Crunching Data for Early Warning: Predicting Renal Conditions in Real Time

Chronic Renal Disease (CRD) poses a significant global health concern, impacting approximately one in five men and one in four women. Its insidious progression and late-stage diagnosis present a formidable challenge within the medical community. In response, our research has been dedicated to the development of a real-time CRD identification system, leveraging a dataset encompassing 798 individuals from a hospital. This dataset laid the groundwork for crafting an exceptionally accurate machine learning model. seamlessly integrated into a user-friendly web application using Streamlittechnology. Among the array of machine learning algorithms we explored, the LightGBM(light gradient-boosting machine)Algorithm distinguished itself by achieving an impressive 98.12% accuracy in predicting CRD. Beyond this accuracy, our investigation delved into the critical attributes that bear the most significance in CRD detection gives the top attributes from each algorithms with providing invaluable insights for healthcare professionals. Our interactive web application not only furnishes accurate predictions but also empowers healthcare providers and individuals to proactively assess CRD and institute early interventions. Our ultimate goal is to make substantial strides in addressing this challenging medical condition, thereby contributing to heightened survival rates and improved overall outcomes for individuals grappling with CRD on a global scale.