Development of performance test bench for pulse oximetry algorithms
Arefyev, Pavel (2023)
2023
All rights reserved. This publication is copyrighted. You may download, display and print it for Your own personal use. Commercial use is prohibited.
Julkaisun pysyvä osoite on
https://urn.fi/URN:NBN:fi:amk-202304034651
https://urn.fi/URN:NBN:fi:amk-202304034651
Tiivistelmä
Pulse oximetry peripheral oxygen saturation (SpO2) is considered one of the vital parameters in modern patient care. SpO2 is actively monitored on all stages of hospitalization: in wards, intensive care units, operation rooms and emergency departments. SpO2 measurement is based on photoplethysmography. The oxygen saturation value is derived from photoplethysmographic waveform using two or more different wavelengths by utilizing pulse oximetry algorithms. The goal of this work was to design and implement a tool for profiling the performance of different SpO2 algorithms and their revisions. This tool is referred as test bench in the context of this work.
The test bench software developed in this work is a software that provides tools for debugging, testing and performance assessment of SpO2 algorithms. The test bench was divided into modules to support various data types and development needs. The modules consisted of statistical and visual data analysis functionality, as well as testing and data processing logic. The software was written using Python programming language and a set of third-party packages. The interface was effectively used with proprietary pulse oximetry algorithm and desaturation study data.
As any other software, pulse oximetry requires proper testing and profiling. Requirements for pulse oximetry devices are listed by International Standardization Organization as well as regulatory agencies, for instance Food and Drug Administration of the United States. These requirements were used as a basis for designing the analyzes that the test bench software provides.
It was concluded that test bench software can be used to profile and visualize SpO2 algorithm performance. In addition, the test bench software has the potential to be used for automation of clinical evidence collection for regulatory submissions.
The test bench software developed in this work is a software that provides tools for debugging, testing and performance assessment of SpO2 algorithms. The test bench was divided into modules to support various data types and development needs. The modules consisted of statistical and visual data analysis functionality, as well as testing and data processing logic. The software was written using Python programming language and a set of third-party packages. The interface was effectively used with proprietary pulse oximetry algorithm and desaturation study data.
As any other software, pulse oximetry requires proper testing and profiling. Requirements for pulse oximetry devices are listed by International Standardization Organization as well as regulatory agencies, for instance Food and Drug Administration of the United States. These requirements were used as a basis for designing the analyzes that the test bench software provides.
It was concluded that test bench software can be used to profile and visualize SpO2 algorithm performance. In addition, the test bench software has the potential to be used for automation of clinical evidence collection for regulatory submissions.