BME100 s2016:Group2 W1030AM L3

From OpenWetWare

Jump to: navigation, search
BME 100 Spring 2016 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
Image:BME494_Asu_logo.png

Contents

OUR TEAM

Logan Luke
Logan Luke
Erik Dragar
Erik Dragar
Earl Brown
Earl Brown
Ishitha Jagadish
Ishitha Jagadish
Bailey Gasvoda
Bailey Gasvoda
Destinee Martin-Karim
Destinee Martin-Karim

LAB 3A WRITE-UP

Descriptive Statistics

Heart Rate (BPM)

Image:Screen_Shot_2016-02-10_at_11.35.47_AM.png


Temperature (Degrees Fahrenheit)

Image:Screen_Shot_2016-02-10_at_11.37.12_AM.jpg





Results

Image:Screen_Shot_2016-02-10_at_11.44.07_AM.png


Image:Screen_Shot_2016-02-10_at_11.44.11_AM.png


Image:Screen_Shot_2016-02-10_at_11.44.16_AM.png


Image:Screen_Shot_2016-02-10_at_11.44.37_AM.png


Two t-tests were conducted, because there were two groups (Spree headband data and gold standard data) in two different sets of data (heart rates and temperatures). The Pearson’s R correlation for the comparison of the scatter plot of heart rates was 0.47721. Since this value is very close to zero, this means that the data is not correlated. The Pearson’s R correlation for the comparison of the scatter plot of temperatures was 0.03634. Since this value is very close to zero, this means that the data is not correlated. Because the t-test conducted on the temperatures data yielded a very small p-value which was less than 0.05 (p = 3.94314*10^-22), the data was statistically significant. However, the t-test conducted on the heart rates data yielded a p value greater than 0.05 (p = 0.427116), meaning that this data was not statistically significant.






Summary/Discussion

The Spree Headband, while great in concept, features many design flaws. The headband is not aesthetically pleasing; most people would not be willing to wear the headband out in public or to the gym. The headband leaves marks on the forehead of the wearer, making sure everyone knows the person has one long after they take it off. This problem could be fixed by making the headband softer and more pliable, this would also make it more comfortable. The temperature sensor in the headband was unreliable at best, especially outside. The front of the headband was open to the air, giving temperature readings colder than the actual temperature of the wearer's body. This could be fixed by enclosing and insulating the Spree. Besides being colder than the actual subject, it would stubbornly read a consistent and inaccurate temperature. The Spree often disconnected from the phone. This could be avoided by keeping the phone close to the headband, but the Bluetooth technology could be updated. The heart rate sensor in the Spree was all over the place, high and low. This could be due the location of the sensor being on the forehead instead of near larger arteries, like in the wrist. The app was not user-friendly and occasionally crashed. The setup of the app needs to be reorganized to make it easier to use and start recording. The Spree headband was, overall, fraught with design flaws.




LAB 3B WRITE-UP

Target Population and Need

Target Population

The group's device will target populations of people in the United States, and other well-developed countries, who are prone to seizures and the family of those who are seizure-prone. The reason why well-developed countries are being targeted is because the device is estimated to have a relatively high cost. These seizure groups include people who have traumatic brain injury, epilepsy, infants with Phenylketonuria, and other patients who are at risk. The current estimate of people with epilepsy in America is 2.2 million people. There is a higher possibility of epilepsy developing in children and young adults. Therefore, the device will target from these two age groups up to elderly adults as well. The device will record the brain activity of the patient and send an alert to the patient and to the patient’s nearby family along with the emergency status based on what type of seizure it is. The patient will have the opportunity to set the device to contact the people of their choosing in case of seizure emergency. These contacted people could be the hospital or family.

Needs

Epilepsy affects people’s lives, because seizures are caused in epileptic patients due to abnormal brain activity. Therefore, there is a need for people with epilepsy to live a normal day-to-day life without having to worry or have family members and friends worry about unexpected seizures affecting routines. It would be very helpful for people with epilepsy to have warnings about their seizures before they experience them, because otherwise they can be in life-threatening situations. For example, if an epileptic patient was driving on the highway, a device that would warn them of an incoming seizure would be very helpful, because then the patient could pull over in order to not put their own life and other lives in danger of a road accident.

Problem Understanding Form

Problem Understanding Form


Customer Needs

Aesthetically pleasing: People with the need for this device would be more likely to purchase the device and therefore satisfy their need if it is aesthetically pleasing.

Battery life: In order for the device to function properly and for as long as possible, it must have a significant amount of battery life.

Non-invasive sensors: The device needs sensors in order to record the brain activity of users; however, the sensors should be non-invasive, because the current technology is invasive and therefore very expensive.

Wireless: The device should be able to function while not being connected to an input source (for example, electricity) for the sake of the user’s maneuverability.

Adjustable:The device needs to be adjustable so that it can fit a wide range of heads that are different shapes and sizes.


The engineering requirements for each customer need is demonstrated in the below problem understanding form:

Image:ER.png




Device Design

The SafeCap:

Image:hat.png

Image:diag.png





Inferential Statistics

Description of the Experiment

Part 1: The SafeCap


1. The SafeCap device will be placed on one mock brain that is hooked up to electricity that produces the same nerve cell signal patterns as a normal brain would.

2. The mock brain will then produce abnormal rhythms to see how well the SafeCap can monitor the abnormal brain waves.

3. The SafeCap will be hooked up to a computer that records the time in which the device identifies the abnormalities and the amount of time it takes to transmit the warning signal.

4. This test will also be run for at least 30 trials for the brain to indicate the precise amount of time is takes the device to transmit the warning signal upon identification.


Part 2: The EEG (Gold Standard)


5. The electroencephalogram test (EEG) will be conducted on one mock brain that is hooked up to electricity that produces the same nerve cell signal patterns as a normal brain would.

6. The mock brain will then produce abnormal rhythms to see how well the electroencephalogram test (EEG) can monitor the abnormal brain waves.

7. The EEG will be hooked up to a computer that records the time in which the device identifies the abnormalities and the amount of time it takes to transmit the warning signal.

8. This test will also be run for at least 30 trials for the brain to indicate the precise amount of time is takes the EEG to transmit the warning signal upon identification.


Application of Inferential Statistics


Based on the data sets that were obtained, one paired t-test was performed, because two sets of data (EEG versus SafeCap) were obtained from the same two mock brains that were prepared. Also, the null hypothesis of the experiment was that there is no difference in the average time of abnormal brain wave activity detection between the SafeCap and the gold standard EEG tests. Because a p value greater than 0.05 was found (p = 0.47), this means that the data obtained is not statistically significant, which suggests that there is no difference in the average time of detection of abnormal brain waves between EEG tests and the SafeCap. However, the analysis of the data from just the point of view of the p value is flawed, because actually 19 out of 30 times, the SafeCap detected abnormal brain waves at the same time or in a shorter time as the corresponding time value recorded from the EEG test. For the rest of the 11 out of 30 times, the SafeCap notified of an incoming seizure in no more than ten seconds of the value in which the EEG test notified of a possible seizure. Therefore, the statistical analysis needs to be corrected with respect to that (however, we did not learn how to do so yet). However, just considering comparing the two data sets within the 30 values, it can be concluded that the SafeCap is relatively as accurate as the existing EEG test detection method. Therefore, this product is validated. However, further statistical tests to confirm the validation of the product will be necessary.




Graph

Image:morestats.png


Image:yellow.png

Note: The highlighted yellow sections are the 19 out of 30 times that the SafeCap performed quicker than the EEG test.


Image:3Bbar1.png


Image:3Bscatter1.png




Personal tools