BME100 f2016:Group11 W8AM L3
From OpenWetWare
BME 100 Fall 2016  Home People Lab WriteUp 1  Lab WriteUp 2  Lab WriteUp 3 Lab WriteUp 4  Lab WriteUp 5  Lab WriteUp 6 Course Logistics For Instructors Photos Wiki Editing Help  
OUR TEAMLAB 3 WRITEUPDescriptive Stats and GraphTemperature Measuring Spree Device: Average (Gold Standard): 96.65269939 Average (Spree): 95.5308642 Standard Deviation (Gold Standard): 1.918462545 Standard Deviation (Spree): 0.8703782993 Count (Gold Standard): 326 Count (Spree): 324 Standard Error (Gold Standard): 0.1062538132 Standard Error (Spree): 0.04835434996
Average (Gold Standard): 98.08977 Average (Spree Band):98.94079 Standard Deviation (Gold Standard):23.03054 Standard Deviation (Spree Band):24.83749 Count (Gold Standard):303 Count (Spree Band):304 Standard Error (Gold Standard):1.3230699533 Standard Error (Spree Band):1.424527749
Significant? No
Inferential StatsTemperature Device Pvalue: 1.097E21 Significant? Yes! Pearson's rvalue: 0.1927980889 (Below minimum .4 required to conclude statistical correlation) Heart Rhythm Device PValue: 0.661778 Significant? No Pearson's rvalue:0.690806
Design Flaws and RecommendationsTemperature Device The spree device produces results significantly different from the gold standard device. The pvalue is much smaller than 0.05, meaning that the differences are in fact statistically significant. In other words, the data from the spree is very different from the gold standard. When comparing the two device measurements through the Pearson's value, which is 0.193, we see that it is much smaller than 0.4 (this is the minimum acceptable value showing correlation). From the data, we can infer that the spree device is inferior in quality to the gold standard and should not be sold in any market due to underlying technical problems causing the data to stray from those of the gold standard. Heart Rhythm Device The data provided in the test between the gold standard and the spree device have unquestionably similar data recorded. The pvalue is larger than 0.05, which tells us that the devices record measurements in no statistically different degree. Testing the reliability between them, we used Pearson's rvalue and need to see if it reaches 0.4. The rvalue found was 0.66 which explains that the spree device is similar to gold standard in terms of how its. Due to this statistical similarity, it would be totally acceptable or even advisable for this device to be sold on the market.
=Experimental Design of Own DeviceOur experimental group would be people of both genders, of all ages that are suffering from high cholesterol related vascular disease. The gender sample will be divided on the gender division in US which is 50.8% female and 42.9% male. The ages will be as follows: youth 018, early adults 1940, adults 4160, and elders 60+. Once again the numbers in our sample will be divided based on amount of people of those age groups in US currently, which is 25%, 40%, 20%, 15% respectively. On top of this, the sample will also include a variety of severity of the diseases, ranging from only slightly above healthy levels of cholesterol to those at life threatening levels which puts them at serious risk of suffering a heart attack or stroke. This will be the only group required in the study to test the efficacy of our device as a means of removing arterial cholesterol plaque as it will be a two tailed, paired, dependent test. Prior to the procedure, a cholesterol test will be taken in the area of the artery that our device will treat for each subject. Next, those same subjects will undergo our devicedependent surgical procedure to remove cholesterol in that same area of the artery tested (this is imperative as the only changes in cholesterol are to be expected in this area). Afterwards, a cholesterol blood test, an ELISA test, will be taken in the area treated for data analysis. a ttest will be conducted for this paired test to determine the pvalue between the two groups of results (pre and postoperation). A value of less than 0.05 will show that there is a statistical difference in the amount of cholesterol present in the treatment are (technically, this could also imply more plaque though a brief look at the graphed data could disprove this) and that our device is in fact effective. A Pearson's rvalue to test correlation between the postop and preop blood cholesterol levels could also be used show the effectiveness of the operation with our device. It would be preferable to see only a slight positive correlation to indicate lowered cholesterol levels among subject of all levels of initial cholesterol plaque in their treatment areas with only a slightly higher level after the operation in those with a higher initial level of cholesterol plaque. This is the realistic, best case expectation, though an rvalue of 0 would be ideal (which would indicate the same level of efficacy in those subjects with life threatening levels of cholesterol as in those with milder, lower inital levels of cholesterol in the same treatment area). An rvalue of 1 would imply the procedure had no effect on cholesterol plaque levels. Stats taken from 2010 Census (http://www.census.gov/prod/cen2010/briefs/c2010br03.pdf).
