Table 1.
Details of the included studies (n=42).
| Study | Country, year | Data type; number of data | Population | Cancer symptoms | Significant predictors | Algorithms | Validation methods |
| Sun et al [20] | China, 2023 | Clinical data; 1152 | People with breast cancer | Pain | Postmenopausal status, urban medical insurance, history of at least 1one operation, underwent general anesthesia with fentanyl and sevoflurane, and received axillary lymph node dissection. | LRa,b, RFc, GBDTd, and XGBe | Random |
| Xinran et al [21] | China, 2023 | Clinical data; 494 | People with advanced cancer | Cognitive impairment | Cancer course, anxiety, and age | LR and ANNf | Random |
| Shaikh et al [22] | United States, 2023 | Clinical data; 1152 | Survivors of cancer with osteoarthritis | Depression | Age, education, care fragmentation, polypharmacy, and zip code–level poverty | XGB | 10-fold CVg |
| Kober et al [23] | United States, 2023 | Clinical data; 1217 | People with cancer receiving chemotherapy | Morning fatigue | 13 individual Li-Fraumeni syndrome items | ENh, RF, LASSOi, LR (filtered/unfiltered), RPARj, and SVMk | Random |
| Du et al [24] | China, 2023 | Clinical data; 565 | People with cancer | Fatigue | Pain score, Eastern Cooperative Oncology Group score, platelet distribution width, and continuous erythropoiesis receptor activator | LR, RF, NBl, and XGB | 5-fold CV |
| Moscato et al [25] | Italy, 2022 | Clinical data; 21 | People with cancer | Pain | N/Am | SVM, RF, MPn, LR, and AdaBoosto | 10-fold CV |
| Masukawa et al [26] | Japan, 2022 | Clinical data; 808 | People with cancer | Social distress, spiritual pain, pain, dyspnea, nausea, and insomnia | N/A | LR, RF, light GBMp, SVM, and ensemble | 5-fold CV |
| Fanizzi et al [27] | Italy, 2022 | CTq image data; 61 | People with oropharyngeal cancer receiving radiotherapy | Xerostomia | Weight preradiotherapy, induction chemotherapy, sex, platinum-based chemotherapy, current chemotherapy, alcohol history, age at diagnosis, smoking history, surgery, clinical tumor, and clinical node | SVM and CNNr | 10-fold CV |
| Ueno et al [28] | Japan, 2022 | Clinical data; 284 | People with breast cancer | Insomnia | General fatigue, physical fatigue, and cognitive fatigue | L2 penalized LR and XGB | 8-fold CV |
| On et al [29] | Korea, 2022 | Clinical data; 935 | People with cancer receiving chemotherapy | Nausea-vomiting, fatigue-anorexia, diarrhea, hypersensitivity, stomatitis, hand-foot syndrome, peripheral neuropathy, and constipation | Earlier history of adverse drug reaction, comorbidity, cancer site and type of chemotherapy, demographics, and antineoplastic therapy–related features | LR, DTs, and ANN | 3-fold CV |
| Li et al [30] | China, 2022 | Clinical data and CT image data; 365 | People with cancer receiving radiotherapy | Xerostomia | Hypertension, age, total radiotherapy dose, dose at 50% of the left parotid volume, mean dose to right parotid gland, mean dose to oral cavity, and course of induction chemotherapy | RF, DT and XGB | External validation |
| Kurisu et al [31] | Japan, 2022 | Clinical data; 668 | People with advanced cancer receiving pharmacological interventions | Delirium | The baseline Delirium Rating Scale-R98 severity score (cutoff of 15), hypoxia, and dehydration | DT | 5-fold CV |
| Guo et al [32] | China, 2022 | Clinical data; 80 | People with lung cancer receiving chemotherapy | Lung infection | Age ≥60 years, length of stay ≥14 days, surgery history, combined chemotherapy, myelosuppression, diabetes, and hormone application | LR and ANN | Random |
| Baglione et al [33] | United States, 2022 | Clinical data; 40 | People with breast cancer | Depressed mood and anxiety | Connectedness, receive support, frequency and duration use of mobile app, and physical pain | RF and XGB | LOOCVt |
| Chao et al [34] | United States, 2022 | Clinical data and CT image data; 155 | People with HNCu receiving radiotherapy | Xerostomia | N/A | SVM, KNNv, NB, and RF | Nested |
| Wakabayashi et al [35] | Japan, 2021 | Clinical data and CT image data; 69 | People with cancer receiving radiotherapy | Pain | Age, numeric rating scale, and biological effective dose 10 | RF | LOOCV |
| Zhou et al [36] | China, 2021 | Clinical data; 386 | People with colorectal cancer after chemotherapy | Cognitive impairment | Age, BMI, colostomy, treatment complications, cancer-related anemia, depression, diabetes, Quality of Life Questionnaire Core 30 score, exercise, hypercholesterolemia, diet, marital status, education level, and pathological stage | RF, LR, and SVM | Random |
| Xuyi et al [37] | Canada, 2021 | Clinical data; 46,104 | Specific cancer site or treatment not mentioned | Pain, depression, and well-being | Lung cancer, late-stage cancer, existing chronic conditions such as osteoarthritis, mood disorder, hypertension, diabetes, and coronary disease | ANN | Random |
| Xu et al [38] | China, 2021 | Clinical data; 598 | People with gastrointestinal tumors after surgery | Postoperative fatigue | Age, higher degree of education, lower personal monthly income, advanced cancer, hypoproteinemia, preoperative anxiety or depression, and limited social support | LR, ANN, CARTw | Random |
| Wei et al [39] | China, 2021 | Clinical data; 533 | People with breast cancer | Lymphedema | N/A | ANN, LR, C5.0, RF, SVM, CART | 10-fold CV |
| Wang et al [40] | United States, 2021 | Clinical data; 823 | People with HNC | Pain, taste, and general activity | N/A | SVM, KNN, and RF; Gaussian NB and MLPx; and ARIMAy and LSTMz | Random |
| Wang et al [41] | United States, 2021 | Clinical data and CT image data; 138 | Specific cancer site or treatment not mentioned | Depression | N/A | Fine tree, medium tree, coarse tree, linear-discriminant, quadratic discriminant, LR, Gaussian NB, kernel NB, linear SVM, quadratic SVM, cubic SVM, Fine Gaussian SVM, Medium Gaussian SVM, Coarse Gaussian SVM, Fine KNN, Medium KNN, Coarse KNN, Cosine KNN, Cubic KNN, Weighted KNN, boosted trees, bagged trees, subspace discriminant, subspace KNN, and random undersampling boosted trees | 5-fold CV |
| Mosa et al [17] | United States, 2021 | Clinical data; 6124 | People with cancer receiving chemotherapy | Nausea-vomiting | Smoking, alcohol status, sex, age, and BMI | NB, LR, ANN, SVRaa, and DT | 10-fold CV |
| Low et al [42] | United States, 2021 | Clinical data; 44 | People with pancreatic cancer after surgery | Diarrhea, fatigue, and pain | Physical activity bouts, sleep, heart rate, and location | LR, KNN, SVM, RF, GBab, XGB, and LightGBM | 3-fold CV and LOOCV |
| Kourou et al [43] | Greece, 2021 | Clinical data; 609 | People with breast cancer | Depression | A set of psychological traits (optimism, perceived ability to cope with trauma, resilience as a trait, and ability to understand the illness) and subjective perceptions of personal functionality (physical, social, and cognitive) | RF, SVM, and GB | 5-fold CV |
| Kober et al [44] | United States, 2021 | Clinical data; 1217 | People with cancer receiving chemotherapy | Evening fatigue | Morning fatigue, lower evening energy, and sleep disturbance | RF, LR (filtered or unfiltered), RPAR, and SVM | 10-fold CV |
| Hu et al [45] | China, 2021 | Clinical data; 238 | People with non-Hodgkin lymphoma receiving chemotherapy | Depression | Education level, sex, age, marital status, medical insurance, per capita monthly household income, pathological stage, Suicide Severity Rating Scale, Pittsburgh Sleep Quality Index, and Quality of Life Questionnaire Core 30 | SVM, RF, and LASSO+LR | Random |
| Haun et al [46] | Germany, 2021 | Clinical data; 496 | People with cancer seen in primary care | Anxiety | Fatigue or weakness, insomnia, and pain appeared | OLSac, RRad, LASSO, ENRae, RF, and XGB | 10-fold CV |
| Lee et al [47] | United States, 2020 | Clinical data and CT Images data; 388 | People with lung cancer after intensity-modulated radiation therapy | Weight loss | Joint Gross tumor volume L1+L2+L3 radiomics, Gross tumor volume, and esophagus L3 dosiomic | SVM, DNNaf, and ensemble classifier | Nested CV |
| Juwara et al [48] | Canada, 2020 | Clinical data; 204 | People with breast cancer after surgery | NPaj | Anxiety, type of surgery, and acute pain | LSah, RR, ENR, RF, GB, and ANN | 10-fold CV |
| Men et al [49] | United States, 2019 | Clinical data and CT image data; 784 | People with HNC receiving radiotherapy | Xerostomia | Feature map visualization | LR and 3D-RCNNai | Random |
| Jiang et al [50] | United States, 2019 | Clinical data and CT images data; 427 | People with HNC | Xerostomia | The patient has human papillomavirus, completed chemotherapy, their baseline xerostomia grade, tumor site, N stage, and use of feeding tube | RR, LASSO, and RF | 10-fold CV |
| Sheikh et al [51] | United States, 2019 | CT images data; 266 | People with HNC | Xerostomia | N/A | Generalized linear model | 10-fold CV |
| Papachristou et al [52] | United States, 2019 | Clinical data; 799 | People with cancer receiving chemotherapy | Sleep disturbance, anxiety, and depression | Age, gender, cancer site, the number of prior cancer treatment, and initial diagnosis | SVR (linear, polynomial, and radial Sigma) and n-CCAaj | 10-fold CV and bootstrap |
| Zhang et al [53] | China, 2018 | Clinical data; 375 | People with cancer receiving radiotherapy | Weight loss | Head and neck tumor location and total radiation dose of ≥70 Gray, and without postsurgery | DT and LR | Random |
| Olling et al [54] | Denmark;2018 | Clinical and CT image; 131 | People with lung cancer receiving radiotherapy | Odynophagia (painful swallowing) | N/A | Multivariable LR, Lasso and elastic net regularized generalized linear models, and SVM | 10-fold CV |
| Gabryś et al [55] | Germany;2018 | Clinical and CT image; 153 | People with HNC after radiotherapy | Xerostomia | The parotid gland volume, the spread of the contralateral dose-volume histogram, and the parotid gland eccentricity, and sex | LRL1ak, LRL2al, LR-ENam, KNN, SVM, ETan, and GTBao | Single and nested CV |
| Lötsch et al [56] | Germany;2018 | Clinical data; 1000 | People with breast cancer after surgery | Pain | Age, chronic pain of any type, number of previous operations, BMI, preoperative pain in the area to be operated on, smoking and psychological factors | Unsupervised MLap | Random |
| Abdollahi et al [57] | Iran;2018 | Clinical and CT image; 47 | People with HNC receiving chemotherapy | Hearing loss | 10 of the 490 radiomic features selected as the associated features with significant sensorineural hearing loss status | Decision stump, Hoeffding, C4.5, NB, AdaBoost, bootstrap aggregating, and LR | 10-fold CV |
| van Dijk et al [58] | United States;2018 | Clinical data and CT image; 68 | People with HNC | Xerostomia | N/A | LR | External validation |
| Cvetković [59] | Serbia;2017 | Clinical data; 84 | People with breast cancer | Depression | N/A | ELMaq, ANN, and Fuzzy Genetic Algorithm | Random |
| van Dijk et al [60] | United States;2017 | CT image features; 249 | People with HNC | Xerostomia | N/A | LR | 10-fold CV |
aLR: logistic regression.
bItalic text in this column indicates the best results used in the study.
cRF: random forest.
dGBDT: gradient boosting decision tree.
eXGB: extreme gradient boosting.
fANN: artificial neural network.
gCV: cross-validation.
hEN: elastic net.
iLASSO: Least absolute shrinkage and selection operator.
jRPAR: recursive partitioning and regression trees.
kSVM: support vector machine.
lNB: Naïve bayes.
mN/A: not applicable.
nMP: multiple perceptron.
oAdaBoost: Adaptive boosting.
pGBM: light gradient boosting machine.
qCT: computed tomography.
rCNN: convolutional neural network.
sDT: decision tree.
tLOOCV: leave-one-out-cross-validation.
uHNC: head and neck cancer.
vKNN: k-nearest neighbor.
wCART: classification and regression tree.
xMLP: multilayer perceptron.
yARIMA: autoregressive integrated moving average.
zLSTM: long short-term memory neural network.
aaSVR: support vector regression.
abGB: gradient boosting.
acOLS: ordinary least square.
adRR: ridge regression.
aeENR: elastic net regression.
afDNN: deep neural network.
agNP: neuropathic pain.
ahLS: least squares.
ai3D-RCNN: 3D region-based convolutional neural network.
ajn-CCA: nonlinear canonical correlation analysis.
akLRL1: L1 penalized logistic regression.
alLRL2: L2 penalized logistic regression.
amLR-EN: logistic regression-elastic net.
anET: extra tree.
aoGTB: gradient tree boosting.
apML: machine learning.
aqELM: extreme linear machine.