Machine Learning Predicts Implant-Based Reconstruction Complications

WEDNESDAY, Nov. 29, 2023 (HealthDay News) -- Machine learning (ML) algorithms can accurately predict periprosthetic infection and explantation following implant-based reconstruction (IBR), according to a study published in the November issue of Plastic and Reconstructive Surgery.

Abbas M. Hassan, M.D., from The University of Texas MD Anderson Cancer Center in Houston, and colleagues conducted a comprehensive review of patients who underwent IBR from January 2018 to December 2019 to develop, validate, and assess the use of ML algorithms to predict IBR complications using readily available perioperative clinical data. Nine supervised ML algorithms were developed; patient data were classified into training and testing sets (80 and 20 percent, respectively).

Data were included for 481 patients who were followed for a mean of 16.1 months. The researchers found that 113 of the reconstructions (16.3 percent) resulted in periprosthetic infection, and explantation was required with 82 (11.8 percent). Good discriminatory performance was seen for predicting periprosthetic infection and explantation with ML (area under the receiver operating characteristic curve, 0.73 and 0.78, respectively); nine and 12 significant predictors of periprosthetic infection and explantation were identified, respectively.

"Our study provides proof of the feasibility, effectiveness, and applicability of artificial intelligence in predicting complications of IBR and should encourage the incorporation of ML in the perioperative assessment of patients undergoing IBR to provide data-driven, patient-specific risk assessment to aid in individualized patient counseling, shared decision-making, and presurgical optimization," the authors write.

Abstract/Full Text