BME100 f2017:Group16 W1030 L3
LAB 3 WRITE-UP
Descriptive Statistics and Graphs
Gold Standard- Mean: 96.64716049; Standard Deviation: 1.922602071
Spree- Mean: 96.5; Standard Deviation: 2.121320344
Heart Rate Data
Gold Standard- Mean: 90; Standard Deviation: 11.3137085
Spree- Mean: 92; Standard Deviation: 14.14213562
Heart Rate: T-test value =0.427116193 Temperature: T-test value= 1.0968 X 10-21
Summary of Results Based on the given data- the gold standard data and spree data for the pulse ox and the thermometer, graphs were generated. For the graphs shown above the correlation coefficient for the thermometer and the spree band and the pulse ox and the spree band was determined. The Pearson's r that compared the thermometer and spree band gave a value of 4.3666921*10-11 (the square root of 1.0968 X 10-21) and for the pulse ox and the spree brand the Pearson's r was approximately 0.6501538 (the square root of 0.477). The trend of this pearson’s r curve, being that is positive, indicates the best validity and best reliability of measurements take from both devices. The mean for the gold standard of the temperature data came out to 96.64716049 and the mean for the spree data of the temperature had the value of 96.5. For the heart rate data, the mean value of the gold standard was 90 and for the spree data of the heart rate 92. The standard deviation, as depicted on the graphs, was also calculated. For the gold standard of the temperature, the standard deviation was 1.922602071 and for the spree data of the temperature it came out to be 2.121320344. For the heart rate data, the gold standard's standard deviation was 11.3137085 and for the spree of the heart rate data the standard deviation was 14.14213562. Based on our experiment scenario, the mean would be the average number of strokes within the one year waiting period of the people with the pressure monitor device, the thermometer patch, and those without any aiding device. Like the example of the pulse ox and thermometer, the validity for this particular stroke experiment will be tested by comparing the results of the two devices for several different trials and seeing if the results are similar to that of this experiment's particular gold standard. Within group variance will tell us the means within the groups so that way information about which device is more/least effective in detecting a stroke can be affirmed.
Experimental Design of Own Device
Goal: To test and determine the reliability of our device in detecting an oncoming stroke by measuring blood pressure changes in the carotid arteries in comparison to devices that measure other bodily functions/factors.
Hypothesis: Our device, which relies on testing blood pressure changes in the carotid arteries, will be more accurate in determining the chances of the patient suffering from a stroke as opposed to competitor products such as the temperature band.
Experiment: We will take 240 people who have already suffered from a stroke and are therefore more likely to suffer from another one, or who are genetically inclined to suffer from one. From these 240 people, 80 people will receive our device which is the pressure monitor, 80 will receive a temperature patch that measures oncoming strokes through heat (competitor), and 80 people will receive nothing but will measure determine the probability of a stroke with identifying symptoms of a stroke (FAST method: facial drooping, arm weakness, slurred speech, and time). The two groups who are being given a device are the experimental groups, whose results will be compared to the control group, the 80 people who did not receive anything. The studies will have to be carried out for at least a year considering the chances of suffering from first time or recurring strokes vary among individuals. At the conclusion of our research period, the patients will be brought in and the amount of strokes detected (or not detected) between the group with the pressure monitor versus the group with the temperature patch and the group with no device will be compared. The reliability and the accuracy of each of these devices will be measured on how many strokes the devices caught in relation to how many they did not detect in a timely manner and alerting the patient.