Laparoscopic Surgery Duration Model Associated With Reduced Operating Room Turnover Time

An artificial intelligence model developed to estimate total operative duration was associated with significant reductions in operating room turnover and patient waiting times during laparoscopic cholecystectomy, according to findings from a non-randomized, quasi-experimental study.

Although artificial intelligence applications have expanded across multiple clinical domains, their implementation for managing operating room efficiency during laparoscopic cholecystectomy procedures has remained relatively underexplored.

To investigate this approach, researchers designed a two-phase study incorporating a historical control cohort. In Phase I, surgical data from 145 patients who underwent laparoscopic cholecystectomy under standard operating practices were analyzed to develop a predictive machine learning framework. The team built a Long Short-Term Memory Network model designed to forecast the min-max-normalized final total operative duration for each surgical case, which was subsequently inverse-transformed into minutes for clinical display.

In Phase II, the deployed artificial intelligence system was evaluated prospectively in an intervention cohort comprising 62 consecutive patients who served as a temporal test group. Data collected from this second phase were entirely separated from the initial development stage, having not been utilized during model training, hyperparameter tuning, or final model selection procedures.

Predictive accuracy was evaluated using the Root Mean Square Error metric to assess the discrepancy between actual operative durations and the inverse-transformed model predictions at the end of the surgical sequence. The neural network achieved an end-of-sequence Root Mean Square Error of 0.593 minutes within the Phase I development-validation subset and 0.637 minutes within the independent Phase II temporal test cohort.

When comparing clinical workflow metrics between the 145-patient traditional cohort and the 62-patient intervention group, the artificial intelligence-assisted mode demonstrated measurable workflow gains. Surgical turnover time decreased from 38.00 ± 2.91 minutes in the traditional group to 23.20 ± 2.97 minutes in the assisted group, representing a statistically significant difference. Patient waiting time also declined significantly from 11.15 ± 4.24 minutes under the traditional schedule to 5.93 ± 1.13 minutes in the assisted group.

Overall patient satisfaction scores were significantly higher in the intervention cohort than in the historical group, measuring 98.39 percent compared with 88.28 percent. In contrast, the analysis revealed no significant difference between the two cohorts regarding the first-surgery delay rate. Surgeon satisfaction outcomes were evaluated and presented descriptively because individual ratings collected at the session level were not statistically independent.

To account for potential confounding clinical factors, the investigators performed multivariable regression analyses controlling for patient age, sex, and total duration of surgery. These adjusted models confirmed that placement in the artificial intelligence-assisted group maintained independent associations with improved surgical turnover time, reduced waiting intervals, and higher patient satisfaction ratings, with all comparisons maintaining significance.

The study authors emphasized several procedural caveats, noting that this preliminary investigation was conducted at a single medical center. Furthermore, the reported Root Mean Square Error metrics reflect end-of-sequence performance and do not demonstrate model accuracy at earlier intraoperative time points. Consequently, the researchers characterized these workflow findings as exploratory, highlighting the need for validation across larger, multi-center trials.

Subject of Research: Artificial intelligence prediction of laparoscopic cholecystectomy operative duration

Article Title: A real-world study on establishing an artificial intelligence-based laparoscopic cholecystectomy final total operative duration prediction model to improve the operational efficiency of the operating room

Article References: Liu, X., Yang, Q., Xia, J., Bai, N., Chen, Q., Liu, Y., & Lin, Y. (2026). A real-world study on establishing an artificial intelligence-based laparoscopic cholecystectomy final total operative duration prediction model to improve the operational efficiency of the operating room. BMC Surgery. https://doi.org/10.1186/s12893-026-04251-x

Image Credits: AI Generated

DOI: 10.1186/s12893-026-04251-x

Keywords: laparoscopic cholecystectomy, artificial intelligence, operating room efficiency, long short-term memory network, operative duration prediction, surgical turnover time, patient waiting time, patient satisfaction, quasi-experimental study, temporal test cohort, root mean square error, first-surgery delay rate, surgeon satisfaction, multivariable regression analysis, clinical workflow, single-center study