DESIGNING AND DATA COLLECTION
I came up with a design that suited my environment more. I call it an all-in-one P.O design. It is cylindrical in shape and has an LED and light sensor attached to it. With this design getting a more accurate reading would be little or no problem because it has stability, and the light sensor and LEDs are tightly fitted with zero percent movement. Also, to ensure perfect stability, I used siliconized acrylic latex sealant to make sure there was not any LED movement while taking readings. Below are the diagrams of the pulse oximeter.
Data collection started immediately after my device was all set up. And currently, a lot of observations have been made and they are as follows:
- No movement of the device or its components while taking readings for an accurate result.
- The tip of the LED must be directly facing the light sensor.
- The position of the finger is important and should not be too deep into the device.
- The IR values from the data has a saturation of 65535 and for the visible 0.
- I constructed a graph that shows the plot of the IR values and Visible values, and it was observed that they both had similar visualization but different values i.e. they moved at the same rate.
- After taking readings on myself, overtime the light ray’s data reduces due to the LED being ON for a long time. But are they ratio significantly close because if they are it does not matter?
So, the ratio between the values in the IR was calculated to see the difference and it turned out they have significant ratios. GOOD NEWS! Then the ratio of the IR and Visible was calculated and they also gave me similar ratios. This is what we wanted so it did not matter if the light ray’s data reduced over time.
- Two samples were acquired next. Sample A was a male and a person of color with very dark skin tone and Sample B was a male Caucasian. The reading was taken from their middle finger on the right hand, and we made some interesting observations. The average of the ratio of IR and Visible light passing through the finger for the person of color and that of the male Caucasian had a large difference. Could this be because of the thickness of the skin or the clear color difference of the skin?
Next week we would be assessing a larger sample class with different skin colors and using both the index and middle finger on the right hand and on the left hand, we would be using the already manufactured pulse oximeter to see the difference. So basically, taking readings at the same time and seeing the difference between what we built and what’s in the market selling. By doing this we would be able to answer some questions that are arising.
