Developing a Pulse Oximetry Sensor that Measures Reflected,
Rather than Penetrating Light, to Evaluate Patient Blood-Oxygen Saturation for use in a Mobile Cardiac Health Assessment Monitor
Background
The United States spends more on healthcare expenses than any other developed nation at an estimated $2 trillion annually[1]. According to the Organization for Economic Cooperation and Development (OECD), 17.4 percent of the U.S. GDP was spent on healthcare costs alone in 2009. This is over 5 percent more than the next highest reported percent GDP spent on healthcare costs by the Netherlands at 12 percent[2]. This figure is estimated to rise to 25 percent by 2025, according to the Congressional Budget Office (CBO)[3].
A significant portion of these healthcare costs can be attributed to medical expenses associated with heart disease, which is defined by the A.D.A.M Medical Encyclopedia as the “narrowing of the small blood vessels that supply blood and oxygen to the heart,” and is highly correlated with heart failure[4]. Statistical data reported by the Centers for Disease Control and Prevention National Center for Health Statistics shows that 26% of 2,423,712 U.S. deaths in 2007 were directly related to heart disease[5].
Rapidly growing healthcare costs could be reduced by educating patients with heart disease of their symptoms and the actions that they can take to reduce serious risks to their health. Encouraging patients to monitor their own health and symptoms reduces the need for medical assistance and, in some cases, could prevent potentially fatal consequences of not recognizing critical symptoms related to heart disease or heart failure.
Mobile Cardiac Health Assessment Monitor
In an effort to promote patient self-care through education and mobile self-monitoring of symptoms associated with heart disease, development has begun on a mobile and non-intrusive cardiac health monitoring device including a suite of integrated sensors such as a pulse oximeter to measure blood-oxygen saturation, an electrode for measuring electrical activity in the heart, and accelerometer to measure movement and deduce physical activity. The device is described as:
a highly mobile collaborative patient-centric, self-monitoring, symptom recognition and self intervention system along with a complementary clinical nursing tool to aid in collaborative patient/clinician chronic cardiac disease management. Our system is composed of a mobile smart phone and wearable sensor suite linked through Bluetooth and cell phone technology to a backend data repository, data mining, knowledge discovery, knowledge evolution and knowledge processing system, providing clinical data collection, procedural collection, intervention planning, medical situational assessment and health status feedback for collaborative users. The system uses a combination of physiologic and psychological instruments to gather patient specific information. The collaborative system aids patients in learning to recognize disease symptoms and understand the effect on their health of adherence to interventions. Secondly the system provides timely clinical data collection, assessment and interventions so clinicians can improve the overall health and lower the re-admission of their patients.[6]
The scope of this thesis deals primarily with the design and integration of a reflectance pulse oximetry sensor as well as development of the software necessary to process and interpret sensor readings. The data generated by this sensor and other integrated sensors will be used in conjunction with relevant demographic, physiological, and psychological information as available to generate a cardiac wellness metric that can be used by both patients and clinicians to improve and facilitate monitoring of patient health. A more in depth description of pulse oximetry sensor function and design is provided below.
Pulse Oximetry
A pulse oximeter is a medical sensor that measures oxygen saturation in a patient's blood. Typical pulse oximeter designs are based on the “red and infrared light absorption characteristics of oxygenated and deoxygenated hemoglobin.” An emitter passes both infrared and red light through a sufficiently translucent part of a patient’s body with good blood flow, such as a finger or earlobe and measures the resulting intensity (see Figure 1). The amount of light that passes through the patient’s body can be measured with a photodetector and, using the known absorption spectrum of oxygenated and deoxygenated hemoglobin (shown in Figure 2), the level of oxygen saturation in patient’s blood can be determined.
[7]
Figure 1: Typical pulse oximetry probe placed on a patient’s finger[8]
A potential drawback of this design, as mentioned previously, is that it requires that light is passed through a sufficiently translucent medium. This limits sensor placement to a finite number of places on the body. For applications implementing pulse oximetry that prohibit adequate sensor placement, an alternative design is required.
Figure 2: Absorption spectrum of reduced hemoglobin (Hb) and oxyhemoglobin (HbO2)[8]
One such alternative design measures reflected, rather than absorbed light, to determine blood oxygen saturation. In such an application, a red/infrared light emitter is placed on the surface of the body in close proximity to a photodetector. The photodetector is used to measure the amount of backscattered, rather than penetrating light from the emitter, from which the blood-oxygen saturation level can be determined (see Figure 3).
Figure 3: Reflectance pulse oximeter probe and the different layers of skin[9]
In a 1992 research paper, published by the International Pediatric Research Foundation, the feasibility of reflectance pulse oximetry was examined. Through use of a prototype reflectance pulse oximeter on fetal lambs and a comparison to simultaneously taken arterial blood samples, it was determined not only to be a feasible solution, but also that blood-oxygen saturation could be measured with a reasonable precision of 4.7%.[10]
Furthermore, research published in IEEE Transactions on Biomedical Engineering shows that reflectance pulse oximeter designs can be optimized to improve performance. For example, local heating of the skin is shown to "increase the pulsatile component of the reflected photoplethysmograms," which can improve sensor accuracy. In addition, the distance between the light source and photodetector can be optimized to improve signal-to-noise ratio.[9]
Problem Statement
A prototype Cardiac Health Assessment Monitor has been created using a typical absorption pulse oximeter. However, due to the nature of the monitor application, namely that its bandage-like packages is intended to be surface mounted on a patient's body, typical pulse oximeter designs will not be useful. This is because the design inhibits the ability to measure penetrating light through the patient's body.
The purpose of this thesis is to evaluate the feasibility of, and subsequently design, implement, and integrate a reflected light sensing pulse oximeter, as well as the software necessary to process sensor readings to produce useful blood-oxygen saturation data that, in conjunction with other components integrated in a mobile device, will be used to assess patient cardiac health.
Approach
The problem statement has been defined above. Further research will be conducted into needs assessment of the device. A set of requirements will be produced. These requirements will define necessary design constraints, such as sustained and maximum allowable power consumption, physical specifications such as size, placement, and interface, minimum sensor accuracy, and constraints relating to physical and software integration of the device.
Various designs will be produced and assessed to determine if design requirements are met. Advantages and disadvantages of each design will be determined and weighted to determine the best design solution. The design that best meets the design requirements will be selected for further development. At this point, work will begin on a prototype reflectance oximeter. During development, the selected design will be refined and reevaluated in its ability to meet design requirements.
Upon completion of a functional prototype, testing will be performed to further refine and reevaluate the design. Test cases reflecting design requirements will be defined and implemented. Test results will influence design modification, and if necessary, development will begin on an improved prototype of the updated design. This processes may be iterated several times until all requirements are satisfied and no further improvements can be made.
A final design will be determined for the reflectance pulse oximeter and integrated in the final design of the mobile cardiac health assessment monitor. It will represent a final, manufacturable design that is ready for production. Additional work will also be necessary to create appropriate documentation, including technical designs and user manuals.
Works Cited
[5]
Jiaquan Xu,
M.D., Kenneth D. Kochanek, M.A., Sherry L. Murphy, B.S., Betzaida Tejada-Vera,
B.S. "Deaths: Final Data for 2007." National Vital Statistics
System (U.S. Department of Health and Human Services, Centers for Disease
Control and Prevention, National Center for Health Statistics) 58, no. 19 (May
2010).
[6]
Paul J. Fortier,
Brendon Puntin, Osama Aljaroudi. "Improved Patient Outcomes through
Collaborative Monitoring and Management of Subtle Behavioral and Physiological
Health Changes." 44th Hawaii International Conference on System
Sciences (HICSS). Kauai, HI : IEEE Computer Society, 2011. 1-10.
[7]
Steven Barker,
M.D., Ph.D., Willian Hay, M.D., Katsuyuki Miyasaka, M.D., Ph.D., FAAP, FCCP, Christian
Poets, M.D. Pulse Oximetry. September 10, 2002.
http://www.oximetry.org/pulseox/principles.htm (accessed September 25, 2011).
[8]
L. M. Schnapp,
N. H. Cohen. "Pulse Oximetry. Uses and Abuses." Chest
(American College of Chest Physicians) 98, no. 5 (November 1990): 1244-1250.
[9]
Y. Mendelson,
B.D. Ochs. "Noninvasive pulse oximetry utilizing skin reflectance
photoplethysmography." IEEE Transactions on Biomedical Engineering,
October 1988: 798-805.
[10]A. Carin Dassel, Reindert Graaff, Jan G. Aarnoudse, Jan M. Elstrodt,
Pieter Heida, Marco H. Koelink, Frits F. de Mul, Jan Greve. "Reflectance
Pulse Oximetry in Fetal Lambs." Pediatric Research, March 1992:
266-269.
