BME100 f2013:W900 Group17 L3
From OpenWetWare
BME 100 Fall 2013  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 3A WRITEUPDescriptive StatisticsIn order to describe this data set, statistical values were taken from both the data set of values collected inside and the data set collected outside as well as the entire data set of the product test. In total, temperature readings from the inside and outside were taken by 15 groups for a total of 15 trials. For the subset of temperature readings taken with the thermometer within the inside data set the average reading was 97.57533333 degrees accross with a standard deviation of 0.778049244 and a standard error of 0.063527455. Then for the subset of temperature readings taken with the RAIIN sensor within the inside data set the average reading was 94.48466667 degrees with a standard deviation of 2.064618419 and a standard error of 0.168575388. It can be observed here that the average reading for the sensor is both on the bottom end of what a normal human body temperature should be and that the standard deviation of the standard reading is high compared with a much lower standard deviation for the readings taken with a thermometer. Now the subset of temperature readings taken with the thermometer within the outside data set the average reading was 97.49666667 degrees with a standard deviation of 0.897227262 and a standard error of 0.066875372. Then for the subset of temperature readings taken with the RAIIN sensor within the outside data set the average reading was 95.57944444 degrees with a standard deviation of 1.479494751 and a standard error of 0.110275028. Unlike the inside data set for the sensor, the outside data set for the sensor displays a more reasonable average in regards to what a normal human body temperature should be and also a somewhat lower standard deviation. However, the average reading and standard deviation is still quite low and high respectively when compared to that of the data set of thermometer readings.
ResultsAnalysisTaking a T test comparing the thermometer readings set of results with the RAIIN sensor set of results, both contained within the inside data set, the result was 1.22112E39 which signifies that there is a high level of confidence in a difference existing within the two data sets. Furthermore, upon taking the Pearson's r correlation test it was found that there is only a positive correlation level of 0.165252944 between the temperature readings data set and the RAIIN sensor data set within the context of the larger inside data set. This is not a high correlation and so it is safe to conclude that the two data sets are significantly different within the context of inside readings taken. Then taking a T test comparing the thermometer readings set of results with the RAIIN sensor set of results, both within the outside data set, the result was 4.04259E38 which signifies that there is a high level of confidence in a difference existing within the two data sets of readings taken outside. Also, upon taking the Pearson's r correlation test it was found that there is only a positive correlation level of 0.225273129 between the temperature readings data set and the RAIIN sensor data set within the context of the larger inside data set. This is not a high correlation and so it is safe to conclude that the two data sets are significantly different within the context of outside readings taken. However, since both the inside data set and the outside data set led to the same conclusion upon running the statistical tests, it is safe to predict that upon running similar statistical tests on all readings, inside and outside, the results will be similar. In fact this is found to be true because when taking the outside data set and inside data set together with the new set separation being solely the set of readings taken with the thermometer and the set of readings taken with the RAIIN sensor, a ttest result was found, 4.18844E72, which again indicates that there is a significant difference between the readings taken with the thermometer and those taken with the RAIIN sensor. And with a correlation of only 0.167395055 found between these two sets, it is safe to further strengthen the conclusion that the readings taken with the two devices are not at all similar.
Summary/DiscussionThe results showed that there is a low pvalue from the ttests, showing that there exists a difference between the values collected by each device. There were various errors with the RAIIN device itself. Phone signal and interference from other phones gets in the way of measuring one person's body temperature. In addition, there should be a different name for the Bluetooth for each device so that the user won't confuse one device's Bluetooth with another. Another problem with RAIIN is its indecipherable, ambiguous instructions. It is difficult to determine where the sensor should be placed on the arm with just a simple picture. The instructions should be more specific so that the user gets the most accurate results. Lastly, the RAIIN sensor is easily affected by the temperature of the environment, increasing in its measurement of body temperature outside and decreasing inside. The difference was more apparent inside as the temperature inside a room is further from normal body temperature than outside in the sun. To fix this, there should be some type of insulation system so that the environment temperature doesn't affect the sensor readings. In conclusion, the RAIIN sensor iPhone app was found to be unreliable in measuring body temperature.
LAB 3B WRITEUPTarget Population and NeedThis new product by JLAN care seeks to fill the need that parents with young children have to monitor their kids's health by providing a new way to effectively and constantly measure body temperature. This product is not only significant because of the need it fills, but also because of the efficient way in which it does so as it is smaller, lighter, and in many cases more accurate than traditional devices such as thermometers. In addition, it can provide a constant data stream which parents can view on their phones anywhere. Seeing as it is a constant monitoring device for body temperature, it is also conceivable that a secondary target population might be athletes. In addition, it can always be used in the traditional way to check for high temperature from anyone. However, this being the case, the main target population would still be parents with young children.
Device DesignThis device is smaller than the RAIIN sensor tested in Lab 3A, decreasing the discomfort from the size the RAIIN sensor was before. To appeal to the child population and encourage children to wear them, the armband will come in a variety of colors, so children can pick their favorite color to wear to complement their tastes.
Inferential StatisticsTaking a T test comparing the thermometer readings set of results with the JLAN Armband set of results, both contained within the inside data set, the result was p = 0.239875, p > 0.05 which signifies that there is a low level of confidence in a difference existing within the two data sets. Furthermore, upon taking the Pearson's r correlation test it was found that there is only a positive correlation level of 0.974094 between the temperature readings data set and the armband data set within the context of the larger inside data set. This is a high correlation and so it is safe to conclude that the two data sets are significantly similar within the context of inside readings taken. Taking a T test comparing the thermometer readings set of results with the JLAN Armband set of results, both contained within the outside data set, the result was p = 0.141642, p > 0.05 which signifies that there is a low level of confidence in a difference existing within the two data sets. Furthermore, upon taking the Pearson's r correlation test it was found that there is only a positive correlation level of 0.982099 between the temperature readings data set and the armband data set within the context of the larger outside data set. This is a high correlation and so it is safe to conclude that the two data sets are significantly similar within the context of inside readings taken. However, since both the inside data set and the outside data set led to the same conclusion upon running the statistical tests, it is safe to predict that upon running similar statistical tests on all readings, inside and outside, the results will be similar. In fact this is found to be true because when taking the outside data set and inside data set together with the new set separation being solely the set of readings taken with the thermometer and the set of readings taken with the armband, a ttest result was found, 0.06348, which again indicates that there is a significant similarity between the readings taken with the thermometer and those taken with the armband. And with a correlation of 0.978159 found between these two sets, it is safe to further strengthen the conclusion that the readings taken with the two devices are very similar, so the JLAN Armband has very accurate readings.
Graph
