Table 2.
Radiomics predicts tumor response and adverse symptom after treatment
| Study | Number of patients | Tumor characteristics | Imaging modality | Therapy/Symptom | Outcome, model | Machine learning algorithm |
|---|---|---|---|---|---|---|
| Yu et al. 2019 81 | Train: 51pts; Validation: 19pts | NPC | T2, CE-T1 | ART | Replan status of patient. Multivariable logistics model | LASSO, logistic regression |
| Zhang et al. 2020 80 | Total: 242pts | NPC | T2, CE-T1 | Early detection of radiation- induced brain injury | Radiation induced temporal lobe injury. Random forest | RF |
| Pota et al. 2017 78 | Total: 37pts (74 parotid glands) | HNC | Before RT (CT1), at the middle of treatment (CT2) and after RT (CT3) | Xerostomia, shrinkage of parotid glands | Parotid shrinkage rate and 12-months xerostomia. Machine learning classifiers | Naïve bayes, LFA |
| Jin et al. 2019 77 | Train: 70pts Validation: 24pts |
EC | Before treatment (CT1), 3 months after CRT(CT2) | Concurrent chemoradiation | 3 months after CRT, Machine learning classifiers | SVM, RBF, XGBoost, PCA |
| Liu et al. 2019 79 | Total: 35pts | NPC | Five CT sets acquired at treatment position during the RT | Acute xerostomia | Patients' saliva was collected every other 10 days during the RT. Multivariate Cox regression | Multivariate machine learning algorithms |
| Bologna et al. 2020 74 | Total: 50pts (25 responders and 25 non-responders) | Sinonasal cancers | T1, T2, ADC | Induction chemotherapy | T1+T2+ADC model displayed the highest; radiomics score | PCA, naïve bayes, SVM, KNN and so on |
| Wang et al. 2018 75 | Total: 120pts | NPC | T2, T2FS, CE- T1 | Induction chemotherapy | Early response to induction chemotherapy; logistic regression | Logistic regression |
| Zhao et al. 2019 76 | Train: 100pts Validation: 23pts |
NPC | T1, T2, CE- T1 | Induction chemotherapy | PFS; multivariable logistic regression | LASSO, SVM, logistic regression |
ART: Adaptive radiotherapy; LFA: Likelihood-fuzzy analysis; RBF: Radial basis function; XGBoost: Extreme gradient boosting algorithm; CRT: Concurrent chemoradiation; T2FS: T2 weighted fat-suppressed; RT: Radiation therapy; EBV: Epstein-Barr virus; ADC: Apparent diffusion coefficient.