Student Projects
Contact Information
University and Department: Penn State, Department of Bioengineering
Team Members: Kate Noa, Elizabeth Shenk, Noah Johnson, Ray Cahill, Jeff Krawiec, Michael Irick
Primary Email Address:ken5005@psu.edu
Project Information
Describe the challenge your project is trying to solve.
Blood pressure is an extremely important vital sign. It must be monitored every-so-often because most of the serious diseases associated with hypertension such as heart attack, stroke and kidney failure will show very few symptoms until it is too late. Two conditions can be diagnosed with readings of blood pressure: hypertension, when a patient has high blood pressure, and hypotension, when a patient has low blood pressure. There are two pressure values that are measured by the device to determine a patient's blood pressure, the first value represents the pressure exerted on the arterial wall during contraction of the ventricle (systole) and the second value represents pressure exerted after the ventricle relaxes (diastole).
Describe how you addressed the challenge through your project.
When the heart beats, it sends blood rushing through your arteries to supply oxygen and other nutrients to the body. After supplying nutrients to the body, the blood is returned to the heart and it then beats again to repeat the process. When measuring blood pressure, two different measurements are taken, the systolic blood pressure and diastolic blood pressure. The systolic blood pressure is the maximum blood pressure which occurs when the ventricles contract and force blood to flow through the arteries. When the ventricles relax, they are filled with blood from the atria so that the process can be repeated. At this point the blood pressure is at its minimum, the diastolic blood pressure.
Preliminary Ideas and Design Alternatives
| Concepts | ||||||||||
Strain Gauge | Gel | Potentiometer | Pressure Sensor | Manometer | |||||||
Selection Criteria | Weight | Rating | Weighted Score | Rating | Weighted Score | Rating | Weighted Score | Rating | Weighted Score | Rating | Weighted Score |
Adjustability/Size | 10% | 4 | 0.4 | 2 | 0.2 | 4 | 0.4 | 3 | 0.3 | 2 | 0.2 |
Durability | 15% | 3 | 0.45 | 2 | 0.3 | 2 | 0.3 | 4 | 0.6 | 3 | 0.45 |
Ease of Use | 10% | 3 | 0.3 | 3 | 0.3 | 2 | 0.2 | 4 | 0.4 | 4 | 0.4 |
Low Cost | 25% | 3 | 0.75 | 1 | 0.25 | 3 | 0.75 | 4 | 1 | 1 | 0.25 |
Accuracy | 25% | 4 | 1 | 2 | 0.5 | 4 | 1 | 4 | 1 | 5 | 1.25 |
Simplicity | 15% | 4 | 0.6 | 2 | 0.3 | 1 | 0.15 | 4 | 0.6 | 4 | 0.6 |
Total score | 3.5 | 1.85 | 2.8 | 3.9 | 3.15 | ||||||
Rank | 2 | 5 | 4 | 1 | 3 | ||||||
Continue? | No | No | No | Develop | No | ||||||
After careful analysis of each of our proposed designs, we chose to develop our design using a pressure sensor. As the above Pugh matrix shows, the pressure sensor ranked the best in overall score, and ranked best in our most important design criteria, including low cost, accuracy, simplicity, and ease of use. The other possible design ideas were disregarded because they were either too expensive, too complex for our use, or simply not as strong of an overall idea. Below, you can read a more detailed description of each of our preliminary ideas. (Note: Each of the preliminary designs would utilize the same cuff, so that was not taken into account in this step.)
Strain Gauge
The strain gauge will be fused with the outer wall of the blood pressure cuff in order to measure displacement due to pressure changes. The sensor is relatively small at about 2 by 3/4 inches and is relatively inexpensive as well.
Electromagnetic Gel
A strain gauge composed of a tube of electromagnetic gel will be attached to the inside of the cuff. As the cuff itself displaces due to pressure changes, the gel compartment will lengthen resulting in a resistance change and therefore a voltage measurement. Leads attached to the tube will be connected to the USB port for a digital readout of this voltage.
Potentiometer
A potentiometer experiences a change in resistance which it converts into a voltage. It does this by linearly converting the length of the resistor to an output voltage. The way this would be attached would be either to connect it externally to the cuff with extension wires or to seal it into an opening within the cuff. They are quite small at about one square inch and are relatively inexpensive at about $3 each.
Pressure Sensor
The pressure sensor will be attached externally to the cuff to measure slight changes in pressure due to the patient's pressure pulse. Once the first sounds of Korotkoff occur the vibration will trigger the sensor to start taking a measurement. This sensor is extremely small at about one square inch and costs about $11 each.
Manometer
This device is approximately seven inches tall and would therefore need to be connected to the cuff externally. Overall it is quite expensive at about $12 for the manometer itself and $6 for the mercury within the tube. The fluid we would use would be an oil substitute for mercury due to mercury's high toxicity and lower availability. The oil will then rise/fall according to the pressure exerted by the artery.
(The idea for this was taken from a project done by Cornell University)
- The first design consideration is to keep the inflatable air pocket to a minimum size so that our oscillometric pressure measurement technique is as accurate as possible.
- One important design consideration is to prevent the outward expansion of the cuff during inflation. Therefore, we need the outer part of the cuff to have a higher resistance than the inside part of the cuff.
- The adhesive used must be airtight, waterproof, and strong enough to tolerate up to 200 mmHg of pressure.
- The weakness in our design occurs where the tubing enters the cuff. This area was reinforced with a washer on the inside of the cuff to distribute the force over a larger area.
- The cuff must be able to deflate relatively quickly (under about 30 seconds) to avoid discomfort in the patient. Therefore, we decided to use 3/8" tubing to enter the inflatable part of the cuff.
- 1/4" tubing was used to connect the t-connector to the pressure sensor.
- With Loctite vinyl
- With neoprene adhesive
- With 5-minute epoxy
- With super glue adhesive
- With super glue adhesive used on roughened surface to provide more adhesive surface.
Initial prototype with purple therapy band (high resistance) and yellow therapy band (lower resistance).
Programming with LabVIEW
The goal for the LabVIEW program is to fully automate the blood pressure reading. In order for this to happen, the program needs to be able to detect the oscillations in the pressure waves. These oscillations occur once the pressure in the cuff has dropped low enough that blood can resume flowing through the arm. This point marks the systolic pressure. As the pressure in the cuff continues to be released, one of the peak-to-peak values will be larger than the rest. When this occurs, this marks the Mean Arterial Pressure, or MAP. By taking the values of Systolic Pressure (SP) and MAP, the Diastolic Pressure (DP) can be determined by the following equation: MAP = (2/3)*DP + (1/3)SP.
As of now, the interface shows both a digital display for Current Pressure and a needle gauge, similar to a physical blood pressure cuff. Once the cuff is inflated, the operator is prompted to press the Continue button. Once pressed, the LabVIEW program begins to search for the oscillations in the signal transmitted by the DAQ device. This part of the code has yet to be added; however, the general idea behind it has been thought out. Once the pressure begins to be slowly released after the circulation has been cut off, small oscillations in the signal can be detected. A peak-to-peak finder block will be used to make sure that this oscillation is within a certain range (approximately 1 or 2 mmHg difference). These oscillations can only occur once the blood can resume flowing, thus marking the systolic pressure. A method for detecting diastolic pressure has yet to be determined.
The first half of the code filters the output and then takes this output (in volts) and converts it to mmHg. It does this by first eliminating the noise present when the sensor reads 0 mmHg (when compared to a commercial blood pressure cuff). Then, this value gets multiplied by the sensitivity of the sensor (in mmHg/V). The "OK Button" in the code refers back to the "Continue" button on the User Interface. This button is to be pressed when the cuff has been inflated and the user is ready to begin deflating the cuff. This button splits the array into two segments. The second segment is the one that should be searched for the maximum peak to peak value. This maximum peak to peak value marks the Mean Arterial Pressure. In this code, however, this step does not happen due to a multitude of errors that could not be solved. Solving one error lead to an error in the next step, and this trend continued up until the last day this project was due.
The original plan to detect the peak to peak values was to also get the "Valleys" and not just the "Peaks" as shown in the bottom portion of that image. By looping through the Peaks and subtracting their corresponding valleys, the difference between them (in other words, the peak to peak value) would be at a maximum at the point where the MAP occurred. An issue that came up when detecting the Peaks and the Valleys was that the blocks in LabVIEW were returning different numbers of Peaks and Valleys, when ideally they should be the same number. For example, it would find 11 peaks and 16 valleys. Without the correct number of both peaks and valleys, this method would not work as intended.
Problems
• Determining the greatest peak-to-peak value does not work. This value is integral to determining the mean arterial pressure. The method used to detect the greatest peak-to-peak value does not return the correct value.
• There's an occasional error that occurs with the DAQ unable to run without running it once first.
• Our pressure readings are consistently off by 4 of 5 mmHg due to the use of a new sensor, despite being re-calibrated multiple times.
• Any sudden change to the pressure, either inside the cuff or near the open-end of the sensor, will skew the readings. This includes moving the sensor while it is acquiring data.
COMSOL Design
E = F Lo
Ao ΔL
The main properties we wanted to model using COMSOL were the stresses exerted on and the y-displacement of the walls. We needed to know where stress on the walls was going to be significant so we could reinforce those areas of the cuff to make it as durable as possible and least likely to pop. We also wanted to know the y-displacement of the walls to ensure that the heavy resistance side was not displacing too much and that most of the force from the pressure in the cuff was focused on the center of the low resistance material which is where the cuff would be contacting the brachial artery area of the arm. I used a pressure of 160mmHg (equivalent to 21333 Pa) which probably be the max pressure put into the cuff. Some of the post-processing images and plots that were important were surface plots of y-displacement, von Mises stress, and first principle strain, and cross sectional plots for the top of the cuff.
It is apparent from this surface plot that the bottom half will displace a very small amount as the cuff pressure goes to its max of 160mmHg. This is exactly what we wanted because we would like all of the force to be focused on the arm which is in contact with the top half. In addition the displacement in the y-direction is shown to be focused on the very middle of the cuff where it needs to be greatest to cut off circulation in the brachial artery.
Quantitatively the plot shows a maximum displacement of 3.9cm in the y direction at the very top of the cuff's curvature. This should be more than enough expansion to cut off circulation in the arm. Also we have to account for the arm impeding the cuff displacement so we will see less than 3.9cm displacement at the top for 160mmHg but a little more than the show value (~0.5 cm displacement) to the left and right of the peak where the arm isn't in contact with.
The plot above shows the strain on the walls under pressure and we can see a large focused stress on the sides of the walls in the lower resistance material. This would cause the walls to bow outwards and begin peeling the two parts of rubber apart at the seal. When inflated to high pressures we found our first few prototypes experienced this rupture at the seam. The Loctite elastic adhesive and Epoxy that we used for these prototypes was not strong enough to resist the high stress levels at the walls. We finally tried superglue and roughed the surfaces of the rubber to increase adhesion power and this was strong enough to hold the cuff together at high pressures.
The max strain at the seams is shown to be 0.226 and while some materials weren't capable of withstanding this amount the superglue was and as long as we don't pump much past 160mmHg the superglue should hold just fine.
- The Cuff:
- The cuff is to be comprised of resistant elastic material. The outside part of the cuff will be stronger than the inside of the cuff so that the cuff will inflate down towards the arm to apply force, instead of out and away from the arm. Two layers of material will be used and sealed together to be air tight.
- The inflatable part of the cuff doesn't cover the whole arm, since it is unnecessary to apply pressure to the back of the elbow as this will not help cut off circulation.
- The Pump:
- The pump is going to be a traditional hand pump used on most blood pressure devices. We found hand pumps to be relatively inexpensive (about $3), to perform quite well, to be quite durable. The pump is able to pump air into the cuff, refill itself so it can pump more air into the cuff, and release air slowly through the use of a small valve.
- A noncommercial hand pump may be designed in the future, but for now a traditional one is being used.
- The Tubing:
- The pump is connected to normal plastic tubing and then to a T-connector where it is split to send air to the pressure sensor and the cuff.
- The tubing leading to the cuff and sensor are different sizes but one can fit over the T-connector and one can fit into the T-connector while still remaining air tight.
- The T-connector:
- This is just a simple brass T-connector found at any home improvement store.
- The Sensor:
- The sensor is an MPX5050DP Motorola differential pressure sensor.
- The desired pressure is measured relative to the atmospheric pressure.
- The sensitivity is 9.0 mV/KPa, so our sensor will supply 1.170 mV for each mmHg.
- Tubing to cuff intersection:
- To get air into the elastic material a small hole was punched in the material and a plastic connector was used to connect to the tubing.
- The plastic connector is wider at the bottom so it doesn't leave the plastic material and has a tip which pokes through the material into the tubing
- This may be reinforced later
- Adjustability of the cuff:
- The cuff will have Velcro attached to the sides of it to allow the cuff to be more adjustable.
- Another feature of adjustability is the small area of inflation on the arm. This allows both children and adults to use the same size cuff.
