BME100 f2015:Group13 8amL3

From OpenWetWare
Jump to navigationJump to search
BME 100 Fall 2015 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: Derek Scott
Name: Alexander Dasilva
Name: Luis Morales
Name: Ksenia Poujlivaia
Name: Shivanshi Shukla
Name: Parker Storry


Descriptive Statistics

Temperature Analysis

Gold Standard Mean: 97.27 Fahrenheit Standard Deviation: 1.024 Standard Error: 0.05274
Spree Headband Mean: 98.79 Fahrenheit Standard Deviation: 1.198 Standard Error: 0.06170
Inferential Statistical Values T-test: 9.5E-59 Pearson's R-Coefficient: 0.07486

Heart-rate Analysis

Gold Standard Mean: 96.44 BPM Standard Deviation: 22.95 Standard Error: 1.230
Spree Headband Mean: 96.70 BPM Standard Deviation: 23.42 Standard Error: 1.245
Inferential Statistical Values T-test: 0.80409 Pearson's R-Coefficient: 0.663709


Temperature Measurement Comparison Results
Graphical Comparison Between Temperature Measurements Between Devices

Heart-rate Comparison Results
Graphical Comparison of Heart-rate Data Between Devices

Note: A line and '*' above bars indicates statistical significance between the two measurements.


We were interested in whether there was statistically significant difference between the two different methods of measurement of the same variable, namely the Spree headband and a thermometer (for temperature), and the Spree headband and an oximeter (for heart rate). Thus, our group performed a two tailed, paired t-test on the statistical data.

Temperature Analysis
When comparing the data collected for the temperature, the P-value for the t-test comparison between the spree headband and the thermometer was 9.54E-59 which is nearly zero. This means that a difference of this amount would almost never occur by random chance and is due to a difference between the two devices. The reason for this difference is likely due to the arbitrary temperature reading given by the spree headband which was not measured in precise values, but was given as either level 1,2,3, or 4 while provided no connection between level and true temperature value. Also, with a Pearson's r-value close to zero, it is clear that there was little correlation between the values of the two devices.

Heart Rate Analysis
Based on the statistical values calculated from the data gathered by the various lab groups, it can be stated that the spree headband provided a fairly accurate measurement of the heart rate of the user. The P-value associated with the t-test for heart rate was 0.80409. This means that 80% of the time we could expect a difference of this magnitude to occur due to random chance. This probability is great enough that it is likely that no real difference exists between the measurements between the spree headband and the oximeter, but that the difference is likely due to random chance. Also, with a Pearson's r-value close to 1 (0.66), there is a strong correlation between the values reported by the two devices. Thus, the spree headband provided a fairly accurate measurement for heart rate.


Based on our findings and analysis of the gathered data, the Spree headband does seem to accurately measure the user's heart rate. It closely matched the data gathered using the pulse oximeter, and the difference between the two values was not found to be statistically significant. Thus, for measuring heart rate, the Spree headband is an acceptable product.

When it comes to evaluating and measuring temperature of the user, the Spree headband loses its usefulness. As its temperature is reported in arbitrary levels (i.e. 1, 2, 3, or 4) rather than individual numerical values in Fahrenheit, there was no useful way to gauge one's temperature accurately. Since there were so few levels, even during exertion the reported values rarely, if ever, changed. Also, for the statistical evaluation, temperature values used to calculate the statistics were assigned to the Spree temperature levels after the fact, and are not reflective of the actual temperature at the time.

The most evident design flaw with the Spree headband is that it provides arbitrary values for temperature without giving any insight as to what each of these value represent. As it is marketed as a useful fitness training tool, it would seem imperative that the user has access to a higher degree of biometric information than was possible with the Spree. What the Spree does offer is an easy interface with one's cell phone via an 'App' that might open the market for biometric monitors to a greater extent than the heart rate monitors typically used by elite athletes.

Since the Spree headband is useful solely as a heart rate monitor, this product would not be used by elite or serious athletes as less invasive methods, such as a band around the chest rather than across the forehead, and more accurate products are available. Thus, it would likely be used for those who are interested in their metrics, but not entirely concerned with specific values and accuracy.


Target Population and Need

The target population for the Swampimeter™ is elite athletes, or athletes who desire an accurate method of measuring their hydration, rather than relying on timing systems or waiting until they feel dehydrated. There is a need in the market for this device as currently, most elite athletes will rely on timing systems that indicate, after a duration of time has passed, that it is time for hydration. While this method works better than simply waiting until one feels thirsty to drink, it could lead to overhydration. Both overhydration and underhydration are serious concerns for endurance or professional athletes. Both will cause injury, and could lead to death, as a result of hyponatremia (overhydration) or dehydration related injury. Our group saw the gap in the market for a device that accurately measures the amount of perspiration from the individual athlete and will determine the precise time to consume water and electrolytes.

Device Design

The Swampimeter's design considerations are displayed in the following Problem Understanding Form:
Swampimeter PUF Form

Swampimeter Design Model
Swampimeter Design

Inferential Statistics

The Following Image Contains the Complete Data Relating to the Statistics Being Measured:
Data Table For Salinity Measurement

Analysis of Hydration of Athletes Over A 10 Kilometer Lab Conducted Test

Gold Standard (Sweat Test) Mean: 1405.0 cM Standard Deviation: 20.270 Standard Error: 3.701
Swampimeter Mean: 1409.5 cM Standard Deviation: 25.873 Standard Error: 4.724
Inferential Statistical Values T-test: 0.777 Pearson's R-Coefficient: 0.989

Concentration of salinity in the sweat was measured in centimolar units (cM).

Gold standard tests: The athlete is placed on a treadmill, and after each kilometer, a piece of absorbant cloth is used to collect sweat. This cloth is placed in a test tube, which is tested for amount of salinity in the sweat.

Swampimeter testing: While each subject is running their 10 kilometer test run, the swampimeter is placed around either of their calfs, or around their biceps. It is always testing the centimolar concentration of the sodium in the sweat, but readings were only taken at each kilometer mark.


Graphical Representation of Device Measurement Comparison

Linear Regression Model Comparing the Sweat Test to the readings of the Swampimeter