Introduction
Literature search and screening
Clinical application of 18F-FDG PET/CT radiomics-based ML analyses in lung or mediastinal tumors
Difference between benign and malignant tumors and between primary and metastatic tumors
Authors | Year | Tumor type | Aim | Sample size | Constructed ML models | Core ML algorithm | Best ML model | Validation | Resulta |
---|---|---|---|---|---|---|---|---|---|
Differentiating benign from malignant tumors or primary from metastatic tumors | |||||||||
Ren et al. [14] | 2022 | SPN | Benign vs. malignant | n = 280 | Clinical model PET radiomics-based model Combined model | LASSO regression | Combined model | Training and validation cohorts | AUC: 0.94 |
Zhou et al. [15] | 2021 | SPN | Primary vs. metastatic | n = 769 | CT radiomics-based model PET radiomics-based model | GBDT | PET radiomics-based model | Training and validation cohorts | AUC: 0.983 |
Salihoğlu et al. [16] | 2022 | SPN | Benign vs. malignant | n = 48 | PET radiomics-based model alone | Deep neural network | – | Internal validation (cross-validation) | AUC: 0.81 |
Zhang et al. [17] | 2019 | SPN | Benign vs. malignant | n = 135 | CT radiomics-based model PET radiomics-based model Combined model | SVM | Combined model | Internal validation (cross-validation) | AUC:0.887 |
Yan et al. [18] | 2020 | SPN | Primary vs. metastatic | n = 445 | CT radiomics-based model PET radiomics-based model Combined model | SMO | Combined model | Internal validation (cross-validation) | AUC: 0.98 |
Agüloğlu et al. [19] | 2023 | Consolidated lesion | Lung cancer vs. infection | n = 106 | PET radiomics-based model only | LR | – | Training and validation cohorts | AUC: 0.813 |
Classifying tumors according to histological subtypes | |||||||||
Zhao et al. [22] | 2022 | NSCLC | ADC vs. SCC | n = 120 | Clinical model PET radiomics-based model Combined model | SVM | Combined model | Training and validation cohorts | AUC: 0.876 |
Han et al. [23] | 2021 | NSCLC | ADC vs. SCC | n = 1419 | PET radiomics-based model only | VGG16 DL | – | Training and validation cohorts | AUC: 0.903 |
Ren et al. [24] | 2021 | NSCLC | ADC vs. SCC | n = 315 | Clinical laboratory model CT radiomics-based model PET radiomics-based model Combination of all models | LASSO regression | Combined model | Training and validation cohorts | AUC: 0.901 |
Koyasu et al. [25] | 2020 | NSCLC | ADC vs. SCC | n = 188 | Combined CT + PET radiomics-based model alone | XGB | – | Internal validation (cross-validation) | AUC: 0.843 |
Hyun et al. [26] | 2019 | NSCLC | ADC vs. SCC | n = 396 | Combined clinical + PET radiomics-based model alone | LR | – | Internal validation (cross-validation) | AUC: 0.859 |
Nakajo et al. [27] | 2022 | TET | Thymic carcinoma vs thymoma | n = 79 | Combined PET radiomics- + CNN-based feature model | LR | – | Internal validation (cross-validation) | AUC: 0.90 |
Ozkan et al. [28] | 2022 | TET | Low-risk thymoma vs. high-risk thymoma | n = 27 | Combined clinical + PET radiomics-based model alone | LASSO + artificial neural network | – | Training and validation cohorts | AUC: 0.88 |
Predicting tumor characteristics | |||||||||
Gao et al. [33] | 2023 | Lung ADC | EGFR status | n = 515 | Clinical model CT radiomics-based model PET radiomics-based model Combined models | RF | Combined model | Training and validation cohorts | AUC: 0.730 |
Chang et al. [34] | 2021 | Lung ADC | ALK status | n = 526 | CT radiomics-based model PET radiomics-based model Combined PET and CT radiomics-based model Combined clinical, PET, and CT models | LASSO regression | Combined clinical, PET and CT model | Training and validation cohorts | AUC: 0.88 |
Shiri et al. [35] | 2020 | NSCLC | EGFR and KRAS status | n = 150 | Combined CT + PET radiomics-based model alone | Stochastic gradient descent | – | Training and validation cohorts | AUC for EGFR: 0.82 AUC for KRAS: 0.83 |
Liu et al. [36] | 2020 | Lung ADC | EGFR status | n = 148 | Combined CT + PET radiomics-based model alone | XGB | – | Training and validation cohorts | AUC: 0.870 |
Agüloğlu et al. [37] | 2022 | NSCLC | EGFR and ALK status | n = 189 | PET radiomics-based model alone | Naïve Bayes algorithm | – | Training and validation cohorts | AUC for EGFR: 0.797 AUC for ALK: 0.814 |
Nair et al. [38] | 2021 | NSCLC | EGFR status | n = 50 | CT radiomics-based model PET radiomics-based model | LR | PET-radiomics model | Internal validation (cross-validation) | AUC: 0.870 |
Li. et al. [39] | 2019 | NSCLC | EGFR status | n = 115 | CT radiomics-based model PET radiomics-based model Combined model | LASSO regression | Combined model | Internal validation (cross-validation) | AUC: 0.822 |
Lim et al. [40] | 2022 | NSCLC | PD-L1 expression | n = 312 | Combined model only (CT + PET radiomics feature) | Naïve Bayes algorithm | – | Internal validation (cross-validation) | AUC: 0.712 |
Mu et al. [41] | 2021 | NSCLC | PD-L1 expression | n = 697 | Combined CT + PET radiomics-based model alone | SRecCNN | – | Training and validation cohorts | AUC: 0.82 |
Tong et al. [42] | 2022 | NSCLC | CD8 expression | n = 1367 | CT radiomics-based model PET radiomics-based model Combined PET and CT scan model Combined clinical, PET, and CT scan model | LR | Combined clinical, PET and CT model | Training and validation cohorts | AUC: 0.932 |
Predicting tumor stage | |||||||||
Wang et al. [44] | 2023 | NSCLC | N stage | n = 192 | Combined clinical, tumor PET, and tumor CT model Combined clinical, lymph node PET, and lymph node CT model Combination of all models | XGB | Combination of all models | Training and validation cohorts | N2 stage, AUC: 0.94 |
Laros et al. [45] | 2022 | NSCLC | LNM | n = 148 | Combined tumor and lymph node PET radiomics-based model alone | XGB | – | Training and validation cohorts | Accuracy: 0.88 |
Onozato et al. [46] | 2023 | Lung cancer | Highly invasive lung cancer | n = 873 | CT radiomics-based model PET radiomics-based model Combined model | Ensemble ML algorithm | Combined model | Training and validation cohorts | AUC: 0.880 |
Predicting treatment response or survival | |||||||||
Zhao et al. [47] | 2022 | Lung ADC | OS | n = 421 | Combined clinical + CT radiomics-based + PET radiomics-based model alone | Ensemble ML algorithm | – | Training and validation cohorts | 3-year OS, AUC: 0.84; 4-year OS, AUC: 0.88 |
Huang et al. [48] | 2022 | Malignant lung tumor | OS | n = 965 | Clinical model CT radiomics-based model PET radiomics-based model Combined PET and CT scan model Combined clinical, PET, and CT scan model | CNN + RSF | Combined clinical, PET, and CT scan models | Training and validation cohorts | C-index: 0.737 |
Ahn et al. [49] | 2019 | NSCLC | Disease recurrence after surgery | n = 93 | PET radiomics-based model alone | RF | – | Training and validation cohorts | AUC: 0.956 |
Kirienko et al.[50] | 2021 | NSCLC | Disease recurrence after surgery | n = 151 | Genomic model Combined PET and CT model Combination of all models | Logic learning machine | Combination of all models | Internal validation (cross-validation) | AUC: 0.87 |
Mu et al.[51] | 2020 | NSCLC | PFS in patients treated with EGFR-TKI | n = 616 | Combined CT + PET radiomics-based model alone | SRecCNN | – | Training and validation cohorts | HR: 0.24 |
Mu et al. [52] | 2020 | NSCLC | PFS and OS in patients treated with ICI | n = 194 | Combined CT radiomics-based + PET radiomics-based + PET/CT scan-based (minimum Kullback–Leibler divergence features) model alone | LASSO + Cox proportional hazard model | – | Training and validation cohorts | PFS, C-index: 0.77; OS, C-index: 0.80 |
Bertolini.et al. [53] | 2022 | NSCLC | 2-year PFS in patients treated with RT | n = 117 | Harmonized CT radiomics-based model Harmonized PET radiomics-based model Combined model | SVM | Combined model | Training and validation cohorts | AUC: 0.77 |
Sepehri et al. [54] | 2021 | NSCLC | OS in patients treated with CRT | n = 138 | Combined CT + PET radiomics-based model alone | Ensemble ML algorithm | – | Training and validation cohorts | Accuracy: 0.78 |
Afshar et al. [55] | 2020 | NSCLC | OS in patients treated with RT | n = 132 | Combined clinical + CT radiomics-based + PET radiomics-based model alone | CNN + Cox proportional hazard model | – | Training and validation cohorts | C-index: 0.68 |
Astaraki et al. [56] | 2019 | NSCLC | OS in patients treated with CRT | n = 30 | CT radiomics-based model PET radiomics-based model Combined model | SVM | Combined model | Internal validation (cross-validation) | AUC: 0.95 |
Park et al. [57] | 2023 | NSCLC | Disease recurrence in patients treated with surgery or RT | n = 77 | Combined clinical + PET radiomics-based model alone | Naïve Bayes algorithm | – | Training and validation cohorts | AUC: 0.816 |
Pavic et al. [58] | 2020 | MPM | PFS in patients treated with surgery | n = 72 | CT radiomics-based model PET radiomics-based model | PCA + cox proportional hazard model | PET radiomics-based model | Training and validation cohorts | C-index: 0.66 |
Classification according to histological types
Prediction of tumor characteristics
Predicting tumor stage
Predicting treatment response or survival
Summary
Clinical application of 18F-FDG PET/CT radiomics-based ML analyses in head and neck tumors
Differentiating benign and malignant tumors and predicting tumor characteristics
Predicting treatment response or survival
Authors | Years | Tumor type | Aim | Sample size | Constructed ML models | Core ML algorithm | Best ML model | Validation | Resultsa |
---|---|---|---|---|---|---|---|---|---|
Differentiating benign from malignant tumors | |||||||||
Aksu et al. [60] | 2020 | Thyroid incidentaloma | Benign vs. malignant | n = 60 | PET radiomics only | RF | – | Training and validation cohorts | AUC: 0.849 |
Predicting tumor characteristics | |||||||||
Haider et al. [62] | 2020 | OPC | HPV status | n = 435 | Tumor PET/CT Lymph node PET/CT Tumor and lymph node PET/CT | XGB | Tumor and lymph node PET/CT | Training and validation cohorts | AUC: 0.83 |
Predicting treatment response or survival | |||||||||
Haider et al. [63] | 2021 | OPC | Locoregional recurrence after RT | n = 190 | Clinical model CT radiomics-based model PET radiomics-based model Combined PET and CT model Combined clinical, PET, and CT model | RSF | Combined model | Internal validation (cross-validation) | C-index: 0.76 |
Nakajo et al. [64] | 2023 | HPC | PFS after RT, CRT, or surgery | n = 100 | Combined clinical + PET radiomics-based model alone | LR | – | Training and validation cohorts | HR: 3.22 |
Lafata. et al. [65] | 2021 | OPC | Recurrence-free survival after RT | n = 64 | Intra-treatment PET radiomics-based model | Unsupervised data clustering algorithm | – | Internal validation | HR: 2.69 |
Spielvogel et al. [66] | 2023 | HNSCC | 3-year OS | n = 127 | Combined genomic + CT radiomics-based + PET radiomics-based model alone | Ensemble ML algorithm | – | Internal validation (cross-validation) | AUC: 0.75 |
Haider et al. [67] | 2020 | OPC | OS after RT, CRT, or surgery | n = 306 | Clinical model CT radiomics-based model PET radiomics-based model Combined PET and CT model Combined clinical, PET, and CT model | RSF | Combined model | Training and validation cohorts | 5-year OS, HPV-associated oropharyngeal cancer (p = 0.02); 5-year OS, HPV-negative oropharyngeal cancer (p = 0.01) |
Zhong et al. [68] | 2021 | HPC and LC | Disease progression at 1 year after chemotherapy or RT | n = 72 | CT radiomics-based model PET radiomics-based model Combined model | RF | Combined model | Training and validation cohorts | AUC: 0.94 |
Du et al. [69] | 2019 | NPC | Local recurrence after chemotherapy or RT | n = 76 | PET radiomics-based model alone | RF | – | Internal validation (cross-validation) | AUC: 0.892 |
Peng et al. [70] | 2019 | NPC | 5-year DFS after chemotherapy or CRT | n = 707 | Combined PET radiomics-based + CNN-based model alone | LASSO regression | – | Training and validation cohorts | C-index: 0.722 |
Liu et al. [71] | 2020 | HNSCC | OS after RT | n = 171 | PET radiomics-based model alone | LASSO regression | – | Internal validation (cross-validation) | C-index: 0.77 |
Summary
Clinical application of 18F-FDG PET/CT radiomics-based ML analyses in lymphatic tumors
Differentiating benign from malignant tumors and primary from metastatic tumors or classifying tumors according to histological types
Predicting treatment response or survival
Authors | Years | Tumor type | Aim | Sample size | Constructed ML models | Core ML algorithm | Best ML model | Validation | Resultsa |
---|---|---|---|---|---|---|---|---|---|
Differentiating benign from malignant tumors and primary from metastatic tumors or classifying tumors according to pathological subtypes | |||||||||
Abenavoli et al. [76] | 2023 | Malignant lymphoma | DLBCL vs. HD | n = 117 | PET radiomics-based model alone | RF | – | Training and validation cohorts | AUC: 0.87 |
de Jesus et al. [77] | 2022 | Malignant lymphoma | DLBCL vs. FL | n = 120 | Combined CT + PET radiomics-based model alone | Gradient boosting | – | Training and validation cohorts | AUC: 0.86 |
Lovinfosse et al. [78] | 2022 | Malignant lymphoma | 1. Malignant lymphoma vs. sarcoidosis 2. DLBCL vs. HD | n = 420 | Combined clinical + PET radiomics-based model alone | RF | – | Training and validation cohorts | 1. AUC: 0.94 2. AUC: 0.95 |
Yang et al. [79] | 2023 | Cervical lymph node | Malignant lymphoma vs. metastasis | n = 165 | CNN model Combined PET radiomics-based + CNN-based model alone | SVM | Combined model | Training and validation cohorts | AUC: 0.948 |
Cui et al. [80] | 2023 | Brain tumor | Malignant lymphoma vs. metastasis | n = 51 | PET radiomics-based model alone | RF | – | Training and validation cohorts | AUC: 0.93 |
Predicting treatment response or survival | |||||||||
Frood et al. [84] | 2022 | DLBCL | Recurrence after chemotherapy | n = 229 | Combined clinical + PET radiomics-based model alone | Ridge regression | – | Training and validation cohorts | AUC: 0.73 |
Cui et al. [85] | 2022 | DLBCL | PFS after chemotherapy | n = 271 | Clinical model PET radiomics-based model Combined clinical + PET radiomics-based model alone | RF + cox proportional hazard | Combined model | Training and validation cohorts | C-index: 0.853 |
Frood et al. [86] | 2022 | HD | Recurrence after chemotherapy or RT | n = 289 | Combined clinical + PET radiomics-based model alone | Ridge regression | – | Training and validation cohorts | AUC: 0.81 |
Ritter et al. [87] | 2022 | DLBCL | Recurrence after chemotherapy | n = 85 | PET radiomics-based model alone | Ensemble ML algorithm | – | Training and validation cohorts | AUC: 0.85 |
Jiang et al. [88] | 2022 | DLBCL | OS and PFS after chemotherapy | n = 383 | Clinical model PET radiomics-based model Combined clinical + PET radiomics-based model alone | Ensemble ML algorithm | Combined model | Training and validation cohorts | PFS, C-index: 0.758, OS, C-index: 0.794, |
Jiang et al. [89] | 2022 | GI DLBCL | OS and PFS after chemotherapy | n = 140 | Clinical model Combined clinical + PET radiomics-based model alone | SVM + cox proportional hazard | Combined model | Training and validation cohorts | PFS, C-index: 0.831 OS, C-index: 0.877 |
Coskun et al. [90] | 2021 | DLBCL | Incomplete response after chemotherapy | n = 45 | PET radiomics-based model alone | LR | – | Internal validation | AUC: 0.81 |
Milgrom et al. [91] | 2019 | HD | Recurrence after chemotherapy | n = 251 | PET radiomics-based model alone | SVM with AdaBoost | – | Internal validation | AUC: 0.952 |
Summary
Clinical application of 18F-FDG PET/CT radiomics-based ML analyses in breast tumors
Differentiating benign from malignant tumors and predicting tumor characteristics or stage
Authors | Years | Tumor type | Aim | Sample size | Constructed ML models | Core ML algorithm | Best ML model | Validation | Resultsa |
---|---|---|---|---|---|---|---|---|---|
Differentiating benign from malignant tumors and predicting tumor characteristics or stage | |||||||||
Eifer et al. [92] | 2022 | Axillary LN | COVID-19 vaccine-associated lymphadenopathy vs. metastasis | n = 99 | CT radiomics-based model PET radiomics-based model Combined model | kNN | Combined model | Training and validation cohorts | AUC: 0.98 |
Chen et al. [93] | 2022 | Breast cancer | HER2 status | n = 217 | CT radiomics-based model PET radiomics-based model PET/CTconcat radiomics-based model PET/CTmean radiomics-based model | XGB | PET/CTmean radiomics model | Training and validation cohorts | AUC: 0.760 |
Song [94] | 2021 | Breast cancer | LNM | n = 100 | PET radiomics-based model alone | XGB | – | Training and validation cohorts | AUC: 0890 |
Krajnc et al. [95] | 2021 | Breast cancer | Triple negative hormone status | n = 170 | Combined clinical + CT radiomics-based + PET radiomics-based model alone | Ensemble ML algorithm | – | Internal validation (cross-validation) | AUC: 0.82 |
Ou et al. [96] | 2020 | Breast tumor | Breast cancer vs. malignant lymphoma | n = 44 | SUV model CT radiomics-based model PET radiomics-based model Combined clinical + PET radiomics-based model Combined clinical + CT radiomics-based model | LASSO + LDA | Combined clinical and PET radiomics model | Training and validation cohorts | AUC: 0.806 |
Predicting treatment response or survival | |||||||||
Li et al. [100] | 2020 | Breast cancer | pCR after NAC | n = 100 | CT radiomics-based model PET radiomics-based model Combined age + CT radiomics-based + PET radiomics-based model | RF | Combined model | Training and validation cohorts | Accuracy: 0.80 |
Gómez et al. [101] | 2022 | Metastatic breast cancer | Metabolic response after treatment | n = 48 | Combined clinical + CT radiomics-based + PET radiomics-based model alone | LASSO + SVM | – | Training and validation cohorts | AUC: 0.82 |
Predicting treatment response or survival
Summary
Clinical application of 18F-FDG PET/CT radiomics-based ML analyses in abdominal tumors
Differentiating benign from malignant tumors and predicting tumor characteristics or stage
Authors | Years | Tumor type | Aim | Sample size | Constructed ML models | Core ML algorithm | Best ML model | Validation | Resultsa |
---|---|---|---|---|---|---|---|---|---|
Differentiating benign from malignant tumors | |||||||||
Zhang et al. [102] | 2019 | Pancreatic tumor | AIP vs. PDAC | n = 251 | CT radiomics-based model PET radiomics-based model Combined model | SVM | Combined model | Internal validation (cross-validation) | Accuracy: 0.850 |
Wei et al. [103] | 2023 | Pancreatic tumor | AIP vs. PDAC | n = 112 | CT radiomics-based + PET radiomics-based model DL feature-based model Multidomain fusion model (radiomics + DL features) | VGG11 DL algorithm | Multidomain fusion model | Internal validation (cross-validation) | Accuracy: 0.901 |
Predicting tumor characteristics or stage | |||||||||
Xing et al. [104] | 2021 | PDAC | Pathological grade | n = 149 | CT radiomics-based model PET radiomics-based model Combined model | XGB | Combined model | Training and validation cohorts | AUC: 0.921 |
Jiang et al. [105] | 2022 | HCC or ICC | MVI | HCC: n = 76; ICC: n = 51 | Clinical model CT radiomics-based model PET radiomics-based model Combined optimal PET and CT radiomics-based model Combined best clinical, PET radiomics-based, or CT radiomics-based model | RF | Combined best clinical and PET feature-based model | Training and validation cohorts | AUC for HCC: 0.88 AUC for ICC: 0.90 |
Liu et al. [106] | 2021 | Gastric cancer | LNM | n = 185 | CT radiomics-based model PET radiomics-based model Combined model | Adaboost | Combined model | Training and validation cohorts | Accuracy: 0.852 |
He et al. [107] | 2021 | Colorectal cancer | LNM | n = 199 | Combined CT + PET radiomics-based model | XGB | – | Training and validation cohorts | Accuracy: 0.7636 |
Li et al. [108] | 2021 | Colorectal cancer | MSI | n = 173 | Combined clinical + CT radiomics-based + PET radiomics-based model alone | Adaboost | – | Training and validation cohorts | AUC: 0.828 |
Predicting treatment response or survival | |||||||||
Toyama et al. [109] | 2020 | Pancreatic cancer | 1-year survival after RT, CRT, or surgery | n = 161 | PET radiomics-based model alone | RF | – | Internal validation (cross-validation) | HR for GLZLM_GLNU: 2.0 |
Liu et al. [110] | 2023 | Gastric cancer | HER2 status Progression after surgery | n = 90 | Combined clinical + CT radiomics-based + PET radiomics-based model | Adaboost | – | Training and validation cohorts | Accuracy for HER2: 0.833 Accuracy for progression: 0.778 |
Lv et al. [111] | 2022 | Colorectal cancer | Recurrence-free survival after surgery | n = 196 | Clinical model CT radiomics-based model PET radiomics-based model Combined model | RSF | Combined model | Training and validation cohorts | C-index for all patients: 0.780 C-index for patients with stage III disease: 0.820 |
Shen et al. [112] | 2020 | Rectal cancer | pCR after NCRT | n = 169 | PET radiomics-based model alone | RF | – | Internal validation | Accuracy: 0.953 |
Agüloğlu et al. [113] | 2023 | Metastatic rectal cancer | 2-year OS | n = 62 | PET radiomics-based model alone | RF | – | Internal validation (cross-validation) | AUC: 0.843 |
Predicting treatment response or survival
Summary
Clinical application of 18F-FDG PET/CT radiomics ML analyses in gynecological tumors
Predicting tumor stage
Authors | Years | Tumor type | Aim | Sample size | Constructed ML models | Core ML algorithm | Best ML model | Validation | Resultsa |
---|---|---|---|---|---|---|---|---|---|
Predicting tumor stage | |||||||||
Lucia et al. [118] | 2023 | Cervical cancer | LNM | n = 178 | Clinical model PET radiomics-based model Combined clinical and PET radiomics-based model Combat PET radiomics-based model Combined clinical and combat PET radiomics-based model | Neural network | Combat PET-radiomics model | Training and validation cohorts | AUC: 0.96 |
Zhang et al. [119] | 2022 | Cervical cancer | COX-2 status N status | n = 148 | PET radiomics-based model alone | LASSO + LR | – | Training and validation cohorts | AUC for COX-2: 0.814 AUC for LNM: 0.817 |
Li et al. [120] | 2021 | Cervical cancer | LVSI | n = 112 | PET radiomics-based model alone | LASSO + LR | – | Training and validation cohorts | AUC: 0.806 |
Chong et al. [121] | 2021 | Cervical cancer | ITB | n = 76 | PET radiomics-based model alone | LASSO + SVM | – | Training and validation cohorts | AUC: 0.784 |
Predicting treatment response or survival | |||||||||
Ferreira et al. [122] | 2021 | Cervical cancer | Disease progression after CRT | n = 158 | Combined clinical + PET radiomics-based model | RF | – | Training and validation cohorts | AUC: 0.78 |
Nakajo et al. [123] | 2022 | Cervical cancer | PFS after RT, CRT, or surgery | n = 50 | Combined clinical + PET radiomics-based model | Naïve base algorithm | – | Internal validation (cross-validation) | HR: 6.89 |
Nakajo et al. [124] | 2021 | Endometrial cancer | PFS and OS after RT, CRT, or surgery | n = 53 | Combined clinical + PET radiomics-based model | kNN | – | Internal validation (cross-validation) | PFS—HR for coarseness: 0.65; OS—HR for coarseness: 0.52 |
Predicting treatment response or survival
Summary
Clinical application of 18F-FDG PET/CT radiomics-based ML analyses in other tumors
Authors | Years | Tumor type | Aim | Sample size | Constructed ML models | Core ML algorithm | Best ML model | Validation | Resultsa |
---|---|---|---|---|---|---|---|---|---|
Differentiating primary from metastatic tumors | |||||||||
Mannam et al. [125] | 2022 | MM | MM vs. skeletal metastasis | n = 40 | CT radiomics-based model PET radiomics-based model Combined model | Multilayer perceptron | Combined model | Training and validation cohorts | AUC: 0.9538 |
Predicting tumor stage, treatment response, or survival | |||||||||
Mesguich et al. [126] | 2021 | MM | Diffuse infiltration in the bone marrow | n = 30 | Combined CT + PET radiomics-based model | RF | – | Training and validation cohorts | AUC: 0.90 |
Li et al. [127] | 2019 | Acute leukemia | Diffuse infiltration in the bone marrow | n = 41 | Combined CT + PET radiomics-based model | RF | – | Training and validation cohorts | Accuracy: 0.886 |
Ni et al. [128] | 2023 | MM | PFS | n = 98 | Clinical model Combined PET and CT radiomics-based model Combined clinical, PET radiomics-based, and CT radiomics-based model | LASSO + cox regression | Combined clinical, PET radiomics-based, and CT radiomics-based model | Training and validation cohorts | C-index: 0.698 |
Feng et al. [130] | 2022 | Neuroblastoma | MKI status | n = 102 | Clinical model Combined PET and CT radiomics-based model Combined clinical, PET, and CT radiomics-based model | XGB | Combined PET and CT radiomics-based model | Training and validation cohorts | AUC: 0.951 |