BME100 f2016:Group7 W1030AM L3

From OpenWetWare

Jump to: navigation, search
BME 100 Fall 2016 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



Name: Koop Bills
Name: Koop Bills
Name: Tyler Lent
Name: Tyler Lent
Name: Maria Hanna
Name: Maria Hanna
Name: Omar Maranon
Name: Omar Maranon
Name: Israel Zaldivar
Name: Israel Zaldivar


Descriptive Stats and Graph

Spree Band (thermometer)
Mean = 95.53086
Standard deviation = 0.870378

Spree Band (heart rate monitor)
Mean = 98.93069
Standard deviation = 24.87795

Mean = 96.6527
Standard deviation = 1.918463

Pulse ox
Mean = 98.08977
Standard deviation = 23.03054

Graphs for Averages:


Pearson's r

Between Spree and Gold Standard thermometer r = 0.172959

Between Spree and Gold Standard pulse ox r = 0.633239

Graphs for Pearson's r:


Inferential Stats

Thermometer x Spree p = 2.82 * 10^-21
Pulse ox x Spree p = 0.48

Both data tests were run with a two tailed test with the null set at p > 0.05. The first test (Spree thermometer x Gold Standard) produced a p= 2.82*10^-21 indicating significant difference, so here we can say that there is a definite issue with the temperature calculation. The second t-test (Spree pulse ox x Gold Standard) produced a p = 0.48 this indicates that there is no real statistical difference so we can say that the spree band works fairly well for determining heart rate!

Design Flaws and Recommendations

Some minor design flaws would be the intended location for the product to be placed at i.e. the head, this is an obvious issue because not every person (if any) would enjoy wearing something like this on their heads. Also the issue with giving false readings for temperature. If we were to give advice for this company we would recommend something more like a bracelet in design or an anklet, pretty much anywhere but the head area. Most importantly we would recommend that the company try and figure out why their product is spitting out false readings for temperature!

Experimental Design of Own Device

For our device we know that we need to run a few different tests against several "gold standards". The first set of tests would be out of patient tests (implying that we would run these without putting the catheter into the patient. First we would have to see if the pressure sensor functioned properly by testing it against a gold standard. Next we would have to ensure that the sensor worked even in wet conditions which means testing the pressure sensor against another gold standard in some type of fluid. Moving on from the sensor, we would have test whether or not our catheter would disallow fluid to flow past it, this test would not need a control group per say as leakage or lack there of is the null that we are testing against. Next we would have to see if our method of releasing fluid works, this too could be done without a control group. Next we would have to check our added watch (its purpose is to give early warnings when fluid reaches a certain pressure) and confirm that it can accurately receive data from our catheter and that once the pressure sensor mounted in the catheter reached a certain amount, that the added watch gave some type of audible notice, (in this case we would be testing to ensure that the watch vibrated in increasing amounts, that it beeped loudly like it was supposed to and that once the pressure reached a certain amount in the bladder the safety feature that we added i.e. to release all liquids without a prior button press, goes through smoothly), we would also have to check to see that the relesase button we have on the watch did in fact release the liquid in the bladder that was being held back. After confirming that our device works correctly we could move on to trials inside an animal. At this point we essentially would be repeating all the trials that we did beforehand but now we would add in a test to ensure that no additional microorganisms would be getting introduced into the host. After completing testing in this area we could move on to increasing numbers of human trials (starting at a test with a dozen or so people and then moving up until the test is done on at least a few thousand individuals) and conclude with marketing and sales!!

Personal tools