Big Data-Driven Health Portraits for Personalized Management in Noncommunicable Diseases: Scoping Review

Area of Science:

• Health Informatics
• Big Data Analytics
• Noncommunicable Disease Management

Background:

• Big data health portraits offer insights for precise risk prediction and personalized management of noncommunicable diseases (NCDs).
• Current adoption is fragmented due to a lack of standardized frameworks for health portraits.
• A systematic review is needed to evaluate big data's role in NCD management through health portraits.

Purpose of the Study:

• To systematically map and categorize research on health portraits for NCD management.
• To evaluate big data utilization using the 3V framework (volume, velocity, variety).
• To assess external validation, comprehensiveness, and identify future research directions.

Main Methods:

• Conducted a scoping review following PRISMA-ScR guidelines.
• Searched multiple databases (PubMed, Embase, Scopus, etc.) for studies from 2014-2024.
• Analyzed data using keyword analysis, the 3V framework, and visual tools.

Main Results:

• 89 studies were analyzed, categorizing health portraits into diagnostic, prognostic, monitoring, and recommender types.
• Only 17.78% of studies met all 3V criteria; recommender portraits showed higher reliance on digital data.
• External validation was reported in 30% of studies, with only 10% meeting both 3V and external validation criteria.

Conclusions:

• Findings provide a framework for evaluating health portraits in NCD management.
• Emphasizes the need for robust data integration, AI-enabled approaches, and enhanced external validation.
• Addressing ethical and privacy concerns is crucial for advancing personalized health management.