A neural network - based algorithm for predicting stone - free status after ESWL therapy.
Ilker Seckiner1, Serap Seckiner2, Haluk Sen1
1Department of Urology, Gaziantep University, Gaziantep, Turkey.
Summary
An artificial neural network (ANN) model accurately predicts kidney stone treatment success after Extracorporporeal Shock Wave Lithotripsy (ESWL). This tool aids in planning ESWL for renal stones, improving patient outcomes.
Area of Science:
- Urology
- Medical Informatics
- Artificial Intelligence
Background:
- Kidney stones pose a significant health challenge, necessitating effective treatment planning.
- Extracorporporeal Shock Wave Lithotripsy (ESWL) is a common treatment modality for renal stones.
- Predicting treatment success is crucial for optimizing patient management and outcomes.
Purpose of the Study:
- To develop a prototype artificial neural network (ANN) model for predicting stone-free status in patients undergoing ESWL.
- To utilize the ANN model to aid in treatment planning for kidney stones.
Main Methods:
- Data from 203 patients with renal stones were collected, including demographic and stone-specific variables.
- Eleven variables such as stone size, density, and patient age were analyzed.
- Regression analysis and ANN methods were employed to predict treatment success.
Main Results:
- The ANN model achieved high prediction accuracy: 99.25% in the training group, 85.48% in the validation group, and 88.70% in the test group.
- Patients were categorized into training, validation, and test groups for model implementation.
- The model demonstrated robust performance across different patient cohorts.
Conclusions:
- The developed ANN model successfully predicts the stone-free rate after ESWL for kidney stones.
- This predictive tool can assist clinicians in planning ESWL treatments.
- The study highlights the potential of AI in optimizing urological interventions.
