Synapse
⌘+K
Synapse
PulseExploreClubsResearchersJournals
Instagram
HomeClubsExplore
May 24, 2026BMJ Public HealthOpen Access

AI applications achieve up to ~97% accuracy for hypertension risk prediction in LMICs.

View Full Paper
Ask AI
Bookmark
Share

Why the study?

Hypertension disproportionately affects low- and middle-income countries, and this review aimed to identify current research, innovations, and developments in applying AI-based tools for hypertension prediction and risk assessment specifically in these settings.

Do artificial intelligence-based tools accurately predict and assess the risk of hypertension in low- and middle-income countries?

Population

5 studies on AI-based risk assessment and prediction of hypertension in LMICs

Design

Scoping review

Key result

Artificial intelligence applications for hypertension risk assessment and prediction in low- and middle-income countries achieved prediction accuracies ranging from 78% to 97%.

Authors

DSDavid SasuTATsatsu AgbettorAAAngela Owusu Ansah

Discussion

Loading...

Member takes

Overview

AI tools may aid hypertension screening in LMICs; leaves open need for prospective validation before clinical adoption.

Key Points

  • The review aims to identify AI-based tools enhancing hypertension prediction and risk assessment in low- and middle-income countries.
  • Conducted a scoping review following PRISMA guidelines.
  • Performed literature searches on multiple databases (PubMed, Scopus, etc.).
  • Selected studies focused on AI applications for hypertension in low- and middle-income countries from a total of 1371 papers.
  • Identified significant AI research in hypertension risk assessment, with prediction accuracies ranging from 78% to 97%.
  • Notable study from Ethiopia achieved an accuracy of 88.81%, precision of 89.62%, and recall of 97.04%.
  • Algorithms analyzed included logistic regression, decision trees, and naïve Bayes.

Structured PICO

Do artificial intelligence-based tools accurately predict and assess the risk of hypertension in low- and middle-income countries?

P
Population
5 studies focusing on AI-based risk assessment and prediction of hypertension in low- and middle-income countries (LMICs)
I
Intervention
Artificial intelligence (AI) algorithms or tools (e.g., machine learning, logistic regression, decision trees, naïve Bayes) for hypertension prediction and risk assessment
O
Outcome
Accuracy of hypertension risk prediction and assessment

AI-based tools demonstrate high accuracy (78-97%) for hypertension risk prediction in LMICs, highlighting their potential to enhance disease prevention and management in resource-limited settings.

Cite This Study

Sasu et al. (2026) conducted a review in Hypertension (n=5). Artificial intelligence (AI) algorithms was evaluated on Prediction accuracy of AI models. Artificial intelligence applications for hypertension risk assessment and prediction in low- and middle-income countries achieved prediction accuracies ranging from 78% to 97%.

synapsesocial.com/papers/6a1295f648a0ea16656725bfhttps://doi.org/10.1136/bmjph-2025-003435
View Full Paper
Ask AI
Bookmark
Share