgery, 2022 (18) | 103 | 68 | 35 | CE-T1W | ML | k-NN, GNB, RF, LR, SVM, MLP, AdaBoost |
| Priya, Neuroradiol J., 2021 (19) | 143 | 97 | 46 | T1W, CE-T1W, T2W, T2-FLAIR, DWI, PWI | ML | Linear regression, multinomial logistic, RR, elastic net, LASSO, NN, SVM with a polynomial kernel, SVM with a radial kernel, MLP, RF, GBRM, AdaBoost |
| Wu, IEEE Transanctions On Medical Imaging, 2018 (20) | 102 | 70 | 32 | CE-T1W, T2W | ML, DL | Sparse Representation, CNN |
| Xia, Journal of Magnetic Resonance Imaging, 2020 (21) | 240 | 129 | 111 | CE-T1W, T2-FLAIR, DWI | ML | LASSO, Multi-variable LR |
| Xia, Journal of Magnetic Resonance Imaging, 2021 (22) | 289 | 153 | 136 | T1W, T2-FLAIR, DWI | DL | CNN |
| Yun, scientific reports, 2019 (23) | 195 | 195 | 119 | CE-T1W, DWI | DL | MLP |
PCNSL, primary central nervous system lymphoma; T1W, T1-weighted; CE-T1W, contrast-enhanced T1 weighted image; T2W, T2-weighted; T2-FLAIR, T2 weighted fluid-attenuated inversion recovery; DWI, diffusion-weighted imaging; PWI, perfusion-weighted imaging; T1-FFE, T1-weighted fast field echo; CT, computed tomography; 18F-FDG-PET/CT, fluorine-18 fluorodeoxyglucose positron emission tomography/computed tomography; ML, machine learning; DL, deep learning; LR, logistic regression; RR, ridge regression; ENR, elastic net regression; LASSO, least absolute shrinkage and selection operator; NN, neural network; SVM, support vector machine; MLP, multilayer perceptron; RF, random forest; GBRM, generalized boosted regression model; AdaBoost, adaptive boosting; k-NN, k-nearest neighbor; NB, naïve bayes; DT, decision tree; LDA, linear discriminant analysis; RBF, radial basis function; GNB, gaussian naïve bayes; CNN, convolutional neural network.