A Covid-19's integrated herd immunity (CIHI) based on classifying people vulnerability

Area of Science:

• Epidemiology and Public Health
• Computational Biology and Bioinformatics
• Infectious Disease Modeling

Background:

• COVID-19's rapid spread and varied individual impact necessitate severity estimation for timely public health responses.
• Asymptomatic individuals act as silent spreaders, highlighting the need for pre-infection detection to control unseen viral transmission.
• Classifying individuals by COVID-19 vulnerability before infection allows for targeted, personalized precautionary measures.

Purpose of the Study:

• To introduce the COVID-19 Integrated Herd Immunity (CIHI) strategy for societal protection with minimal losses.
• To develop and validate a Distance Based Classification Strategy (DBCS) for predicting COVID-19 susceptibility and identifying asymptomatic cases.
• To enable proactive public health interventions by classifying individuals into distinct vulnerability types.

Main Methods:

• The Distance Based Classification Strategy (DBCS) was developed, comprising three sequential phases: Outlier Rejection Phase (ORP) using Hybrid Outlier Rejection (HOR), Feature Selection Phase (FSP) using Hybrid Feature Selection (HFS), and Classification Phase (CP) using Accumulative K-Nearest Neighbors (AKNN).
• DBCS classifies individuals into six distinct vulnerability types, enabling tailored precautionary measures and identification of future symptomatic and asymptomatic cases.
• The proposed strategy was evaluated against existing COVID-19 diagnostic techniques using the 'NileDS' dataset.

Main Results:

• Experimental results demonstrated the efficiency and applicability of the DBCS strategy.
• The proposed DBCS achieved superior classification accuracy compared to other contemporary COVID-19 diagnostic methods.
• The strategy successfully identified individuals likely to be asymptomatic or severely affected upon infection.

Conclusions:

• The developed DBCS, as part of the CIHI strategy, offers an effective approach to manage COVID-19 spread by predicting individual susceptibility.
• Early identification of asymptomatic and high-risk individuals allows for timely and targeted interventions, enhancing herd immunity efforts.
• The DBCS strategy provides a robust framework for personalized risk assessment and public health preparedness against infectious diseases.