BME100 s2014:W Group7 L3

From OpenWetWare
Jump to: navigation, search
Owwnotebook icon.png BME 100 Fall 2013 Home
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
Wiki Editing Help
BME494 Asu logo.png


Name: Haley Sivertson
Name: Elizabeth Hansen
Name: Velia Butruz
Name: Matthew Welz
Name: Valerie McDonald


Descriptive Statistics

In this experiment, we compared the results between two different devices to determine their accuracy. Temperature and blood pressure were the two variables used in this experiment. For temperature, the two devices compared were an oral thermometer and an external thermometer. For the blood pressure, a wrist watch and blood pressure cuff were compared. Finding the average through excel, we were able to plot a correlation graph and a t-value test graph with the error bars as standard deviation. We also determined the significance of the study with its p-value.


In the study for blood pressure, the average value for the blood pressure cuff was 118.27 mmHg and the watch sensor average was determined to be 112.31 mmHg which gave us a standard deviation of 18.95 and 13.54 respectively. Using a t-test to compare the two different groups, we were able to determine the value to be 2.97*10^(-5). This proved its significance when comparing to the P-value of 0.05. The value was below the p-value therefore it shows a statistic significance. To determine the type of correlation in the experiment, Pearson's R correlation's value was given at 0.15. Because this was between -1 and 1, it showed a slightly positive correlation. Although, it was near 0, therefore there was little to no correlation meaning that one of the devices is not calibrated correctly.

In the study for temperature, the average value for the oral thermometer was 97.07 degrees Fahrenheit and the sensor determined an average temperature of 96.50 degrees Fahrenheit which gave us a standard deviation of 1.07 and 2.03 respectively. Using a t-test to compare the two groups, the t-value resulted in a value of 3.7*10^(-6). This proved its significance when comparing that to the P-value of 0.05. The value was below the p-value therefore it shows a significant statistic. To determine the type of correlation in the experiment, Pearson's r correlation's value was given to be 0.21. Because this was between -1 and 1, it shows a slightly positive correlation. However, it is still so close to 0 showing little to no correlation.

The standard deviation of the blood pressure cuff was larger than that of the watch sensor. The graph of the data in the temperature showed that the oral thermometer showed a smaller deviation than the RAIIN sensor.


In both studies, according to the data and results that was collected, it was determined that there was a significant source of error in one of the devices. Considering the r-correlation was so close to zero, it gave the graph no correlation. Due to the lack of correlation, it shows that one of the device was faulty. When measuring temperature, it is unlikely that the oral thermometer was the inaccurate device as it is considered to be the gold standard device. Therefore, it is likely that the external RAIIN sensor was the source of the data's inaccuracy. When referring to the results regarding blood pressure, the source of inaccuracy was likely to be from the Omron Watch Sensor rather than the blood pressure cuff.

The difference in deviation between the blood pressure cuff and the watch sensor shows that the watch sensor was more inaccurate because we were studying a large range of blood pressure. Therefore, there would be a larger range of results as evidence by the blood pressure cuff.

However, with the temperature, there should be less deviation due to the reasoning that the average human body temperature is 98.6 degrees Fahrenheit. The sensor shows a larger deviation that the oral thermometer, therefore the sensor is more inaccurate than the oral thermometer.


In this experiment, it was determined that the Omron Watch Sensor was an inaccurate source when determining blood pressure and was deemed not yet fit for public use. There were various flaws in the measurement process. There were multiple trials when the watch would either stop working and turn off or would say there was an error when trying to determine the subject's blood pressure. The subject's arm position also had to be positioned at a specific level, with the subject holding their arm up in one place. A way to improve the Omron Watch Sensor is to make the sensor more accurate. The company would make a more adjustable strap for a high pressure to allow for a higher measurement. Another problem that could be fixed is the glitches and bugs within the system. Also, the battery problems could be looked at; possibly a better battery.

The RAIIN sensor was determined to be inaccurate when testing a subject's body temperature. There were constant problems with the bluetooth and the device always had to be close to the cellphone which read the data. It turned off multiple and during the testing process and could take up to five minutes to turn back on. Because the device was not invasive, it could only determine an external temperature rather than a core temperature. To improve the RAIIN sensor, the company would need to look at improving the Bluetooth connection between the phone and the device or find a better form of connection. In order to get the most accurate measurement, one would need an invasive thermometer which will directly determine one's inner core temperature. With this being a noninvasive determination, it would be difficult to improve on this device.


Target Population and Need

For this new product, the targeted audience are athletes and any individual that participates in physical activity. This product is for anyone who would like to monitor their electrolyte count during physical activities to prevent dehydration and other health risks from heat and water loss. There is a need for this product because of the number of dehydration-related incidents. It will make the user aware of their condition. With this product, it will be successful in preventing many from becoming incredibly dehydrated and preventing dehydrated-related deaths.

Device Design

The product will be a microchip in the shirt that is engineered to monitor the subject's electrolyte levels through various sensors that are attached to the chip throughout the entire shirt. The logo on the front of the shirt will contain the microship and will change color to indicate subject's hydration levels. The logo is white when levels are normal and will slowly darken as electrolyte levels deplete.

Inferential Statistics

In the experiment, the new device was compared to that of a gold standard device. This test was based on measuring sodium levels in the body after various periods of physical activity. The sodium levels were monitored in four different stages with 10 subjects in each stage: at rest, after 60 minutes with no rehydration, 120 minutes with no rehydration and 180 minutes with no rehydration. The average of both the gold standard device and the new product was determined for each stage. There was a t-test applied to the different groups and the results of each test varied between 0.42 and 0.83 respectively. Therefore, this concluded that as the calculations approached zero, the new device was determined to be very accurate. To measure if the new device was accurate with the t-test results, a Pearson's r value was ran. This test gave a value of 0.999, almost 1. This means that the device is very similar to the gold standard test, giving a 99% accuracy in reading.


The bar graphs shown are the results of the averages between the gold standard and the device with the standard deviation being the error bar. Each graph represents a different time trial with one being at rest and then at intervals of 60 minutes. The line graph shows the results from the Pearson's r test showing two lines almost correlating with one another, proving the new device's accuracy.