Diabetes Lite

My product measures blood glucose levels non-invasively using NIR spectroscopy and stores the obtained results in an app. 

“Health organization shows that the amount of people that got diabetes disease in 1980 is 108 million and became 422 million in 2014. Prevalence has been rising more rapidly in low and middle-income countries than in high-income countries. Between 2000 and 2016, there was a 5% increase in premature mortality from diabetes. In 2019, diabetes was the ninth leading cause of death with an estimated 1.5 million deaths directly caused by diabetes”. After I saw these numbers and the number of people that suffer from needles and injections, I wanted to help the present generation and the coming generation have an easy life without the pain of needles whenever they want to measure their blood glucose. India is the capital of diabetes with 134.2 million people battling diabetes every single day. While the severity of the disease changes within that number, still a very significant portion of people go through pain every day while testing their blood sugar levels. My grandmom is one such person. She checks her blood sugar levels 5 times a day and its pains me to watch her go through that. If just watching it pains me I can't imagine the number of pain she's going through. because of this and my ambition to become a scientist in the pharmaceutical field, I went into this line of creating devices that will help people with chronic conditions. I have created a non-invasive method to check blood glucose levels using NIR spectroscopy and I plan on implementing it worldwide to end the suffering of Diabetic patients. The device that I have created is cost-effective, user-friendly, pocket friendly, and is also eco-friendly as it does not require the constant changing of needles or test strips. I have implemented this on a small scale but getting adequate funding and more people to come and survey will improve the accuracy of the test and will benefit people. 

The proposed work is based on NIR optical technique. A NIR light (Near-infrared Light) source of 940 nm wavelength is chosen because it is suitable for measuring blood glucose concentration. The sensing unit consists of a NIR emitter and a NIR receiver (photodetector) positioned on either side of the measurement site (fingertip). When the NIR light is propagated through the fingertip in which it interacts with the glucose molecule, a part of NIR light gets absorbed depending on the glucose concentration of blood, and the remaining part is passed through the fingertip. The amount of NIR light passing through the fingertip depends on the amount of blood glucose concentration. The transmitted signal is detected by the photodetector. The output current of the photodetector is converted into a voltage signal and then it is filtered and amplified in the sensor. This amplified signal is fed into the Arduino. This signal is processed by using second-order regression analysis to predict the blood glucose value and the blood glucose value is displayed on the LCD display. A mobile application (App) is created to view and store the predicted blood glucose value after receiving it via Bluetooth. The Arduino UNO communicates to the mobile app via Bluetooth by connecting a Bluetooth module (HC-05) to it.

Diabetes is a common chronic disease in mostly all countries worldwide. The most used method to measure the glucose level in blood is an invasive method which is painful, expensive, and dangerous in spreading infectious diseases. During prolonged use, the invasive method results in damage to finger tissues. As an alternative, the non-invasive method can be used which facilitates frequent testing and relieves pain and discomfort caused by frequent finger pricks. Considering all these issues with the current system, I planned on introducing a non-invasive method of glucose level measurement.

I have done this project individually. 

I have surveyed the students in my grade and some teachers regarding this and I have also asked some of my diabetic relatives about their experience with the disease and if they would want a non-invasive blood glucose monitor on the market soon. My grandmom who has battling diabetes for a while told me about how she has to prick her fingers 5 times a day to check her blood glucose levels and also told me how calloused her fingers are. So, I engaged her and some other relatives who are also battling diabetes to research and fine-tune my instrument. I took the analog voltage levels and the invasive blood glucose levels from them to modify the working of the code and to give more accurate results. So far my grandmom has been a regular user of my machine and gives results that have a deviation of ±5. 

Right now in the market, there are only invasive methods of checking blood glucose levels and diabetic people need to do this 5-7 times a day depending on the severity of the disease. Having this machine will help them by easing the pain that is there while pricking and checking the levels. It is also painful, expensive, and dangerous in spreading infectious diseases. During prolonged use, the invasive method results in damage to finger tissues. This product could be revolutionary because of the sheer amount of people that are battling diabetes every day. This is also a number that is growing at a staggering amount and keeps multiplying each year. Having a product like this will reduce the pain and suffering of these people and will provide them with an easier life. 

My goal for this project is to implement this worldwide so that people would not need to suffer each day. Next year I hope to go to doctors or medical professionals to get recommendations and to see if this is something they'd be interested in. I also hope to collect data from more diabetic people and improve the existing accuracy of this product. 

Glucose is a kind of monosaccharide ( simplest carbohydrates ) with the molecular formula C6H12O6 in the form of pyranose. It has several absorption peaks in the NIR region where light possesses its maximum penetration depth in tissue is referred to as the Near Infrared window. Glucose has light absorption peaks at wavelengths of 940 nm, 970 nm, 1197 nm, 1408nm, 1536nm, 1688nm, 1925 nm, 2100nm, 2261nm, and 2326nm, but at 940 nm wavelength the attenuation of optical signals by other constituents of the blood like water, platelets., red blood cells etc. is minimum, hence a desired depth of penetration can be achieved, and actual glucose concentration can be predicted.

When a light ray interacts with human body tissues, it is attenuated by scattering as well as by absorption by the tissues. Due to the mismatch between the refraction index of extracellular fluid and the cell membrane, light scattering occurs in tissues. The refraction index of extracellular fluid varies with the glucose concentration whereas the cellular membrane index is assumed to remain relatively constant. BeerLambert Law ( The Beer-Lambert law states that there is a linear relationship between the concentration and the absorbance of the solution, which enables the concentration of a solution to be calculated by measuring its absorbance ) plays a major role in absorbance measurement which states that absorbance of light through any solution is in proportion with the concentration of the solution and the length path traveled by light ray.

More glucose tissue results in less scattering, less optical path length, and thus more absorption than before tissues. Due to increased uptake in elevated glucose tissues Reflected light is less intense compared to tissue with lower glucose content. 

I haven't yet launched my solution but next year I hope to serve around 200 people with this product as it's still a prototype and needs more development to implement on a wide scale. With the data that I will collect from the people, I hope to improve the accuracy more and implement it on a larger scale. 

I found coding a little bit challenging, but I’ve been able to write the perfect coding with some help from a famous website like GitHub. I tried to build the whole sensor on my own, but the process was too long and ruff because of the amplification stage and error of the amplification when it multiplied with a high gain. So, after digging more into the subject I found a sensor that was similar to what I wanted to build. The big challenge was how to obtain the glucose equation by taking samples from the people. 

None.