Related Experiment Video
Updated: Jun 6, 2025
12:10
Retzius-Sparing Robot-Assisted Radical Prostatectomy
Published on: May 19, 2022
7.5K
Development and Assessment of an AI-based Machine Learning Model for Predicting Urinary Continence and Erectile
S Saikali1, S Reddy1, M Gokaraju2
1AdventHealth Global Robotics Institute, FL, USA.
Computer Methods and Programs in Biomedicine
|December 3, 2024
Summary
Artificial intelligence models can predict urinary continence and erectile function after nerve-sparing robotic radical prostatectomy (RARP). These models identified age and comorbidities as key predictors, offering potential for improved patient outcome prediction.
Area of Science:
- Urology
- Artificial Intelligence
- Oncology
Background:
- Prostate cancer treatment, including radical prostatectomy, presents challenges in predicting postoperative functional outcomes like urinary continence and erectile function.
- Artificial intelligence (AI) offers a promising avenue for developing advanced predictive models in this domain.
Purpose of the Study:
- To develop and validate AI-based models for predicting continence and potency after nerve-sparing robotic radical prostatectomy (RARP).
Main Methods:
- Analysis of 8,524 patients undergoing RARP, collecting preoperative variables.
- Training two Artificial Neural Network (ANN) models to predict 12-month postoperative continence and potency.
- Assessing model performance using Area Under the Curve (AUC) and comparing with other machine learning algorithms.
Main Results:
- ANN models achieved AUCs of 0.74 for potency and 0.68 for continence, outperforming other algorithms.
- Feature importance analysis highlighted preoperative factors like age, comorbidities, and baseline scores as significant predictors.
Conclusions:
- AI models demonstrate potential for predicting functional outcomes post-RARP.
- Further optimization, exploration of additional factors, and multi-center studies are necessary to enhance predictive accuracy and clinical utility.
- AI holds promise for improving clinical decision-making in prostate cancer management.
