BME100 f2013:W900 Group8 L3

From OpenWetWare
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

File:BME103student.jpg
Name: Kendall Hickie
Role(s)
File:BME103student.jpg
Name: Connor McCoy
Role(s)
Name:
Kavin M. Watson
Role(s):
Design Critic
File:BME103student.jpg
Name: Samuel Butler
Role(s)
File:BME103student.jpg
Name: Ashley Ivany
Role(s)
File:BME103student.jpg
Name: student
Role(s)

LAB 3A WRITE-UP

Descriptive Statistics

Morning Group 8 Temp lab1.png Morning group 8 temp lab2.png Morning group 8 temp lab3.png Morning group 8 temp lab4.png

Morning group 8 temp lab5.png Morning group 8 temp lab6.png Morning group 8 temp lab7.png Morning group 8 temp lab8.png

Morning group 8 temp lab9.png Morning group 8 temp lab10.png Morning group 8 temp lab11.png Morning group 8 temp lab12.png

Morning Group 8 Temp analysis4.png

By comparing in the data in the data charts above, it is clear to see a significant difference between the sensor, and traditional oral readings. The data shows that the traditional oral readings have much better accuracy, and have a lower standard deviation. In comparison, the sensor gave data that was more sporadic, giving it a higher standard deviation. The data for the sensor also contained points which were so low, it would be impossible to be considered "accurate" unless the subject was dead. From this, it is clear to see that the sensor was not accurate, or reliable at all throughout this experiment.

Results

The following graph gives a better perspective of the data collected and the range of temperatures recorded from each device.
Morning Group 8 Temp Graph1.png

The following graph shows the average temperature of both the sensor and the oral thermometer. The "p" value derived from the T-Test shows that there is less that a 0.05 difference between the graphs, showing that there is a statistical difference in the data. The error bars show the standard deviation between the two data sets. Morning Group 8 Temp graph 23.png

The following graph represents the correlation between the two devices. The Pearson's r is 0.16739506, which represents poor correlation between the two devices.
Morning Group 8 Temp lab pearsons r1.png


Analysis

The T-Test value for this set of information is 2.9459E-80. Since this value is less that 0.05, there is a statistical difference between the two values. This statistical difference shows the significant invalidity of the sensor. Its average of 95 degrees shows that there are a number of outliers held in the data. These outliers skew the data drastically, proving the sensors unreliability. However, the oral readings prove to show more validity, due to its lower standard deviation and more accurate readings. Through the data, it is shown that the oral readings never drop below 97 degrees, where as the sensor readings drop as low as 92 degrees. A reading of 92 degrees Fahrenheit would only prove to be accurate if a subject was suffering hypothermia. Since none of the subjects in the experiment suffered through this, it is safe to assume this data is invalid. The Pearson's R value for this set of data is 0.16739506, which shows the correlation between the sensor and the oral thermometer. This number describes the poor correlation between the sensor and the oral thermometer.




Summary/Discussion

As one can see above, the Body Temperature per Trial graph shows that the temperature sensor displayed temperatures that were consistently lower than the standard oral thermometer. Only rarely did it approach the same temperature as the oral thermometer. When considering the results of the T-test, one can see a marked statistical difference in measured temperature between the sensor and the control, a standard oral thermometer. The implications of these results along with the flaws of, and ways to improve the device are described in more detail below.

To begin with, the device was difficult to turn on. It required one to use their thumbnail or some other flat object. This could be to prevent the subject from inadvertently shutting the device down, but there are already other, easier to access, methods to prevent shutdown in many other electronics. Next, the device would not communicate over Bluetooth with the subject's iPhone(tm), this could be due to user error, or a defect in the subject's iPhone(tm), but other subjects were having the same issues. When one could get the product to operate correctly, it took a significant amount of time for the unit to calibrate, much longer than a simple, under the tongue, thermometer would take. Further, one must rely on the adhesives that come with the device to attach the device to the subject's arm. The adhesive wasn't very comfortable to wear, it was irritating to the skin, especially after left on for an extended period of time. With such a bulky device, it may be more comfortable to have an arm band built into the device.

Finally, the device takes an extremely long time to reach some semblance of accurate readings. When it has reached its maximum accuracy it is still off by 2 to 4 degrees Fahrenheit when compared with a reliable mouth thermometer. For everyday situations, this error may be acceptable, but in life threatening situations where fevers are involved, this error can be deadly. Even when it has reached its best measuring capabilities, its mechanical unreliability comes out, and it shuts down spontaneously.

The best paths to take to improve this device is to decrease the size of the device, small enough to fit under a patch that could be twice the size of a nicotine patch. Make the edges more round to improve the comfort of the wearer. Make the power on indicator light more visible, especially to the subject who may be wearing it. Perhaps, increase the surface area of the metal contacts to decrease warm-up time of the device for quicker accurate measurement. If one cannot decrease the size of the device for any reason, it may be best to include a Velcro(tm) arm band to make the device more comfortable to wear, and to keep the device more secure.

LAB 3B WRITE-UP

Target Population

Ninja Med's Thermopatch was targeted towards the elderly and athletes. These separate groups were targeted because the Thermopatch allows the elderly to have more freedom allowing them to care for themselves, and athletes are able to keep track of their body temperature during training and games. Ninja Med's Thermopatch provides great convenience for both of these target populations.





Device Design

Below is a diagram of the Thermopatch. The Thermopatch is 0.25 inches thick to provide comfort for users, ensuring users hardly notice it while on. The device comes loaded with a pen on-off switch, this ensures the device stays on throughout the day. Due to this, the device should only power down by users intentions. With a green arc shaped LED light, users can easily determine if the device is turned on, or off. The device also includes adhesive patches similar to band aid's or nicotine patches. These patches are used to keep the device to the user's skin, ensuring the device records accurate body temperature.




Device Description.png

Inferential Statistics

Morning Group 8 Thermo Patch Ninja Med data1.png Morning Group 8 Thermo Patch Ninja Med data2.png Morning Group 8 Thermo Patch Ninja Med data3.png Morning Group 8 Thermo Patch Ninja Med data4.png Morning Group 8 Thermo Patch Ninja Med data5.png Morning Group 8 Thermo Patch Ninja Med data analysis 1.png
The data above shows the comparison of traditional oral thermometers. With the Thermopatch, users are able to obtain more accurate data. Results show that the Thermopatch not only show more accurate data through lower standard deviation, but the data contains less standard error. By seeing how the Thermopatch compares to traditional oral readings, one can easily see that the Thermopatch is the better product to measure one's body temperature.





Graph

The graph below displays the data and shows the reliability of the ThermoPatch compared to the reliability of the oral thermometer. The graph portrays that there is more reliability from the ThermoPatch than the oral thermometer. Morning Group 8 Thermo Patch Ninja Med data graph 12.png

This graph displays the average temperature for each device. The p value is 1.81555E-26 which is less than 0.05 which shows that there is a statistical difference between the two devices. Morning Group 8 Thermo Patch Ninja Med average graph1.png

This graph displays the correlation between the two devices. The Pearson's r value is 0.11219213 which is close to zero and shows that there is almost no correlation between the two devices. Morning Group 8 Thermo Patch Ninja Med corelation graph 12.png