Probabilistic Integration Random Forest Decision Tree Fusion Model
Ayesha Butalia, Debarshi Basu Bhattacharjee, Keerti Satpute · Advances in computational intelligence and robotics book series · 2024
Our study proposes an Integration Random Forest Decision Tree Fusion Model (IRFDTFM) for detecting kidney stones, utilizing two datasets comprising 414 and 89 records, respectively. These datasets include measurements such as gravity, pH, osmolality, conductivity, urea, calcium, and a binary target variable indicating stone presence. Unlike existing methodologies, which primarily focus on detection, our approach aims to uncover the underlying factors contributing to kidney stone formation by leveraging Decision Tree and Random Forest methodologies, coupled with exploratory data analysis and feature engineering. Through this, we aim to enhance the efficiency and accuracy of kidney stone detection, with objectives including identifying crucial factors responsible for kidney stone formation and contributing to early and accurate detection to improve healthcare outcomes. By integrating probabilistic methods with advanced machine learning techniques, our model offers a solution for kidney stone detection, with potential usability and applicability in clinical settings.