BME100 f2014:Group20 L3

From OpenWetWare
Jump to navigationJump to search
BME 100 Fall 2014 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


Karthik Nambiar
Ryan Giudice
Jeremy Ellis
Connor Seiferth
Name: Bjorn Houman
Name: Your name


Descriptive Statistics



Because there are two groups being studied, the appropriate test is a t-test:

Pre-walk Gold vs. Spree [Heart Rate] p = 0.4287, not statistically significant

Walk Gold vs. Spree [Heart Rate] p = 0.9460, not statistically significant

Post-walk Gold vs. Spree [Heart Rate] p = 0.6644, not statistically significant

However, for all measurements comparing the gold standard thermometer to the Spree our p value was less than .05 and thus statistically significant.

This would indicate that while the Spree headband does not produce statistically significant values from the gold standard for Heart Rate, it would for the thermometer, emphasizing the conception that the temperature monitor is not incredibly accurate. This seems to suggest that the Spree headband provides comparable results to the heart rate monitor, and could reliably measure heart rate.


Some flaws with the device are noticeable that prevent it from excelling. For example, the sensor headband is very tight-fitting on the user's head. If the user has a large head, it can often leave marks on their head and cause serious discomfort. Not only is this a negative in terms of comfort for the customer, it also is harmful when considering the actual measurement. Constriction of the head (especially in such a prominent manner) can seriously increase blood pressure due to this constriction, thus removing the possibility of measuring blood pressure accurately. Similarly, the temperature measurement can be affected by the extra work that the head must now do to keep blood flowing in those regions. Another problem is the bluetooth signal. If this product were ever to get seriously popular, there would be a large chance that measurements might become difficult, as the proliferation of iPhones looking for the specific sensor necessary might have some problems. Similarly, the lack of Android functionality removes a sizable market from using the device. A final note might be the vagueness and lack of specificity of the temperature sensor. The temperature sensor measures with 4 values, "1", "2", "3", and "4". These ranges house a large variety of temperatures, which might make it difficult to get any relevant information from the sensors on the slight variations in temperature that are relevant.

There are some simple fixes that could be made to help avoid the problems we experienced. With regards to the constriction on the head, a larger strap could be used. It would thus be on the customer to make sure they understood what level was too tight and what level of tightness would not allow them to have proper tightness. Similarly, the mesh pattern on the rubber evident in the Spree band might make it look more attractive and have interesting structural properties, in reality, it hurts the customer in situations where the band is placed on. In general, a simpler design with the bare basics would allow the customer to work out without thinking too much on the actual headband. A device that resembles current headbands, with little to no rubber would be important, and rubber would be used to protect important parts of the product. Android functionality could be implemented depended on programming expertise in the division, but would improve customer base. One way to circumvent the problem with bluetooth would be to force users to create an account, for which they could register their devices. Only phones that had the headband registered to their account could receive the data necessary from the headset. It would make it slightly less customer-friendly in the ability to pass along headbands to each other, but that idea is not widely prevalent, and those at such a level to trade headbands would find no issue in signing into an account that does little more than monitor which headbands can be accessed. There are two easily-seen reasons on why the company used simple temperature statistics that did not tell much. One could be the idea that this statistic was not particularly interesting to customers, and could clog up space, and the other would be that their sensors are not complex enough to get a reasonable degree of precision. The former is debatable, but in general, these statistics would not be carefully noted during exercise, simply afterwards. Not putting in these events due to clutter seems silly considering the technology is present. If the technology is not present, and there is demand, there are many suitable, relatively small temperature sensors for a higher price.


Target Population and Need

The target population for this product is anyone who feels drowsy when woken up by an alarm clock. This encompasses almost everyone in the modern world (i.e. students, parents, workers) as waking up on time is essential to almost everyone's daily duties. Our product monitors brainwaves and states of sleep, it then uses that data to determine when to wake you up within a set time frame in which you will feel the most awake and energized. The need for this product is that most people feel a sense of drowsiness and lack of energy even after getting a full night's rest, leading to a slow start to the day and loss of productivity. The product would allow people to avoid these feelings by adjusting the time in which they are awoken. This would lead to people being more productive and ready to take on the challenges of the day.

Device Design

Inferential Statistics

This data set refers to the data used for the t-test

The data set below refers to the data set used for the Pearson's R:

This shows the t-test as Prob>abs(t)

This shows Pearson's R