BME100 s2014:W Group12 L3

From OpenWetWare
Jump to: navigation, search
Owwnotebook icon.png BME 100 Fall 2013 Home
People
Lab Write-Up 1 | Lab Write-Up 2 | Lab Write-Up 3
Lab Write-Up 4 | Lab Write-Up 5 | Lab Write-Up 6
Course Logistics For Instructors
Photos
Wiki Editing Help
BME494 Asu logo.png

OUR TEAM

Name: Sara Jerez
Name: Austin Doyle
Name: Anthony Zlaket
Name: Jacqueline Stokes
Name: RJ Parkinson


LAB 3A WRITE-UP

Descriptive Statistics

Average systolic blood pressure reading using blood pressure cuff - 118.05

Average systolic blood pressure reading using Omcron watch sensor - 112.2833

Average pulse oximetry reading using pulse oximeter - 87.9625

Average pulse oximetry reading using Omcron watch sensor - 86.74167

Average temperature recorded using oral thermometor (Fahrenheit)- 97.07729

Average temperature recorded using sensor - (Fahrenheit) 96.49492



Results

P value's of the two tailed paired t-test's

Systolic blood pressure reading's of the blood pressure cuff compared to Omcron watch sensor P= 5.19E-5

Pulse Oximetry reading's of pulse oximeter compared to Omcron watch sensor P= 0.105127

Temperature readings (Fahrenheit) of oral thermometer and sensor P= 2.71E-6




Analysis

Using an alpha level of .05, the p values measured for the systolic blood pressure readings and the temperature readings are well below alpha therefore we reject the null hypothesis and can assert that the average blood pressure reading of the Omcron watch sensor and the temperature readings of the RAIIN sensor are not equal to the average values measured by the gold standard devices.

Comparing the pulse oximetry readings from our gold standard device to the readings obtained by the Omcron watch sensor the p value was .105127. Because this value is greater than the alpha level of .05 we cannot reject the null hypothesis and can not conclude that the Omcron oximetry values varied significantly from the gold standard oximetry values.




Summary/Discussion

The first device that our group tested was the RAIIN sensor, a sensor for measuring body temperature. This was tested by comparing it to an already proven device in the oral thermometer. In order to test this device we performed an experiment in which we took the body temperature of one group member with both devices simultaneously every two minutes for a total of five times while he was sitting at rest in a chair. The RAIIN sensor was measured by placing it and securing it under the arm of our group member. Through a smart phone app we used Bluetooth to record the temperature of our teammate. After these five readings we all went outside and sat in the sun. We then recorded his body temperature twelve more times with both devices simultaneously every five minutes. After these readings we returned to our lab and had our group member take his body temperature five more times, every two minutes. At the conclusion of our experiment we ran a paired T-test and Pearson’s R data analysis to ascertain as to any similarities between the two products. We were able to determine that there were differences in the data. The Pearson’s R was 0.3885 +/- 58.77. The one tailed T-test was 1.36 E-6 and the two tailed T-test was 2.71 E-6. This shows that the differences were not considerable but still not enough to validate the RAIIN sensor. Perhaps the readings were not similar because of the many faults that the RAIIN sensor has. A few problems we had were: first, it was difficult to place the sensor under our teammates arm, it took more than 10 minutes for him to secure it close enough to his skin so that it could read his temperature; second was that it gave us a temperature well below what his body temperature was, it took a few readings for it to rise to a normal body temperature level; and thirdly, the phone and device disconnected a few times and thus this could have affected the true body temperature reading. In order to improve this device the most important addition would be something to more securely hold it to a person’s body but at the same time remaining un-invasive. Perhaps this could be done by adding sticky arms to each side. This would help keep the sensor in place throughout the time it takes to read the temperature and it would also be quick and easy.


The second device that our group tested was the Omcron Watch Sensor. We performed this test by comparing it with two proven devices, the Blood Pressure Cuff and the Pulse Oximeter. Both of these products are proven in the medical field to give accurate pulse measurements. We performed these tests by measuring the pulse of one member in our group every two minutes, with all three devices simultaneously, five times. These five measurements of the pulse were taken while our group member was at rest, sitting in a chair in the lab. Then we took his pulse twelve times every five minutes after he ran up and down the stairs twice, causing the blood pressure to increase. After those five measurements we returned to the lab and obtained his pulse five more times, while at rest. Upon reviewing this data it was clear that his pulse went up after running; all three devices were able to record higher pulses. There was, however, a large amount of inconsistency in the readings. Rarely at one time did all three devices read the same pulse. We then performed a paired T-Test and Pearson’s R data analysis between the Watch Sensor and Pressure Cuff, followed by the same between the Watch Sensor and the Pulse Oximeter. Upon reviewing the data we were able to conclude that the Watch Sensor and Pressure Cuff were more similar in their pulse readings. The Pearson’s R was 0.1055. The one tailed T-test was 2.6E -5 and the two tailed T-test was 5.19E-5. These ratings show that there are differences between the two devices but not considerable differences. This cannot be said of the paired T-test and the Pearson’s R data analysis between the Watch Sensor and the Pulse Oximeter. This data showed that there was a considerable difference between the two devices. The one tailed T-test was 0.0526 and the two tailed T-test was 0.105. The Pearson’s R was 0.505. Perhaps this data occurred in this fashion because of the similarities in function between the Watch Sensor and the Pressure Cuff. Both measure the pulse by being wrapped around the arm near major arteries. Both of these are subject to like inconsistencies. For example they could both be wrapped loosely or placed too far from the artery. The Pulse oximeter is a little different for the fact that it is placed around the tip of the finger. Since the Pressure Cuff is the longest lasting device and has the most history it could be that the Watch Sensor functions correctly and the Pulse Oximeter is the faulty device. Since the Pulse Oximeter is proven as well this may not be the case. In order to improve on the Omcron Watch Sensor perhaps the most impacting difference could be by assuring that it is wrapped tightly around the wrist like the Pulse Oximeter is around the finger.





LAB 3B WRITE-UP

Target Population and Need

The StressTech device targets college students who find themselves stressed frequently. This device allows students to monitor their stress levels and use that data to manage their stress and work more efficiently.




Device Design

BME Device.png

Headband part: Adjustable; able to fit snugly into hats, headphones, etc. Comfortable fit. Sensors: Very small; do not add any discomfort; reads the brainwaves and stress signals to calculate an easy-to-understand stress level.




Inferential Statistics

To compare the effectiveness of our StressTech monitoring device we collected sleep EEG readings from a healthy group of individuals and compared that to readings collected using the StressTech monitor. Each group consisted of 35 healthy individuals, we used two separate groups because it is impossible to collect data using an EEG device and the StressTech device at the same time because the devices would interact and create artifacts in the EEG readings. To quantify EEG readings the StressTech monitor uses an algorithm to compute relative wave type frequencies to infer whether the wearer is in a stressed state. For the purposes of this comparison the device was measuring sleep data because the sleep patterns produced by participants are likely to be similar, versus waves generated in an awake state which vary dramatically depending on the environment.

Average Relative frequency of alpha waves measured with EEG- 0.325543

Average Relative frequency of alpha waves measured with StressTech monitor-0.325829

T-test - P value=0.966192 Because this P value is very large and larger than the standard alpha level 0.05 we can not conclude that the average relative frequency of alpha waves measured by the StressTech device is significantly different from the readings collected from a standard EEG device.



Graph

Graph.JPG