STEM at Mass Academy is a unique course led by Dr. Crowthers. The course is split into 2 parts: STEM 1 and STEM 2. STEM 1 revolves around the creation of an individual science fair project. During this time, students are required to generate their own idea for a project and develop their idea, with many intermediate check-in points along the way. Throughout the year, students will also learn how to properly read and analyze scientific journal articles, perform statistical tests to analyze data, effectively present information, and complete pieces of technical writing.
An Assistive Device for the Prevention of Nocturnal Hypoglycemia in Type 1 Diabetic Patients
Diabetes is one of the most prevalent chronic illnesses in the United States, affecting roughly 10.5% of the population.
Diabetic patients must consistently monitor their blood glucose levels to ensure that they remain within their target range.
Failure to properly regulate blood glucose levels may result in more frequent cases of out-of-range readings, which can have
detrimental effects as time progresses. Nocturnal hypoglycemia poses a significant threat to diabetic patients as it becomes
impossible to control blood glucose levels in a state of fatigue. To address this problem, a system was created using a
temperature probe and a water pump to mimic a glucose sensing system. A program was created using the Arduino platform to
trigger the water pump to begin excreting water when the temperature of the water rose above the set high point. Since the
device was able to properly collect and respond to a temperature reading outside of the target range, the same principle could
be applied to a current continuous glucose monitoring system to prevent an instance of nocturnal hypoglycemia. The combination
of multiple means of glucose monitoring continues to allow for more precise and effective responses to changes in glucose levels.
Integrating this concept into a monitoring system would allow for patients to be less reliant on their own ability to control
their blood glucose levels. Future applications may involve the creation of a similar program that is able to interact with
glucose levels from a sensor rather than a temperature reading. Additionally, the device could be altered to pump specific
amounts of glucose that are tailored to the patient rather than one static amount.
Keywords: Type-1 diabetes, auto-injection, nocturnal hypoglycemia, Arduino, glucose
The objective of this project is to create a device that can release a bolus dose of glucose when blood glucose levels are at a detrimentally low level.
Diabetic patients are unable to properly monitor their blood glucose levels while sleeping, resulting in nocturnal hypoglycemia and the inability to return their blood-glucose levels to a normal state.
Diabetes in the United States is on the rise. In 2020, around 1.6 million Americans suffered from
Type 1 diabetes. This number is only expected to grow as time progresses, to an estimated 5 million people by 2050
(National Diabetes Statistics Report 2022). Type 1 diabetes is a life-long chronic illness which prevents the pancreas
from creating and delivering insulin to the rest of the body, making it difficult to control the amount of glucose in
the bloodstream (blood glucose levels). Patients who suffer from Type 1 diabetes must be able to monitor and regulate
their blood glucose levels at all times to ensure that they remain within their individualized target range. Remaining
within this target range is essential to the well-being of a diabetic patient because frequent and repeated instances
of out-of-range readings can lead to detrimental long term health conditions such as kidney disease, vision loss, and
heart disease (CDC, 2021).
To aid diabetic patients in regulating their blood-glucose levels, multiple devices and methods were engineered and implemented. Most means of regulation can be classified into three groups: multiple daily injections (MDI), continuous glucose monitoring (CGM), and continuous subcutaneous insulin infusion (CSII). Prior to the rise of continuous glucose monitoring, many diabetic patients relied on MDI to control their blood-glucose levels. An MDI regimen consists of the use of a glucose meter and a small blood sample to obtain a glucose reading anywhere from 1 to 10 times per day. Following this, a patient must respond to an out-of-range reading through the self-administration of glucose or insulin. By collecting a reading using a blood sample, patients will be able to obtain a very accurate reading. However, a downside of an MDI regimen is that patients are only able to see one reading at the specific time they take it, making it harder to see and predict trends and the current direction of the blood-glucose levels. Continuous glucose monitoring aimed to solve this issue by maximizing the amount of information a diabetic receives regarding their glucose level. The CGM device is able to collect a continuous glucose reading by using interstitial fluid, found between the cells in a patient’s arm. This information can be stored and analyzed using a handheld device, which can track trends and display this information to the patient. CGM still requires the patient to be able to interpret and respond to the reading through the self-administration of the proper drug. Additionally, the reading from a CGM device is less accurate than that of a glucose meter (Bergenstal et al., 2010). Glucose meters can provide a “snapshot” of a diabetic’s blood-glucose levels while a CGM device provides a much fuller view of the whole picture. CSII is vastly different from an MDI or CGM regimen. CSII utilizes an insulin pump to deliver a continuous supply of insulin to the body. The pump delivers a basal dose of insulin and administers a bolus dose before mealtimes to account for the predicted rise. CSII allows for patients to be less involved in the self-administration of the insulin needed to regulate their blood-glucose levels. Still, the insulin pump does not account for cases of extremely low blood-glucose levels, in which a patient would need to self-administer glucose.
A study conducted Garg et al. provided evidence that combining multiple means of blood-glucose regulation could increase
the efficacy and performance of these devices. A group of individuals who relied on either an MDI regimen or a CSII regimen
were observed. The control group continued to rely on their traditional methods of blood-glucose regulation while the study
group used a CGM device in conjunction with their traditional method. It was observed that in both the MDI study group and
the CSII study group that the overall time spent within the target range increased (0.7 hr/day for CSII group, 1.4 hr/day
for MDI group, P = 0.009). Similarly, the time both groups spent below their target range was reduced by 21% in the CSII
group and 30% in the MDI group (P = 0.05). Based on this data, it can be concluded that the addition of a CGM device in
conjunction with another method of blood-glucose regulation could dramatically increase the accuracy of the readings, and
the patient’s awareness of their needs at a given time. Continuing to explore the possibilities of connecting multiple devices
could allow for the creation of more advanced and accurate monitoring systems.
The current state of diabetes technology still ultimately relies on the patient’s ability to respond the data provided by their given device. Unfortunately, when a diabetic patient experiences a low blood-glucose reading while they are asleep (an instance of nocturnal hypoglycemia), they are unable to comprehend and react to this data. Nocturnal hypoglycemia poses as a severe threat to patients who live alone, as there is no other way to receive glucose other than self-administration. Currently, no device exists on the market that is able to simultaneously monitor glucose and react to a low reading by delivering glucose to the body.
A battery-powered water pump was created using two 30 milliliter syringes, one 10 milliliter syringe, one 6-volt
DC motor, one sheet of iron, one empty ink cartridge from a ballpoint pen, and one metal rod of a 1/5 cm diameter. The metal
rod was first inserted into one of the 30 ml syringes through the nozzle. It was then trimmed so that the rod extended 2 cm
above the tip of the nozzle. The iron sheet was cut into a small square, with each side measuring to be the same width of the
diameter of the 30 ml syringe. The base of the syringe was traced onto the iron sheet and cut so that it could fit within the
syringe. The sheet was then laid flat and marked at its center. This point was then drilled using a drill bit with a 1/5 cm
diameter. The iron sheet was bent into a propeller shape and attached to the rod using superglue and steel epoxy as shown in
Figure 1. This was left to set overnight. 8 small, evenly spaced holes were drawn onto the 30 ml syringe at the top near the nozzle.
Additionally, the 5 ml syringe was traced near the bottom of the syringe and a hole of the same diameter of the syringe was
created using a drill. This syringe was cut at the nozzle, creating a tube-like shape to act as the spout. The secondary 30
ml syringe was cut at the 5 ml mark, and the piece with the nozzle was discarded. The DC motor was inserted and glued to the
syringe so that the rod faced the bottom of the syringe as shown in Figure 2. The rubber stopper from one of the 30 ml syringes
was drilled at its center using the same 1/5 cm diameter drill bit. The pump was the assembled as shown in Figure 3, using a
small piece of the ballpoint pen to connect the rod to the motor axel.
The water pump was tested through 2 different procedures: Methylene Blue testing and Temperature Testing.
Methods and Materials
Methylene Blue Testing
The water pump was used in conjunction with a 0.02% solution of methylene blue and water. The solution was prepared by first creating a 2% stock from methylene blue concentrate, and further diluting the stock by a factor of 50. 300 mL of the 0.02% solution was poured into a beaker, and the water pump was held so that the propeller was completely submerged. The spout was oriented underneath a 50 mL tube. The pump was then connected to the 9-Volt battery so that it began to pump the solution into the tube. A stopwatch was run from the time that the water initially began to pour out of the spout until 5 seconds had passed. This process was repeated twice more, with time intervals of 10 and 15 seconds. Following this, the tubes were filled to the 50 mL mark, therefore creating dilutions of the original 0.02% solution. The concentrations of each of these 3 samples along with a sample of the 0.02% solution were collected through the use of a Vernier Spectrophotometer. A blank cuvette was prepared using a sample of water. Then, 4 additional cuvettes were analyzed with each of the corresponding samples. The data was collected in LoggerPro and then transferred to Excel for analysis.
The device was tested using a temperature-based simulation to represent the fluctuation of blood glucose levels. A beaker of water was placed onto a Bunsen burner to mimic the dynamic changes of a diabetic patient’s blood glucose levels, where an increase in temperature is correlated with a decrease in blood glucose levels. The purpose of the water pump in this test is to pump a continuous supply of cold water into the beaker to represent a dose of glucose to increase the simulated blood-glucose levels. The device was programmed to use the detected temperature of the water to determine whether the motor was supplied power. When the temperature of the water reached 100 degrees Celsius, the program allowed for the power to be supplied to the motor, allowing the cold water to be pumped and counteract the simulated low blood-glucose reading.
Figure 1. Absorbency as a function of Wavelength
Figure 3. Prototype of Water Pump
Figure 2. Absorbency as a function of Time
Figure 4. Sample Cuvettes, blank to 0.02% solution (Left to Right)
The data from the Absorbency vs. Wavelength was extrapolated from the 661.8 wavelength as this is where the concentration of the solution had the most direct impact on the absorbency. The data was moved into Excel and a linear regression model was created to demonstrate the line of best fit. The following equation was derived: y = 0.0599x - 0.011 (Correlation Coefficient of 0.999513). This model suggests that the predicted line would be aptly able to predict the necessary pump injection time in order to achieve any given dosage of glucose. The linear model could be altered by equating the methylene blue solution to that of a dose of glucagon. The data is highly correlated, however, the test should be run with a larger sample size in order to more accurately reflect the pump activation time.
Discussion and Future Extensions
Unfortunately, this device was ultimately unable to be created using a Continuous Glucose Monitor as a testing apparatus due to lack of proper materials. Applying this technology to a CGM would allow for the efficacy of the program to detect a low glucose reading to be tested. Moving forward from this, the device could be combined with other methods of diabetes management in order to continue to improve and build upon current technologies. Prototypes could also be improved and designed to account for portability, wearability, and comfort. Safety mechanisms would be put in place to ensure that the device would only inject glucose when a low glucose reading was detected, and could not be triggered by physical touch.
Ben Mohammadi, L., Klotzbuecher, T., Sigloch, S., Welzel, K., Goeddel, M., Pieber, T. R., & Schaupp, L. (2015). Clinical performance of a low cost near infrared sensor for continuous glucose monitoring applied with subcutaneous microdialysis. Biomedical Microdevices, 17(4). https://doi.org/10.1007/s10544-015-9983-4
Bergenstal, R. M., Tamborlane, W. V., Ahmann, A., Buse, J. B., Dailey, G., Davis, S. N., Joyce, C., Peoples, T., Perkins, B. A., Welsh, J. B., Willi, S. M., & Wood, M. A. (2010). Effectiveness of sensor-augmented insulin-pump therapy in type 1 diabetes. New England Journal of Medicine, 363(4), 311–320. https://doi.org/10.1056/nejmoa1002853
Buckingham, B. A., Raghinaru, D., Cameron, F., Bequette, B. W., Chase, H. P., Maahs, D. M., Slover, R., Wadwa, R. P., Wilson, D. M., Ly, T., Aye, T., Hramiak, I., Clarson, C., Stein, R., Gallego, P. H., Lum, J., Sibayan, J., Kollman, C., & Beck, R. W. (2015). Predictive low-glucose insulin suspension reduces duration of nocturnal hypoglycemia in children without increasing ketosis. Diabetes Care, 38(7), 1197–1204. https://doi.org/10.2337/dc14-3053
Centers for Disease Control and Prevention. (2021, November 16). What is diabetes? Centers for Disease Control and Prevention. Retrieved December 21, 2021, from https://www.cdc.gov/diabetes/basics/diabetes.html
Effectiveness of sensor-augmented insulin-pump therapy in type 1 diabetes. (2010). New England Journal of Medicine, 363(11), 1092–1092. https://doi.org/10.1056/nejmx100057
Garg, S. K., Voelmle, M. K., Beatson, C. R., Miller, H. A., Crew, L. B., Freson, B. J., & Hazenfield, R. M. (2011). Use of continuous glucose monitoring in subjects with type 1 diabetes on multiple daily injections versus continuous subcutaneous insulin infusion therapy: A prospective 6-month study. Diabetes Care, 34(3), 574–579. https://doi.org/10.2337/dc10-1852
Kubihal, S., Goyal, A., Gupta, Y., & Khadgawat, R. (2021). Glucose measurement in body fluids: A ready reckoner for clinicians. Diabetes & Metabolic Syndrome: Clinical Research & Reviews, 15(1), 45–53. https://doi.org/10.1016/j.dsx.2020.11.021
Maahs, D. M., Calhoun, P., Buckingham, B. A., Chase, H. P., Hramiak, I., Lum, J., Cameron, F., Bequette, B. W., Aye, T., Paul, T., Slover, R., Wadwa, R. P., Wilson, D. M., Kollman, C., & Beck, R. W. (2014). A randomized trial of a home system to reduce nocturnal hypoglycemia in type 1 diabetes. Diabetes Care, 37(7), 1885–1891. https://doi.org/10.2337/dc13-2159
Pickup, J. C. (2012). Insulin-pump therapy for type 1 diabetes mellitus. New England Journal of Medicine, 366(17), 1616–1624. https://doi.org/10.1056/nejmct1113948
Statistics about diabetes. Statistics About Diabetes | ADA. (n.d.). Retrieved December 21, 2021, from https://www.diabetes.org/resources/statistics/statistics-about-diabetes
World Health Organization. (n.d.). Diabetes. World Health Organization. Retrieved December 21, 2021, from https://www.who.int/news-room/fact-sheets/detail/diabetes