AI-Powered Genetic Algorithm-Based System for Predicting Diabetics Risk in Future Generations

Abstract

Diabetes is a global health concern leading to complications such as high blood pressure, kidney failure, and vision impairment. Effective management of diabetes involves maintaining a healthy lifestyle with proper diet and exercise. This research focuses on predicting the transmission of diabetes to future generations using classical Mendelian genetics combined with a probabilistic model enhanced by Artificial Intelligence (AI) and machine learning (ML).  Real-time data was collected from endocrinologist associated with various families maintaining the ethical privacy of their datasets. The study employs Mendelian laws of genetics to analyze the inheritance patterns of diabetes focusing on dominant and recessive alleles. Integrating Mendelian genetics with probabilistic rules (Naïve Bayes), the methodology offers a robust framework for predicting diabetes transmission. The findings highlight significant generational differences, revealing that diabetic parents have a strong likelihood in a sample of 400 individuals of passing diabetes to their offspring. This intelligent application allows individuals without technical or biomedical knowledge to easily predict the risk of diabetes in their next generation. This application validated through Mendelian laws and probabilistic models using AI and ML, holds promise for future expansion to address related diseases like kidney failure, eye problems, high blood pressure, and heart disease. By making this tool accessible to a lay audience, it enhances understanding of diabetes inheritance and offers a powerful resource for genetic prediction.