BIG DATA ANALYTICS IN GLOBAL NON-COMMUNICABLE DISEASE TREND PREDICTION

Abstract

This study draws on advanced big data analytics to make global predictions of non-communicable diseases (NCDs) including cardiovascular diseases, diabetes, chronic respiratory diseases, and cancers using built-in datasets of epidemiological surveillance systems, electronic health records, demographic repositories, and socioeconomic indices.  The research applies machine learning algorithms, including long short-term memory (LSTM) networks, random forest regression and gradient boosting to identify new patterns in the occurrence of diseases, the mortality rate and the variations in risk across areas.  The findings indicate that LSTM models performed the best in predicting five-year incidence trajectories of NCDs (R 2 = 0.91) and ensemble models were the best in identifying high risk geographic clusters.  Evidence of data mining of world health indicators indicated that things such as urbanisation, nutrition change, ageing population, and income inequality had significant correlations with the burden of noncommunicable diseases (NCDs).  The data streams provided by multiple sources indicate that predictive heatmaps display that the population of people with NCDs is prone to grow substantially in low- and middle-income countries. By 2035, South Asia and Sub-Saharan Africa are likely to experience a 28 per cent increase in the number of people with NCDs.  Furthermore, early-warning analytics were effective in discovering the hypertension, diabetes, and obesity trends about 18-24 months before the traditional surveillance systems.  These findings demonstrate that big data analytics can transform the manner in which we plan to handle the global health by enabling us to implement evidence-based resource mobilization, focused prevention initiatives, and instant surveillance to decelerate the expanding global NCD epidemic.