BME100 f2014:Group32 L6
BME 100 Fall 2014 | Home People 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 Photos Wiki Editing Help | ||||||
OUR COMPANY
LAB 6 WRITE-UPBayesian StatisticsOverview of the Original Diagnosis System
These results were then analyzed using Bayes statistics to calculate the probablity of the results. The final data resulted in 23 positives and 45 negatives however, the data that was given only required 30 positive samples and 24 negative samples. Some of the data was either discarded or blank because groups made errors within the experimental method that biased the results.
The Bayes Statistics approach to this implied that for calculations 1 and 2 the results seemed to be reliable in determine whether or not the patients had the disease SNP or whether they did not have the diseased SNP. The calculations for both were close enough to 1 to be considered alright, but the discrepancies in the data accounted for the large error away from 1. The calculations from 1/2 compared the diagnosis from the fluorimeter and the PCR machine as positive or negative, gave a numerical value for how accurate the PCR readings were. Because these numbers were fairly close to 1 within the error, one can say that the diagnosis method used was fairly accurate. However the discrepancy in the calculations cannot be ignored. The discrepancies were because of idealizations inherent in the theory as well as errors and uncertainties in the experimental method. The camera placement each time in the experiment might have been slightly off more and more each time, thus propagating the error of the experiment exponentially. The box itself was not closed properly during all pictures as well which could explain the discrepancies in the data; light let into the box distorted the images and made the data processing error ridden. The images taken by the phone itself were neither of the highest nor clearest qualities, thus the images themselves had slight distortion by the camera as well. Error within the experiment could have occurred in the human portion as well; during the experiment the samples could have gotten contaminated by oils or foreign material, different materials not supposed to be combined could have been combined or incorrect amounts of material between the tubes could have been placed in the PCR machine causing unsatisfactory results. The experimental method could be wrong as well; not enough sample could have been used or maybe there was not enough SYBR green solution per droplet in order to test for DNA sampling. Incorrect analysis of the results leading to an incorrect calibration curve could have led to incorrect results as well and thus a less reliable method. Calculations 3 and 4 give a direct numerical accuracy of the device used in this diagnostic method for the detection of the disease. The values were not statistically significant because of how small they were and thus one cannot say that this device is extremely accurate at measuring the disease based on the data taken. Because these values were not as good as the first two calculations, this suggests that this experimental method design is wrong or needs significant improvement in order to raise the standards of this device and experimental data. Both of the values calculated for the test were very small and below 0.5 which means that there was less than a 50% chance of detecting the correct diagnosis given the positive or negative test. 1)The calculation of the probability that a patient will get a positive test (SNP), if he/she were given a positive PCR reaction is close to one. Therefore, the PCR can be considered reliable for these calculations within the limits of accuracy. Possible reasons for error:
2)The probability of a patient getting a negative test (SNP) given a negative diagnostic signal is less than the first value yet close to one within limits of accuracy. Possible reasons for error in values:
Computer-Aided DesignTinkerCAD
Feature 1: Consumables KitFeature 2: Hardware - PCR Machine & Fluorimeter |