Publicação
Vital signs acquisition and processing system for self-diagnosis
| Resumo: | Vital sign monitoring offers several advantages, both in hospital settings and in everyday life. Among these, the continuous patient monitoring and early detection of clinical disturbances. This project aimed to create a modular system to monitor a set of vital signs (body temperature, oxygen saturation, heart rate and respiratory rate) found in fitness and wellness wearable devices, finding a compromise between accuracy and comfort for a wearable device for prolonged use. Throughout the project, different algorithms for measuring these signals were developed and tested, in the frequency and time domains, as well as the development of signal processing and selection of the most appropriate configurations to ensure the project requirements. Finally, the resources used by the different algorithms were also studied: flash memory, RAM and execution times. The system achieved for resting heart rate an absolute error less than 7 bpm for the frequency domain and less than 15 bpm for the time domain. For respiratory rate, the time domain algorithm performed poorly, with a constant maximum error of 7 brpm for higher frequencies. In comparison, in the frequency domain, it has a maximum error of 5 brpm and an average absolute error of 1.4 brpm. For oxygen saturation, the algorithm performed well, presenting an average error of 1.9%.The AS6221 sensor also works as expected with a maximum absolute error of less than 0.38 ◦C. These results allow us to conclude that satisfactory quality of results is guaranteed for all vital signs, taking into account the requirements and limitations of the system resources. |
|---|---|
| Autores principais: | Leite, Inês Fernanda Azevedo |
| Assunto: | Heart rate Photoplethysmography SpO₂ Vital signals Wearables |
| Ano: | 2025 |
| País: | Portugal |
| Tipo de documento: | dissertação de mestrado |
| Tipo de acesso: | acesso aberto |
| Instituição associada: | Universidade de Aveiro |
| Idioma: | inglês |
| Origem: | RIA - Repositório Institucional da Universidade de Aveiro |
| Resumo: | Vital sign monitoring offers several advantages, both in hospital settings and in everyday life. Among these, the continuous patient monitoring and early detection of clinical disturbances. This project aimed to create a modular system to monitor a set of vital signs (body temperature, oxygen saturation, heart rate and respiratory rate) found in fitness and wellness wearable devices, finding a compromise between accuracy and comfort for a wearable device for prolonged use. Throughout the project, different algorithms for measuring these signals were developed and tested, in the frequency and time domains, as well as the development of signal processing and selection of the most appropriate configurations to ensure the project requirements. Finally, the resources used by the different algorithms were also studied: flash memory, RAM and execution times. The system achieved for resting heart rate an absolute error less than 7 bpm for the frequency domain and less than 15 bpm for the time domain. For respiratory rate, the time domain algorithm performed poorly, with a constant maximum error of 7 brpm for higher frequencies. In comparison, in the frequency domain, it has a maximum error of 5 brpm and an average absolute error of 1.4 brpm. For oxygen saturation, the algorithm performed well, presenting an average error of 1.9%.The AS6221 sensor also works as expected with a maximum absolute error of less than 0.38 ◦C. These results allow us to conclude that satisfactory quality of results is guaranteed for all vital signs, taking into account the requirements and limitations of the system resources. |
|---|
Atividades financiadas
Não foram encontrados projetos financiados.