International journal of urology : official journal of the Japanese Urological Association · 2025
Operative Time Prediction by Machine Learning for Robot-Assisted Laparoscopic Radical Prostatectomy.
СтатьяМногоцентровое исследованиеНазвание и аннотация: Перевод в очереди
Обработка статьи · этапы, ошибки и управление
Загрузка состояния…
Ошибки и ограничения
История · было → стало
Состояние обновляется каждые 5 секунд, пока блок открыт. Процент показывается только там, где известен объём работы. Пауза останавливает продолжение этапа; текущий запрос к источнику может завершиться.
Suzuki Yu, Sonobe Shinya, Kawasaki Yoshihide, Ikumi Saori, Takaya Eichi, Miyauchi Carlos Makoto, Kobayashi Tomoya, Iwazaki Junya, Adachi Hisanobu, Kawamorita Naoki, Ito Akihiro
Аннотация
OBJECTIVES: Operative time prediction is crucial for efficient operating room scheduling. This study developed and validated an operative time prediction system for robot-assisted laparoscopic radical prostatectomy using data from two institutions.
METHODS: Retrospectively identified 557 and 150 patients who underwent RARP at Tohoku University Hospital and Miyagi Cancer Center were analyzed. The following variables were collected as explanatory variables for the prediction system: Age, height, body mass index, comorbidities, performance status, prostate volume, tumor extent of local invasion, grade group, prostatitis, preoperative treatment, history of intraperitoneal surgery, lymphadenectomy, and nerve-sparing. In addition, the observed operative time was collected as a source variable for the objective variable. The observed operative time was calculated as the sum of components related to institutional, operator, and patient factors. Approximation curves were applied to the first two components, while a random forest-based machine learning model was applied to the latter, resulting in the development of an integrated prediction system.
RESULTS: The normalized root mean square error between the observed and predicted operative times was 0.107 in internal validation and 0.148 in external validation, demonstrating greater reliability than the expected operative time by operators in advance. Factors that are known to impact operative time, such as lymphadenectomy, grade group, prostate volume, and body mass index, were identified as contributing variables.
CONCLUSIONS: The system provides a reliable and robust prediction focusing on factors known to impact operative time. It has the potential to improve the efficiency of operating room scheduling.
METHODS: Retrospectively identified 557 and 150 patients who underwent RARP at Tohoku University Hospital and Miyagi Cancer Center were analyzed. The following variables were collected as explanatory variables for the prediction system: Age, height, body mass index, comorbidities, performance status, prostate volume, tumor extent of local invasion, grade group, prostatitis, preoperative treatment, history of intraperitoneal surgery, lymphadenectomy, and nerve-sparing. In addition, the observed operative time was collected as a source variable for the objective variable. The observed operative time was calculated as the sum of components related to institutional, operator, and patient factors. Approximation curves were applied to the first two components, while a random forest-based machine learning model was applied to the latter, resulting in the development of an integrated prediction system.
RESULTS: The normalized root mean square error between the observed and predicted operative times was 0.107 in internal validation and 0.148 in external validation, demonstrating greater reliability than the expected operative time by operators in advance. Factors that are known to impact operative time, such as lymphadenectomy, grade group, prostate volume, and body mass index, were identified as contributing variables.
CONCLUSIONS: The system provides a reliable and robust prediction focusing on factors known to impact operative time. It has the potential to improve the efficiency of operating room scheduling.