Weaknesses: Not fully accurate, can't calculate or compile data on it's own, isn't as protected from error as computerized and electronic devices.
Strengths: Compiles accurate data, does multiple steps in the PCR process (such as heating and cooling), easy to use.
Weaknesses: Expensive, data collection is rather slow (2-3 hours), can only fit 12 samples at a time, requires power source.
Strengths: Data collection is quick, cheap.
Weaknesses: Hard to use, requires many pieces (slides, smartphone, stand, box), requires software (image J).
What Bayes Statistics Imply about This Diagnostic Approach
The Bayes Statistics imply, through the probability results, that this diagnostic approach is not highly accurate. The predictability of the disease through this method is 72%, while the detection of the disease is 93%. In an ideal system, a favorable Bayesian probability would be close to 99% or better.
Calculation 1: What is the probability that a patient will get a positive final test conclusion, given a positive PCR reaction?
Variable
Description
Numerical Value
A
A patient getting a positive test conclusion
0.25
B
A positive PCR Reaction
0.25
P(B\A)
Probability of B given A
.933
P(A\B)
Probability of A given B
.933
Calculation 2: What is the probability that a patient will get a negative final test conclusion, given a negative diagnostic signal?
Variable
Description
Numerical Value
A
A patient getting a negative test conclusion
0.70
B
A Negative PCR Reaction
0.72
P(B\A)
Probability of B given A
.953
P(A\B)
Probability of A given B
.927
Calculation 3: What is the probability that a patient will develop the disease, given a positive final test conclusion?
Variable
Description
Numerical Value
A
A positive final test conclusion
.30
B
A patient developing the disease
.25
P(B\A)
Probability of B given A
.60
P(A\B)
Probability of A given B
.72
Calculation 4: What is the probability that a patient will not develop the disease, given a negative final test conclusion?
Variable
Description
Numerical Value
A
A negative final test conclusion
.70
B
A patient not developing the disease
.70
P(B\A)
Probability of B given A
.857
P(A\B)
Probability of A given B
.857
Which calculation describes the sensitivity of the system regarding the ability to detect the disease SN?
Probability of A given B in Table 1 (0.933)
Which calculation describes the sensitivity of the system regarding the ability to predict the disease SN?
Probability of A given B in Table 3 (0.72)
Which calculation describes the specificity of the system regarding the ability to detect the disease SN?
Probability of A given B in Table 2 (0.927)
Which calculation describes the specificity of the system regarding the ability to predict the disease SN?
Probability of A given B in Table 4 (0.857)
Discuss what the results for calculations 1 and 2 imply about the reliability of the individual PCR replicates for concluding that a person has the disease SNP or not.
As far as the relationship between the test conclusion and the result from the PCR reaction, they were highly accurate, being close to 1.00 (100%). While you cannot rely on this method alone to conclude whether the disease is present or not, it is a very reliable source to detect the disease.
Discuss what the results for calculations 3 and 4 imply about the reliability of PCR for *predicting the development of the disease* (diagnosis).
As far as the reliability of this method to predict whether the disease is present or not, it not highly accurate, being close to 0.75 (75%). This is not a very reliable method and you cannot confidently conclude whether the disease is present or not.
Discuss three possible sources of human or machine/device error that could have occurred during the PCR & detection steps that could have affected the Bayes values in a negative way.
There are many errors that could have occurred throughout this lab. Because we are all fairly novice with a micropipette, many people made mistakes when extracting and inputting the various liquids into different tubes. This could have cause problems by not having the correct amount of liquid in each tube causing problems later in the lab.
Another error could have been in the ability of the phone camera to take a high enough quality picture for ImageJ to be able to detect the fluorescence in the droplet. Many phone cameras don't have the ability to adjust the individual settings necessary for the perfect picture.
Another error in the ability to take the picture was the lack of perfect darkness. To be able to take the picture the lid on the box had to be cracked open slightly and allowed rays of light to affect the picture as well. This could have decreased the amount of green fluorescence that was detected making our method less accurate.
Intro to Computer-Aided Design
3D Modeling
Our team utilized Solidworks to build the 3D model of our product because multiple members of our group were well versed and experienced in using it, so we had no problem designing our product. When we first ran the experiment, we had trouble keeping the phone one the stand at a consistent distance from the droplets since the stand would move and the phone had to be touched to take the pictures. Because of this, we designed our product with stability in mind. We started out our design by keeping intact most aspects of the original design, such as the slot for glass slides and the laser. We first created a basic box and used the boss extrude tools in Solidworks to create the part that holds the slides and contains the laser. We then created the adjustable stand in a separate file and attached it to the primary fluorimeter box to make it into one cohesive unit. All in all, by using software like Solidworks, we were able to conceptualize our product in 3 dimensions and get a good idea of how it would actually work.
Improved Fluorimeter Design
Our improved fluorimeter design features characteristics that make it not only easier to use, but also more effective in taking the perfect picture. First, we have a phone dock that is adjustable to height and distance from the fluorimeter to allow for optimal distance from the droplet to take a picture with the least amount of background noise. The phone dock is going to have a replaceable component that will allow for an iPhone plug as well as a micro-USB plug. This dock will be connected to a wire leading outside of the dark, enclosed box to a button that takes a picture when pushed. This will eliminate any exterior light from reaching the droplet because the user won't have to crack the box open to take the picture on their phone.
The second major component to our improved design is the conveyor belt system on the main fluorimeter box. This will have exterior buttons to move the conveyor belt forwards and backwards and allow the glass slide to move the perfect distance to not only be able to maximize its use but also allow the droplet to get perfect exposure to the laser going across it.
Overall, our design is an all in one piece that doesn't require any exterior components to make it work. It allows for the best picture to be taken with the ultimate ease.
Feature 1: Consumables
Our PCR kit will include similar consumables from the previous PCR reaction lab. These materials include: 8 tubes filled with 500 microliters of buffer, 2 tubes filled with 1000 microliters of SYBR Green solution, a tube filled with 1000 microliters of water at pH 8, 5 labeled tubes filled with strands of DNA, a micropipette, micropipette tips, and glass slides. These specific amount of consumables is chosen under the assumption that the experiment will be performed on the same amount of DNA samples as the PCR lab. The consumables should be packaged neatly and labeled to avoid any confusion on the user's part.
Feature 2: Hardware - PCR Machine & Fluorimeter
We will be using a similar PCR machine that was used in the lab due to its effective and ease of use. There seems to be no real reason to change the design of the PCR machine so we focused our redesign on the fluorimeter. The detailed use and SolidWorks drawing and shown in the section above, but essentially our fluroimeter design makes the device more user friendly and allows for a better picture to be taken.
The major weaknesses we saw while using the fluorimeter was the lack of user friendliness. Just setting up our phone to take the picture required external bulk just to raise the fluorimeter to the height of the phone. Additionally, to be able to take the picture, the user had to lift up the lid and click on their phone to take the picture which allowed external light to reach the droplet. Our design will address these weaknesses and implement them in a way that will still allow the device to be cheap and durable. First, we will have an adjustable phone dock that will be able to move in all directions allowing for optimal height and distance from the droplet for the perfect picture will the least amount of background noise. This dock will be attached to a wire that will lead outside the enclosed box to a button that can be pressed to take the picture. This will allow the droplet to be in complete darkness save for the light exposure from the fluorimeter to allow for the best picture. Additionally, our device would be all in one piece, so there aren't two or three pieces that need to be kept together to allow for the device to be used.