Triggering of Ambulatory Blood Pressure Measurement Based on Patient Status: Software Architecture and Implementation
Herranen, Juho (2022)
Herranen, Juho
2022
Sähkötekniikan DI-ohjelma - Master's Programme in Electrical Engineering
Informaatioteknologian ja viestinnän tiedekunta - Faculty of Information Technology and Communication Sciences
This publication is copyrighted. You may download, display and print it for Your own personal use. Commercial use is prohibited.
Hyväksymispäivämäärä
2022-05-17
Julkaisun pysyvä osoite on
https://urn.fi/URN:NBN:fi:tuni-202205044358
https://urn.fi/URN:NBN:fi:tuni-202205044358
Tiivistelmä
Blood pressure is an important physiological parameter that is used for both assessing immediate health status of hospitalized patients and providing indications of various cardiovascular diseases. Invasive blood pressure measurement has stayed as gold standard of blood pressure measurement while oscillometric measurement has established its position as the primary measurement method in hospital wards and home care. However, research around continuous non-invasive blood pressure measurement (CNIBP) methodologies have been growing, and blood pressure monitoring devices using CNIBP have developed recently. Applied CNIBP methods include, but are not limited to, pulse wave velocity and pulse wave analysis.
In this thesis, a prototype software system for detecting significant and sustained changes in a patient’s blood pressure was designed and implemented. The system is based on pulse wave analysis based continuous blood pressure measurement algorithm. The goal was to either trigger a cuff-based measurement automatically or to prompt the user to take a new cuff measurement when needed. Characteristics of the applied CNIBP method set requirements for the system. CNIBP measurement is affected by the patient’s posture as well as movement, and therefore, information about the activity of the patient was needed. Furthermore, the ambulatory patient monitoring system, in which the prototype was integrated, set architectural requirements for the developed system. Signal fault conditions were essential to recognize and handle by the implemented software.
The implemented system consists of four parts: continuous blood pressure estimation, patient activity detection, evaluation of the need for the blood pressure measurement, and notifier. The system uses a photoplethysmographic signal from an oxygen saturation sensor as an input for the blood pressure estimator. Accelerometer signals from the patient’s chest and wrist are used to detect the patient’s posture and activity. Continuous blood pressure estimate and patient activity information are used in assessing the need for a cuff-based blood pressure measurement. The system is designed to operate alongside auto-cycling ambulatory blood pressure monitoring.
The algorithm that estimates blood pressure changes was provided by an external partner while the algorithm classifying the patient’s activity was developed in GE Healthcare. The algorithm that estimates the need for the blood pressure measurement was developed in a collaboration with a team of engineers working on the project. The parts of the system mentioned above were combined into the functional system and integrated into the ambulatory monitoring system.
It was demonstrated that the system can detect significant and sustained blood pressure changes reliably, while at the same time discarding false readings in continuous blood pressure, as well as the blood pressure changes caused by the subject’s activity. Therefore, the system can provide actionable information about the changes in patient blood pressure and adds new value to patient monitoring.
In this thesis, a prototype software system for detecting significant and sustained changes in a patient’s blood pressure was designed and implemented. The system is based on pulse wave analysis based continuous blood pressure measurement algorithm. The goal was to either trigger a cuff-based measurement automatically or to prompt the user to take a new cuff measurement when needed. Characteristics of the applied CNIBP method set requirements for the system. CNIBP measurement is affected by the patient’s posture as well as movement, and therefore, information about the activity of the patient was needed. Furthermore, the ambulatory patient monitoring system, in which the prototype was integrated, set architectural requirements for the developed system. Signal fault conditions were essential to recognize and handle by the implemented software.
The implemented system consists of four parts: continuous blood pressure estimation, patient activity detection, evaluation of the need for the blood pressure measurement, and notifier. The system uses a photoplethysmographic signal from an oxygen saturation sensor as an input for the blood pressure estimator. Accelerometer signals from the patient’s chest and wrist are used to detect the patient’s posture and activity. Continuous blood pressure estimate and patient activity information are used in assessing the need for a cuff-based blood pressure measurement. The system is designed to operate alongside auto-cycling ambulatory blood pressure monitoring.
The algorithm that estimates blood pressure changes was provided by an external partner while the algorithm classifying the patient’s activity was developed in GE Healthcare. The algorithm that estimates the need for the blood pressure measurement was developed in a collaboration with a team of engineers working on the project. The parts of the system mentioned above were combined into the functional system and integrated into the ambulatory monitoring system.
It was demonstrated that the system can detect significant and sustained blood pressure changes reliably, while at the same time discarding false readings in continuous blood pressure, as well as the blood pressure changes caused by the subject’s activity. Therefore, the system can provide actionable information about the changes in patient blood pressure and adds new value to patient monitoring.