Novel Gaidai hypersurface multidimensional biosystem nonstationary dynamics prognostics: an algorithmic approach

Area of Science:

• Biostatistics
• Epidemiology
• Public Health

Background:

• Forecasting mortality rates for major diseases like cancer, cardiovascular disorders, and diabetes is vital for public health planning.
• Existing statistical methods struggle with high-dimensional, multi-regional spatiotemporal data, particularly beyond bivariate analyses.

Purpose of the Study:

• To apply a novel multi-failure-mode Gaidai hypersurface spatiotemporal prognostics concept to analyze clinical surveillance data.
• To benchmark a multimodal methodology for assessing the reliability of complex biosystems.

Main Methods:

• A multicentre, population-based biostatistical methodology was applied to 685-dimensional biosystems data from 195 nations.
• The study addresses challenges in extending Extreme Value Theory (EVT) from univariate to high-dimensional spatiotemporal data.

Main Results:

• The proposed multimodal methodology effectively handles complex spatiotemporal public health data.
• Projected 15- and 100-year mortality rates with 95% Confidence Intervals (CI) for cancer, cardiovascular, and diabetes disorders were reported.

Conclusions:

• The findings have implications for digital health and multimodal prognostics tools, especially for analyzing big data in public health.
• The multivariate bio-reliability concept enhances bio-risk estimates, particularly with limited data sample sizes.