Culturally Adapted and Trust-Enhanced Biometric Protocols for Reliable Cardiovascular Data Collection in Low-Income Populations

Abstract Background: In low-income settings, cardiovascular data collection faces low adherence and poor-quality data owing to cultural mismatches and a possible lack of trust in conventional modes of data collection. We propose a culturally adapted and trust-enhanced biometric protocol to improve cardiovascular data collection in low-income populations. Methodology: We developed a culturally adapted and trust-enhanced biometric protocol integrating three interconnected modules: (1) cultural belief mapping using ethnographic interviews and participatory workshops to create a weighted bipartite graph linking cultural factors to device parameters; (2) trust-enhanced biometric adaptation employing a constrained optimisation function that maximises acceptability while maintaining clinical accuracy and (3) iterative implementation with federated learning architecture for privacy-preserving feedback integration. The system was evaluated across three diverse low-income communities in Nairobi, Bangladesh and Mexico. Results: The proposed method achieves higher reliability and acceptability compared to traditional approaches, as demonstrated by preliminary results from low-resource contexts. Moreover, the framework supports modular firmware adjustments in state-of-the-art wearable devices, enabling real-time interface customisation without compromising measurement precision. Conclusion: This work bridges a critical gap in cardiovascular epidemiology by embedding cultural and trust dynamics into the core of data collection, thereby fostering equitable health monitoring in underserved communities. The significance of this approach lies in its scalability and potential to inform future biometric interventions tailored to diverse sociocultural settings.