Your browser is unsupported

We recommend using the latest version of IE11, Edge, Chrome, Firefox or Safari.

BME.12 Pulse Oximeter Without Racial or Ethnic Bias

Team Members Heading link

  • Ayo Adekoya
  • Destiny Cruickshank
  • Amina Nazimuddin
  • Behnam Shamsaddin
  • Jannah Yusuf

Project Description Heading link

Pulse oximeters are essential devices in clinical and home settings as they are used to provide consistent and reliable oxygen saturation, SpO2, readings to the user. The accuracy of pulse oximeters has been questioned due to recent recordings of higher than actual readings for individuals with darker skin pigmentation. These inaccuracies became more prominent during the beginning of the COVID 19 pandemic, where many people sought the use of pulse oximeters to measure their vitals. This bias against individuals with darker skin pigmentation has been found to be a result of darker individuals melanin absorbing some of the light emitted by the pulse oximeter, thus yielding a higher SpO2 reading than their true value. In this project, we sought a method of increasing the reliability of the use of pulse oximeters in darker pigmented individuals. To accomplish this, we focused on developing a device that considers the needs and requirements of our expected users. This involved creating a device that focuses on providing accurate oxygen saturation measurements for patients with darker skin pigmentation. As the device will be used by doctors, nurses, EMTs, and medical technicians, it is made with the intent to have minimal set up and a user interface similar to currently available pulse oximeter devices. A mathematical approach to uncover and correct bias in pulse oximeters was executed. Based on the results of our mathematical model, we designed and prototyped a calibrated pulse oximeter device using an RGB sensor chip to determine the users skin pigmentation and adjust their SpO2 reading accordingly. With additional resources, data robustness, and approved protocol for extensive research, we can test and verify the statistical significance between commercially available pulse oximeters, direct blood oxygen analysis, and our prototype which considers racial/ethnic bias. Future studies would include other methods to account for the individual patient’s pigmentation and discussion over the efficacy of the methods with respect to what to pursue further for practical implementation. Additional aspects for this project would include considerations of higher quality materials, different points for calibration, and data collection methods.