Table 1.
Characteristics of articles included in final analysis. ACS-NSQIP: American College of Surgeons National Surgical Quality Improvement Program, NIS: National Inpatient Sample, ASES: American Shoulder and Elbow Surgeons score
| Title | Author, year | Study design | Database | Sample size (N) | Sex (%male) | Mean age | Machine learning algorithm | Training/test split | Minors |
|---|---|---|---|---|---|---|---|---|---|
| A novel machine learning model developed to assist in patient selection for outpatient total shoulder arthroplasty | Biron et al., (2020) [26] | Retrospective Study | ACS-NSQIP | 3,128 | 44.90% | 69.4 | Random Forest | 70:30 | 14 |
| Construct validation of machine learning in the prediction of short-term postoperative complications following total shoulder arthroplasty | Gowd et al., (2019) [27] | Retrospective Study | ACS-NSQIP | 17,119 | 56.20% | 69.5 |
-Logistic regression -K-nearest neighbor -Random forest -Naive-Bayes -Decision tree -Gradient boosting trees |
80:20 | 16 |
| The value of artificial neural networks for predicting length of stay, discharge disposition, and inpatient costs after anatomic and reverse shoulder arthroplasty | Karnuta et al., (2020) [28] | Retrospective Study | NIS | 111,147 | 40.80% | 69 | Artificial Neural Network | 70% for training, 10% for validation, 20% for testing | 14 |
| What Is the Accuracy of Three Different Machine Learning Techniques to Predict Clinical Outcomes After Shoulder Arthroplasty? | Kumar et al., (2020) [22] | Retrospective Study | MultiCenter | 4,782 | 39.90% | 69.6 |
-Linear regression -XGBoost -Wide and Deep |
66.7:33.3 | 15 |
| Using machine learning to predict clinical outcomes after shoulder arthroplasty with a minimal feature set | Kumar et al., (2021) [31] | Retrospective Study | MultiCenter | 5,774 | 39.30% | 70.1 | XGBoost | 66.7:33.3 | 14 |
| Use of machine learning to assess the predictive value of 3 commonly used clinical measures to quantify outcomes after total shoulder arthroplasty | Kumar et al., (2021) [32] | Retrospective Study | MultiCenter | 2,790 | 59.10% | N/A | XGBoost | 66.7:33.3 | 15 |
| Using machine learning to predict internal rotation after anatomic and reverse total shoulder arthroplasty | Kumar et al., (2022) [30] | Retrospective Study | MultiCenter | 6,468 | 38.80% | 48.7 |
-Linear regression -XGBoost -Wide and Deep |
66.7:33.3 | 15 |
| Development of a predictive model for a machine learning–derived shoulder arthroplasty clinical outcome score | Kumar et al., (2022) [29] | Retrospective Study | MultiCenter | 6,468 | 38.80% | 48.7 |
-Linear regression -XGBoost -Wide and Deep |
66.7:33.3 | 14 |
| Using machine learning methods to predict nonhome discharge after elective total shoulder arthroplasty | Lopez et al., (2021) [33] | Retrospective Study | ACS-NSQIP | 21,544 | 44.70% | 69.1 |
-Boosted Decision Tree -Artificial Neural Network |
80:20 | 14 |
| Using machine learning methods to predict prolonged operative time in elective total shoulder arthroplasty | Lopez et al., (2022) [34] | Retrospective Study | ACS-NSQIP | 21,544 | 44.70% | 69.1 |
-Boosted Decision Tree -Artificial Neural Network |
80:20 | 14 |
| Machine Learning Can Predict Level of Improvement in Shoulder Arthroplasty | McLendon et al., (2021) [35] | Retrospective Study | Single Institution | 472 | 56% | 68 | N/A | N/A | 14 |
| Development of supervised machine learning algorithms for prediction of satisfaction at 2 years following total shoulder arthroplasty | Polce et al., (2020) [36] | Retrospective Study | Single Institution | 413 | 58.60% | 66 |
-Stochastic gradient boosting -Random forest -Support vector machine -Neural network -Elastic-net penalized logistic regression |
80:20 | 13 |
ACS-NSQIP American College of Surgeons National Surgical Quality Improvement Program, NIS National Inpatient Sample, ASES American Shoulder and Elbow Surgeons score