Pulse Waveform Analysis for Cardiovascular Disease; The Balance of Biomedical Assessment and Social Determinants of Health for Medical Devices

The focus of the technical and STS research paper is on the role of medical devices within the medical field, and the impacts this can have on the patient-physician interaction. The goal of the technical project is to develop a noninvasive manner to monitor peripheral pulse waveform with the end goal of using machine learning to correlate key features of the waveform with cardiac failure severity. The goal of the STS research paper is to specifically investigate the role of medical technology in mitigating healthcare disparities. As the technical project is focused on the development of a novel medical device to act as a diagnostic tool in the future, the STS research paper investigates the impact this type of device could have on society and the ethics and biases of medical devices.
In my technical project, I worked on a device to gather pulse waveform measurements to be correlated with underlying cardiac diseases through the use of machine learning algorithms. Collaborating with Dr. Mazimba, cardiologist at the University of Virginia hospital, and Dr. Hossack, biomedical engineering assistant professor, I developed an instrument for an accurate measurement of pulse waveform through photoplethysmography. Improvements to the Arduino and Matlab code were made to increase the efficiency, and use of a screen with the hardware allows for more intuitive use of the device. ECG leads were added as a method of validation of the collected peripheral waveform data. 40 sets of inpatient data collection were obtained from patients with acute heart failure from the Cardiology unit of the UVA Hospital. The collected peripheral pulse waveforms were then transformed using a transfer function validated by literature review to a central waveform. Key traits from patient charts, including ejection fraction and New York Heart Association (NYHA) classification of heart failure, were used in conjunction with the collected and transformed pulse waveforms to help create the learning algorithm to monitor the severity of cardiac disease.
In my STS research paper, I investigated the interactions physicians have with patients of different socioeconomic status, and how the use of medical technology either alleviates or magnifies this difference. Specifically, the role of socioeconomic status and social determinants of health on health outcomes was investigated within the medical field, as well as the impact that the use of medical technology can have on mitigating these disparities. This research found there is an established relationship between factors such as poverty rate and breast cancer survival, posing ethical concerns over how the current system places implicit values on human life based on one’s background. Additionally, this research emphasizes the fact that biases within healthcare are impossible to avoid, with both interaction with a person and a medical device. The main conclusion from this research was that medical technology and its development is not causing healthcare inequities itself, as this is more of an issue of how the healthcare system is structured and the accessibility patients have to medical care.
In conclusion, both the technical and STS research portion of this project emphasize the impact that medical devices and technology can have on the care patients receive. While the technical project focuses on the improved effects new diagnostic technology can have on monitoring cardiac health, the STS research question investigates how some patients are placed in more advantageous positions to receive the medical care they need even though the medical technology exists for all.

School of Engineering and Applied Science
Bachelor of Science in Biomedical Engineering
Technical Advisor: Dr. Sula Mazimba, Dr. John Hossack
STS Advisor: Dr. Richard Jacques
Capstone Team Members: Lauren Orr