Table 4 A summary on the details of machine learning, strengths and limitations of the study on prediction of periprosthetic joint infection
First author [Ref.] | Machine learning application | Algorithm | Statistical performance | Strengths | Limitations |
|---|---|---|---|---|---|
Yeo, I. (2022) [7] | Preoperative prediction of the risk of periprosthetic joint infection following primary total knee arthroplasty | Artificial neural network | Area under the receiver operating characteristic curve: 0.84. Accuracy: >94.0%. Brier score: 0.054. Calibration intercept: 0.09. Calibration slope: 1.06 | Important variables for the risk prediction were identified and expressed in a bar graph which showed the relative importance value | No external validation |