Application of Data Science to Quantify the Effect of Propofol Infusion on Postoperative Nausea and Vomiting.

The effectiveness of propofol infusion on postoperative nausea and vomiting (PONV) is poorly understood in relation to various patient and procedure characteristics. This retrospective cohort study aimed to quantify the effectiveness of propofol infusion when administered either via total intravenous administration (TIVA) or combined intravenous anesthesia (CIVA) with inhalational agents on PONV. The relationship between propofol infusion and PONV was characterized controlling for patient demographics, procedure characteristics, PONV risk factors, and antiemetic drugs in adult patients (age ≥18 years) undergoing general anesthesia. Learned coefficients from multivariate regression models were reported as "lift" which represents the percentage change in the base likelihood of observing PONV if a variable is present versus absent. In a total of 41,490 patients, models showed that propofol infusion has a naive effect on PONV with a lift of -41% (P < .001) when using TIVA and -17% (P < .001) when using CIVA. Adding interaction terms to the model resulted in the loss of statistical significance in these relationships (lift of -30%, P = .23, when using TIVA, and -42%, P = .36, when using CIVA). It was further found that CIVA/TIVA are ineffective in short cases (CIVA * short surgery duration: lift = 49%, P < .001 and TIVA * short surgery duration: lift = 56%, P < .001).

Machine learning approach to predict postoperative opioid requirements in ambulatory surgery patients.

Opioids play a critical role in acute postoperative pain management. Our objective was to develop machine learning models to predict postoperative opioid requirements in patients undergoing ambulatory surgery. To develop the models, we used a perioperative dataset of 13,700 patients (≥ 18 years) undergoing ambulatory surgery between the years 2016-2018. The data, comprising of patient, procedure and provider factors that could influence postoperative pain and opioid requirements, was randomly split into training (80%) and validation (20%) datasets. Machine learning models of different classes were developed to predict categorized levels of postoperative opioid requirements using the training dataset and then evaluated on the validation dataset. Prediction accuracy was used to differentiate model performances. The five types of models that were developed returned the following accuracies at two different stages of surgery: 1) Prior to surgery-Multinomial Logistic Regression: 71%, Naïve Bayes: 67%, Neural Network: 30%, Random Forest: 72%, Extreme Gradient Boost: 71% and 2) End of surgery-Multinomial Logistic Regression: 71%, Naïve Bayes: 63%, Neural Network: 32%, Random Forest: 72%, Extreme Gradient Boost: 70%. Analyzing the sensitivities of the best performing Random Forest model showed that the lower opioid requirements are predicted with better accuracy (89%) as compared with higher opioid requirements (43%). Feature importance (% relative importance) of model predictions showed that the type of procedure (15.4%), medical history (12.9%) and procedure duration (12.0%) were the top three features contributing to model predictions. Overall, the contribution of patient and procedure features towards model predictions were 65% and 35% respectively. Machine learning models could be used to predict postoperative opioid requirements in ambulatory surgery patients and could potentially assist in better management of their postoperative acute pain.

Hyper-G: An Artificial Intelligence Tool for Optimal Decision-Making and Management of Blood Glucose Levels in Surgery Patients.

BACKGROUND: Hyperglycemia or high blood glucose during surgery is associated with poor postoperative outcome. Knowing in advance which patients may develop hyperglycemia allows optimal assignment of resources and earlier initiation of glucose management plan. OBJECTIVE: To develop predictive models to estimate peak glucose levels in surgical patients and to implement the best performing model as a point-of-care clinical tool to assist the surgical team to optimally manage glucose levels. METHODS: Using a large perioperative dataset (6,579 patients) of patient- and surgery-specific parameters, we developed and validated linear regression and machine learning models (random forest, extreme gradient boosting [Xg Boost], classification and regression trees [CART], and neural network) to predict the peak glucose levels during surgery. The model performances were compared in terms of mean absolute percentage error (MAPE), logarithm of the ratio of the predicted to actual value (log ratio), median prediction error, and interquartile error range. The best performing model was implemented as part of a web-based application for optimal decision-making toward glucose management during surgery. RESULTS: Accuracy of the machine learning models were higher (MAPE = 17%, log ratio = 0.029 for Xg Boost) when compared with that of the linear regression model (MAPE = 22%, log ratio = 0.041). The Xg Boost model had the smallest median prediction error (5.4 mg/dL) and the narrowest interquartile error range (-17 to 24 mg/dL) as compared with the other models. The best performing model, Xg Boost, was implemented as a web application, Hyper-G, which the perioperative providers can use at the point of care to estimate peak glucose levels during surgery. CONCLUSIONS: Machine learning models are able to accurately predict peak glucose levels during surgery. Implementation of such a model as a web-based application can facilitate optimal decision-making and advance planning of glucose management strategies.

Improving Operating Room Efficiency: Machine Learning Approach to Predict Case-Time Duration.

BACKGROUND: Accurate estimation of operative case-time duration is critical for optimizing operating room use. Current estimates are inaccurate and earlier models include data not available at the time of scheduling. Our objective was to develop statistical models in a large retrospective data set to improve estimation of case-time duration relative to current standards. STUDY DESIGN: We developed models to predict case-time duration using linear regression and supervised machine learning. For each of these models, we generated an all-inclusive model, service-specific models, and surgeon-specific models. In the latter 2 approaches, individual models were created for each surgical service and surgeon, respectively. Our data set included 46,986 scheduled operations performed at a large academic medical center from January 2014 to December 2017, with 80% used for training and 20% for model testing/validation. Predictions derived from each model were compared with our institutional standard of using average historic procedure times and surgeon estimates. Models were evaluated based on accuracy, overage (case duration > predicted + 10%), underage (case duration < predicted - 10%), and the predictive capability of being within a 10% tolerance threshold. RESULTS: The machine learning algorithm resulted in the highest predictive capability. The surgeon-specific model was superior to the service-specific model, with higher accuracy, lower percentage of overage and underage, and higher percentage of cases within the 10% threshold. The ability to predict cases within 10% improved from 32% using our institutional standard to 39% with the machine learning surgeon-specific model. CONCLUSIONS: Our study is a notable advancement toward statistical modeling of case-time duration across all surgical departments in a large tertiary medical center. Machine learning approaches can improve case duration estimations, enabling improved operating room scheduling, efficiency, and reduced costs.