Table 1.
Table summarizing the main studies of Radiomics applied to lung cancer.
| Study | N Patients | Objective | Types of Evaluation | Model Performance | Imaging Modality | Features Selected | Nature of Study |
|---|---|---|---|---|---|---|---|
| Beig N. et al., Radiol. 2019 [7] |
Total 290 | AD vs. Granulomas | Intranodular and Perinodular radiomic analysis |
0.80 0.76 0.60–0.61 |
CT | 12 | Multicentric Restrospective |
| Linning E. et al., Acad. Radiol. 2019 [8] | Total 278 | SCLC vs. NSCLC SCLC vs. AD SCLC vs. SCC |
Primary lesion radiomic analysis |
End. 1 AUC: 0.74 End. 2 AUC: 0.82 End. 3 AUC: 0.66 |
CT | 20 | Monocentric Restrospective |
| Cong M. et al., Lung Cancer 2020 [9] |
Training 455 Validation 194 |
Assessment nodal metastases | Predictive Radiomics on primary lesion | AUC: 0.91 AUC: 0.86 |
CT | 7 | Monocentric Restrospective |
| Zhang J. et al., Eur. J. Nucl. Med Mol. Imaging 2020 [10] |
Training 175 Validation 73 |
EGFR status | Radiomic signature Fusion models |
AUC: 0.86 AUC: 0.87 |
18F-FDG PET/CT | 10 | Monocentric Restrospective |
| Zerunian M. et al., Sci. Rep 2021 [11] |
Total 21 | OS PFS |
Volumetric Textural analysis |
End. 1 AUC: 0.72 End. 2 AUC: 0.74 |
CT | 6 | Monocentric Prospective |
| Khorrami M. et al., Lung Cancer 2019 [12] |
Training 45 Validation 45 |
Pathological response | Intranodular and perinodular radiomic analysis |
AUC: 0.90 AUC: 0.86 |
CT | 13 | Monocentric Restrospective |
AD, Adenocarinoma; SCLC, small cell lung cancer; NSCLC, non-small lung cancer, SCC, squamous cell carcinomas; EGFR, epidermal growth factor receptor; OS, overall survival; PFS, progression free survival; SVM, support vector machine; ROC, receiver operating characteristic; CNN, convolutional neural networks; AUC, area under the curve; CT, computed tomography.