Advancing Chronic Kidney Disease Prediction with Quaternion GANs for IoT and GGO Algorithm

Chronic disease detection using IoT devices involves leveraging secure, decentralized networks to collect, store, and share patient health data. Ensures data integrity, security, and transparency, while IoT devices provide real-time health monitoring. This combination enhances data accuracy, facilitates seamless information sharing among healthcare providers, and supports personalized, efficient care for chronic conditions. But the existing methods have lot of disadvantages such as, low precision, low accuracy and high error rate. To overcome these problems, Quaternion Generative Adversarial Networks with Greylag Goose Optimization Algorithm (QGADN-GGOA) is proposed. In this, IoT based input data is taken from a dataset such as MIMIC III dataset. It pre-processed data using Correlation Coefficients with Min-Max Normalization (CC- MMN), Following that Feature selection using Sea-Horse Optimizer (S-HO) and classification Quaternion Generative Adversarial Networks (QGAN) and the optimization using Greylag Goose Optimization Algorithm (GGOA) for detecting the type of Chronic Disease Management and to find the No CKD AND CKD. The introduced system is executed in python. The efficiency of the proposed QGADN-GGOA is analysed using a dataset and attains 99.9% accuracy, 0.1% error rate and 0.1s PT and attains better results compared with the existing methods. This demonstrates the approach's superior efficiency and the potential for future progress in the field.