Artificial intelligence in clinical care amidst COVID-19 pandemic: A systematic review

Abstract

The worldwide health crisis caused by the SARS-Cov-2 virus has resulted in>3 million deaths so far. Improving early screening, diagnosis and prognosis of the disease are critical steps in assisting healthcare professionals to save lives during this pandemic. Since WHO declared the COVID-19 outbreak as a pandemic, several studies have been conducted using Artificial Intelligence techniques to optimize these steps on clinical settings in terms of quality, accuracy and most importantly time. The objective of this study is to conduct a systematic literature review on published and preprint reports of Artificial Intelligence models developed and validated for screening, diagnosis and prognosis of the coronavirus disease 2019. We included 101 studies, published from January 1st, 2020 to December 30th, 2020, that developed AI prediction models which can be applied in the clinical setting. We identified in total 14 models for screening, 38 diagnostic models for detecting COVID-19 and 50 prognostic models for predicting ICU need, ventilator need, mortality risk, severity assessment or hospital length stay. Moreover, 43 studies were based on medical imaging and 58 studies on the use of clinical parameters, laboratory results or demographic features. Several heterogeneous predictors derived from multimodal data were identified. Analysis of these multimodal data, captured from various sources, in terms of prominence for each category of the included studies, was performed. Finally, Risk of Bias (RoB) analysis was also conducted to examine the applicability of the included studies in the clinical setting and assist healthcare providers, guideline developers, and policymakers.

Keywords: ABG, Arterial Blood Gas; ADA, Adenosine Deaminase; AI, Artificial Intelligence; ANN, Artificial Neural Networks; APTT, Activated Partial Thromboplastin Time; ARMED, Attribute Reduction with Multi-objective Decomposition Ensemble optimizer; AUC, Area Under the Curve; Acc, Accuracy; Adaboost, Adaptive Boosting; Apol AI, Apolipoprotein AI; Apol B, Apolipoprotein B; Artificial intelligence; BNB, Bernoulli Naïve Bayes; BUN, Blood Urea Nitrogen; CI, Confidence Interval; CK-MB, Creatine Kinase isoenzyme; CNN, Convolutional Neural Networks; COVID-19; CPP, COVID-19 Positive Patients; CRP, C-Reactive Protein; CRT, Classification and Regression Decision Tree; CoxPH, Cox Proportional Hazards; DCNN, Deep Convolutional Neural Networks; DL, Deep Learning; DLC, Density Lipoprotein Cholesterol; DNN, Deep Neural Networks; DT, Decision Tree; Diagnosis; ED, Emergency Department; ESR, Erythrocyte Sedimentation Rate; ET, Extra Trees; FCV, Fold Cross Validation; FL, Federated Learning; FiO2, Fraction of Inspiration O2; GBDT, Gradient Boost Decision Tree; GBM light, Gradient Boosting Machine light; GDCNN, Genetic Deep Learning Convolutional Neural Network; GFR, Glomerular Filtration Rate; GFS, Gradient boosted feature selection; GGT, Glutamyl Transpeptidase; GNB, Gaussian Naïve Bayes; HDLC, High Density Lipoprotein Cholesterol; INR, International Normalized Ratio; Inception Resnet, Inception Residual Neural Network; L1LR, L1 Regularized Logistic Regression; LASSO, Least Absolute Shrinkage and Selection Operator; LDA, Linear Discriminant Analysis; LDH, Lactate Dehydrogenase; LDLC, Low Density Lipoprotein Cholesterol; LR, Logistic Regression; LSTM, Long-Short Term Memory; MCHC, Mean Corpuscular Hemoglobin Concentration; MCV, Mean corpuscular volume; ML, Machine Learning; MLP, MultiLayer Perceptron; MPV, Mean Platelet Volume; MRMR, Maximum Relevance Minimum Redundancy; Multimodal data; NB, Naïve Bayes; NLP, Natural Language Processing; NPV, Negative Predictive Values; Nadam optimizer, Nesterov Accelerated Adaptive Moment optimizer; OB, Occult Blood test; PCT, Thrombocytocrit; PPV, Positive Predictive Values; PWD, Platelet Distribution Width; PaO2, Arterial Oxygen Tension; Paco2, Arterial Carbondioxide Tension; Prognosis; RBC, Red Blood Cell; RBF, Radial Basis Function; RBP, Retinol Binding Protein; RDW, Red blood cell Distribution Width; RF, Random Forest; RFE, Recursive Feature Elimination; RSV, Respiratory Syncytial Virus; SEN, Sensitivity; SG, Specific Gravity; SMOTE, Synthetic Minority Oversampling Technique; SPE, Specificity; SRLSR, Sparse Rescaled Linear Square Regression; SVM, Support Vector Machine; SaO2, Arterial Oxygen saturation; Screening; TBA, Total Bile Acid; TTS, Training Test Split; WBC, White Blood Cell count; XGB, eXtreme Gradient Boost; k-NN, K-Nearest Neighbor.

Graphical abstract

Fig. 1

AI-based clinical prediction models.

Fig. 2

PRISMA (preferred reporting items for systematic reviews and meta-analyses) flowchart.

Fig. 3

Number of studies per year 2020 quarter.

Fig. 4

Included AI based prediction models.