TY - GEN
T1 - Gender-Specific Machine Learning Models to Predict Unplanned Return to Operating Room Following Primary Total Shoulder Arthroplasty
AU - Letter, Christina
AU - Gupta, Puneet
AU - Kim, Annie
AU - Cong, Guang Ting
AU - Liu, Hongfang
AU - Tafti, Ahmad P.
N1 - Publisher Copyright:
© 2023 IEEE.
PY - 2023
Y1 - 2023
N2 - Reoperation is one of the most significant complications following any surgical procedure. Developing machine learning strategies that can predict the need for reoperation involves building computational models to understand if patients are likely to require a second surgery after an initial surgical procedure. These models use machine learning algorithms to analyze patient data, including demographics, medical history, lab results, and prior complications, to identify those at higher risk for complications or additional procedures. Such models offer potentials to healthcare providers to take proactive measures in reducing reoperation risks and improve patient outcomes. While several machine learning models have been developed to predict the need for a return to operating room (OR) within 30-days following primary total shoulder arthroplasty (TSA), implementation of gender-specific models has been very limited so far. This study aimed to build, train, and evaluate several return to OR predictive models trained on each gender separately and compare their accuracy performance with the traditional models trained on all genders together.
AB - Reoperation is one of the most significant complications following any surgical procedure. Developing machine learning strategies that can predict the need for reoperation involves building computational models to understand if patients are likely to require a second surgery after an initial surgical procedure. These models use machine learning algorithms to analyze patient data, including demographics, medical history, lab results, and prior complications, to identify those at higher risk for complications or additional procedures. Such models offer potentials to healthcare providers to take proactive measures in reducing reoperation risks and improve patient outcomes. While several machine learning models have been developed to predict the need for a return to operating room (OR) within 30-days following primary total shoulder arthroplasty (TSA), implementation of gender-specific models has been very limited so far. This study aimed to build, train, and evaluate several return to OR predictive models trained on each gender separately and compare their accuracy performance with the traditional models trained on all genders together.
KW - Gender-Specific Models
KW - ML Fairness
KW - Reoperation Risk Analysis
KW - Total Shoulder Arthroplasty
UR - http://www.scopus.com/inward/record.url?scp=85181578634&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85181578634&partnerID=8YFLogxK
U2 - 10.1109/ICHI57859.2023.00131
DO - 10.1109/ICHI57859.2023.00131
M3 - Conference contribution
AN - SCOPUS:85181578634
T3 - Proceedings - 2023 IEEE 11th International Conference on Healthcare Informatics, ICHI 2023
SP - 717
EP - 721
BT - Proceedings - 2023 IEEE 11th International Conference on Healthcare Informatics, ICHI 2023
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 11th IEEE International Conference on Healthcare Informatics, ICHI 2023
Y2 - 26 June 2023 through 29 June 2023
ER -