Enhancing Health Data Assessment Performance: A Multi-Sensor Focused Study for Public Health

Abstract

The term health data assessment performance described the process of assessing efficacy of acquiring and reporting health data. Multi sensor approach for predicting diabetes in public health scenarios, multi sensor for public health assimilates assorted data source and addresses community health challenges widely and efficiently. In this study, we proposed a Horse Herd Fused Discriminant Generative Adversarial Network (HHF-DGAN) algorithm for diabetes prediction. Sensor-based diabetes data is gathered from kaggle, and then data preprocessing using z-score normalization is inaccurate records from a dataset. To extract feature from the pre-processed data, Short-Time Fourier Transform (STFT) is used for dimensionality reduction. HHF-DGAN is a public health technique that addresses diabetes prevention and management while also including multi sensor assessment data. The proposed methods are to evaluate F1-score (86%), precision (92%), recall (93%), and accuracy (93%). The effectiveness of the suggested strategy in improving prediction reliability for diabetes detection in public health circumstances has been established by comparing it with existing methods. The study concludes field of healthcare assessments with a robust framework for leveraging multi sensor data and deep learning methods in disease prediction.