DYNAMIC BRAIN LAB
2015 First Hardware Prototypes
Until 2015 we were researchers in cognitive science, measuring and testing hypotheses. None of us had deep experience in building hardware at this point. With a little bit of confidence, and thanks to the many years of experience recording physiological signals in research laboratories, we started building our own measurement and signal acquisition hardware. Many small steps ahead, after billions of consults with friends and colleagues, we arrived at our first functional breadboard. It was a simple heart-beat counter. With a great boost to our confidence we carried on building, testing and eventually experimenting with several electro physiological measurements, such as photo plethysmography (PPG), ECG and EEG acquisition boards. Later on, we have built and evaluated other devices; temperature, light pressure, galvanic skin response and vibration sensors.
Autumn 2015 The Glucose Problem
We started noticing more and more people around us occasionally pricking their fingers to draw a drop of blood. This slightly painful and irritating procedure is considered normal in the life of a diabetic, a drop of blood is required multiple times a day to monitor blood glucose. We know now, from own experience, how irritating and painful this becomes with time. Only after looking at the statistics of Diabetes and asking around we realized how huge and common is this problem. Many more friends than we ever thought before, prick their fingers daily, multiple times. Some diabetic patients have to draw a drop of blood up to 12 times a day. Our parents had glucose monitors. The chance of needing these to control body glucose grows with age, but is quite high even in younger population either due to bad luck, genes or bad lifestyle choices.
Summer 2016 The Glucose Measurement Problem
This was a period of long discussion and current research review. Non-invasive methods for blood glucose monitoring were researched before. We knew this from the work on device we had build up to this point. We have reinvented the history and current research and decided to try replicate some of published studies. Meanwhile, we kept improving our skills in building hardware.
November 2016 1st Generation of Glucoscanner TM
Our first breadboard prototype for sensing glucose was an educated guess based on a few research papers we have deemed plausible. At this moment only optical sensors were used, they gave us first positive signs of a working principle.
February 2017 2nd Generation of Glucoscanner TM
At this point our first PCB prototype was still a bunch of wires and a cuff made with a cardboard. We have performed first experiments on people other than the two researchers. Results were promising and we learned that our method worked better when adjusted for each individual subject. We started adding sensors to measure influences of the environment and subjects’ condition (temperature, perspiration). And results kept improving, so we felt that we had enough knowledge and experience to put some real and useful product together.
April 2017 Laboratory (“DBL”)
In April we moved from a shared office in a basement to first office of our own. We set up a lab which we didn't have to clean up every time before we leave. We got professional tools for prototyping and measuring, a 3d printer, various development boards with micro-controllers, stocked on electronic parts.
May 2017 3rd Generation of GlucoScanner TM- First Compact USB Connected Device
Our plans were to start gathering experimental data from outside subjects within weeks from establishing our lab. In May we were ready and after weighing options for the next step, we contracted a company to bring test subjects to our lab.
June 2017 Calibration/Validation Experiment 1 (Outside Subjects)
Experiment took 7 days. We have recorded data from 48 external subjects, using two hand assembled prototypes of GlucoScanner. Each subject provided at least 3 recordings on each device over a period of roughly 80 minutes. Recordings were obtained at varying levels of blood glucose (subject came with empty stomach instructed to bring food/snacks).
July and August 2017 Data Analysis and plans for prototype manufacture
We kept running a series of experiments aimed at 1) validation of our measurement method, 2) calibration for various type of skin, finger size and other physical and environmental factors, 3) gathering data on usability and 4) study of the interaction of subjects with our prototypes. In the meantime, we have started discussions with potential partners to bring our prototype closer to a real portable product for mass production.
September 2017 Calibration/Validation Experiment 2 (Outside Subjects)
Using the same company, we have revisited 10 subjects from June. We applied a series of improvements to our devices and to the experimental procedure and conducted a more tightly controlled experiment.
November 2017 Planning Next Steps, Large Scale Testing
We were getting sick of pricking our own fingers and we also needed more measurements from different individuals in arbitrary but common conditions. Friends and supporters came to donate their time and blood for our good cause. Thanks to these supporters we could speed up our progress but we still needed more help from outside of the lab to fine tune our devices.
We decided to try crowd funding. Our target is to get just enough money to pay for the last prototype before mass production. It shall be smaller, truly portable and practical. We want to allow anybody to check glucose quickly and conveniently in everyday life situations, even in the while you wait to pay your restaurant bill. At first, we will need just a few boxes landed to carefully selected future customers, which would allow us to continuously gather data for a number of weeks.
We need to select subjects mainly for two reasons. 1) We need to ensure that data collected with the final device to be will cover the full range of commonly measured blood glucose levels. 2) Individual differences, such as skin type and finger size, are an important parameter in our measurements. We will take these into consideration too. Until now we were able to group people with similar traits into just a few groups. For a new subject we identify a subgroup and apply appropriate adjustment to our measurement. The greater the variety of subjects we choose the better performance we can expect on new subjects.
Ultimately, we need to learn about user's habits, usage and user experience on the portable device we are about to manufacture. We want to get the device right, we need tests and feedback. We are just a few steps away.
2018/2019 4th Generation of Glucoscanner TM
DYNAMIC BRAIN LAB
© 2020 All right Reserved