Frugal Heart Rate Correction Method for Scalable Health and Safety Monitoring in Construction Sites

• DOI: 10.3390/s23146464
• Publication Date: 2023-07-18T05:46:02Z
• Authors: Piotr Sowiński; Kajetan Rachwał; Anastasiya Danilenka; Karolina Bogacka; Monika Kobus; Anna Dąbrowska; Andrzej Paszkiewicz; Marek Bolanowski; Maria Ganzha; Marcin Paprzycki
• Keywords: Smartwatch; Wearable Technology

Continuous, real-time monitoring of occupational health and safety in high-risk workplaces such as construction sites can substantially improve the workers. However, introducing systems practice is associated with a number challenges, scaling up solution while keeping its cost low. In this context, work investigates use an off-the-shelf, low-cost smartwatch to detect issues based on heart rate privacy-preserving manner. To smartwatch’s low measurement quality, novel, frugal machine learning method proposed that corrects errors, along new dataset for task. This method’s integration remaining parts system (built ASSIST-IoT reference architecture) are presented. was evaluated laboratory environment terms accuracy, computational requirements, frugality. With experimentally established mean absolute error 8.19 BPM, only 880 bytes required memory, negligible impact performance device, meets all relevant requirements expected be field-tested coming months. support reproducibility encourage alternative approaches, dataset, trained model, implementation were published under free licenses.