II. Introduction
Figure 1: Cost per person with diabetes (adapted from [2])Current treatment options for Type I diabetes include insulin injections, an insulin pump, or an insulin inhaler. Insulin injections, however, can cause pain and allergic reactions, the pump can be inconvenient and can cause infections, and the inhaler must be used with insulin injections.
To alleviate the pains associated with the current treatment options, the new device is a small implant that would be placed in the user’s arm, and insulin would be delivered as needed, depending on the measured insulin levels. With the implant, Type I Diabetics will no longer be required to monitor their insulin levels and administer injections. A doctor would implant the device, and then the user would be able to access information about the device, such as battery level from their mobile device. The device creates economic and social value for stakeholders including the user, caregivers, insulin suppliers, and manufacturers. For example, the device will save time and money for the user and caregivers. The device also creates economic value for manufactures and insulin suppliers.
The next section, methodology, describes the user needs, the materials the device is made from, and the key research in developing the device. The results section includes the development of the nanoparticle coating and how the aspects of the device were tested. The discussion section describes how the prototype was scored, the limitations of the device, and how the device compares to other options. The conclusion section explains the overall device function and a summary of the testing and results.
III. Methodology
The goal for this device is to improve the lives of those afflicted by Type I diabetes through a device that incorporates nanotechnology. The criterion for this device had to be as such: the device has to be easy to use and not intrude on one’s time, the ability to get insulin into the body had to be painless, and the device also had to be user friendly. This led us to create a device that is implantable and will use an app (detailed in Figure B3) to share data about blood sugar levels and would also be able to control how much insulin will be releasing into the body at a time.
The device will use the following materials in its function: A titanium outer shell & insulin reservoir, PVDF coating (a type of polyelectrolyte hydrogel), poly(ethylene glycol) (PEG) hydrogel, watch batteries, insulin, a refilling port [4], glucose sensing probe [5], wires, a board and a Bluetooth transmitting device. The device was then designed in SolidWorks with room for certain parts being created as if they have yet to be placed in the device, this will show where each part goes and how they will all fit together. Since titanium was not accessible, the model of the device was 3D printed and scaled up so that it can be easily analyzed due to how small the model would be if it was printed to what its intended size would be.
Figure 2: Refilling port (adapted from [4])
Figure 3: Glucose sensing probe (adapted from [5])
The orifice that opens to deliver insulin is based on polymeric artificial muscles that surround and control micrometer-sized holes that open to release drug [6]. The polymeric ring expands or contracts in response to an electrical signal transmitted through a conducting polymer that contacts a swellable hydrogel [6]. Polyelectrolyte hydrogels are stable in the body’s temperature, pH, and are known to be biocompatible. They have shown to have high durability in the body along with short response time to electric signals. Because polyelectrolyte hydrogel can absorb water to over a thousand times its size, the porous gel will cover the orifice of ideally 1 mm, opening and closing based on the delivery of 1-5 V [7]. Constant low voltage is required to maintain the hydrogel in its swelled state when insulin is not being delivered. Poly(vinylidene fluoride) (PVDF) has been chosen to surround the pores in the hydrogel due to its “mechanical and chemical resistance, thermal stability, and suitable mechanical properties,”[7]. The PVDF mechanism is explained in more detail in Appendix B2.
The main issue with the implant was the way it would be powered. Since a pacemaker battery was too big for the implant, it was decided that watch batteries would be used in the implant instead. This is because the demand for electricity in high voltages was not needed, rather the voltage needed to be constant over long periods of time. The polyelectrolyte hydrogel requires 1-5V to constantly be in its swelled state, which is its state when insulin is not being delivered. This electricity is roughly half of what is needed to power the whole device, including the Bluetooth transmitter that broadcasts all of the data from the implant to an app. This Bluetooth transmitter requires 3.3-5.0 V of input to function. This requirement will be fulfilled by one of the two watch batteries. The size of the transmitter is not as issue either with it having the dimensions of 23mm of length, 15mm of width, and 2mm of height, making it an easy addition to the inside of the device [8].
Additional testing was done in SolidWorks for the device in a fluid flow simulation of the tube that the insulin is going to be running through. The necessary information & procedures needed to conduct the flow simulation were obtained through an exploration in which the flow of insulin through a tube was analyzed. Because the SolidWorks material library did not have insulin as a pre-loaded fluid, it was created as a user-generated fluid. The flow simulation was conducted using an arbitrary pressure difference between the insulin reservoir and the end of the tube. Due to fluid flowing perpendicular to an opening creating a lower pressure around the orifice, the pressure differential was reasoned to be a difference of 25 Pa. The simulation was run at 310.15K (internal body temperature) for a total of 150 iterations. The mesh size was 150, 150, 75 (x, y, z). The flow simulation was not done to prove the exact flow rate of the insulin, but rather to demonstrate that the tube design for insulin transport would allow for effective insulin delivery to the patient.
Since the device has many parts that could not be obtained, there will not be a test of the device on a Type I diabetes patient, but rather the fluid flow simulation results and the data regarding the watch battery, hydrogel, and Bluetooth transmitter [8] proving the efficacy of the device.
IV. Results
In order to increase biocompatibility within the body, the exterior of the device would undergo the PEGylation process. This coating, which is made from polyethylene glycol, has been proven to provide superior biocompatibility between implantable devices and the body [3]. In this study, it was found that immune cells would block the device and stop insulin release unless the surface was PEGylated. This coating limits cell and platelet adhesion and allows the device to essentially remain in the body without becoming surrounded by tissue, which would lead to complications with insulin release [3]. Due to existing literature supporting the successful application of the poly(ethylene glycol) coating for bio-stability of implantable devices, no further coatings or stabilizing agents were explored within the scope of this experiment. Within the devices internal reservoir, the insulin exists in a highly pure form, which eliminates the need for additional stabilizing agents. The delivery tube that carries the insulin from the reservoir to the external orifice of the device is created from silicon that has hydrophilic properties, which allow for limited interactions between the insulin and the tube walls.
Figure 4: PEGylated external surface (adapted from [3])
The device was prototyped using ABS-plastic that was 3D-printed into the various components of the device. As discussed earlier, the cost limitations and processes needed to create a working and fully functional product greatly limited the physical deliverables of the insulin delivery device. However, a final representative model was created. Designed using SolidWorks three-dimensional modeling software, all of the physical parts of the device were created and printed. Prior to this, however, the design underwent many iterations. These iterations dealt primarily with the pumping mechanism and the blood-sugar level detection methods. Briefly, these iterations included: mechanical pumping mechanism, osmotic pumping mechanism, pH-responsive hydrogel, electronic sensors, and CNTs. The SolidWorks model changed slightly with each iteration, but the overall shape and location of each piece were not changed. The ports, holes, and interfaces on the device were slightly changed in size and location to make the modeling process easier, not because they were deemed to have any functional advantage.
Figure 5: SolidWorks rendering
Insulin stability was also examined, and it was found that based on existing literature pure insulin would not require much, if any, stabilizing agents. Especially due to the sterile nature of the reservoir, no reason was discovered during the research process to support stabilizing agents either on the surface of the reservoir or within the insulin itself.
One design feature that was changed was the insulin re-filling mechanism. The refilling mechanism allows for a needle to inject insulin into the reservoir without removing the device. This specific mechanism was adapted from M. Zaki AJ [4]. Another changed mechanism within the final device was the blood sugar detector. This device, explored originally by Yoon et al., measures the potential difference across two nano-porous platinum electrodes placed in interstitial fluid [5]. In our design, the difference would then transfer to an internal circuit board that transmits the reading to the user via Bluetooth and determine if the hydrogel should open or remain closed. Because of material constraints and costs, no actual circuits or circuit diagrams were included in our final device and deliverables.
Fluid Flow simulations revealed that the tube design, and by extension the overall device design, would work effectively and as intended. At a pressure differential of 25 Pa, the flow simulation revealed an average velocity of 0.248 m/s. Surface plots of where the insulin exits the tube revealed that maximum flow occurs towards the center of the tube, with laminar flow affecting the flow rate of the insulin at the very edges of the tube (shown below in Figure 6). After all iterations of the flow simulation were run, a clear asymptote was displayed on the graph which plotted iterations against velocity (figure 7).
Figure 6: SolidWorks flow simulation surface plot 
Figure 7: velocity vs. iterations
V. Discussion
To determine the success of our prototype, our score card includes requirements created based on the values calculated in the correlation matrix. The score card and correlation can be found in appendix A. Battery stability and biocompatibility are the two most important design features included in the final product compared to small size, insulin delivery rate, user interfacing ability, and refill ability. Based on our score card, our design meets all requirements with full points besides the small size category. Our design is four centimeters in length as opposed to less than four centimeters, receiving twenty-four out of twenty-five points.
Because the implant scored a 136 on the Pugh Scoring Matrix compared to the Insulin inhaler, 129, and the traditional method of insulin injections, 88, our final concept selection is the Implantable Insulin Device as it also monitors glucose levels. Full Pugh Scoring Matrix can be found in appendix A. The use of the inhaler and insulin injections still require patient monitoring of glucose levels and control of insulin delivery. The implantable device creates a closed-loop system, removing the patient from the process of glucose monitoring and insulin delivery while greatly reducing patient involvement in treatment cycles.
The end users of this project are people with Type I diabetes, specifically children. Important user needs include convenience, comprehensive, cheap, long term, reliable, and painless. The device dramatically decreases patients’ involvement in treatment, reducing the time, energy, stress, and pain involved in the process. Because the device is a onetime purchase, it is more affordable in comparison to traditional methods such as a pump or injections. This lessens the financial burden on caregivers who are responsible for purchasing diabetes treatment for someone. Additionally, because insulin is refilled every two weeks, the batteries are replaced every year, and all data regarding treatment is communicated through our app, the time involved in treatment is dramatically reduced, leaving patients and caregivers more time and energy for their jobs and other activities, improving productivity and potentially income. Children afflicted with Type I diabetes can focus on school, friends, and playing with friends.
Comparing our device to traditional treatment options, the market currently has several options for treatment, such as insulin injections, insulin inhalers, and insulin pumps. Insulin injections can cause irritation and possibly a possible allergic reaction when administered. Insulin inhalers are not comprehensive and must be paired with insulin injections. The insulin pump is another option available for Type I diabetics, but it remains outside of the body, forcing the patient to be cautious and always aware of its presence. Each of these treatment options requires a significant amount of time and energy.
The explorations biosensing and flow simulations contributed to our design through aiding our team in increasing our knowledge on the topics that correlated with our device design. The biosensing exploration increased our knowledge and experience with biosensors, helping us choose the one to modify that most closely aligns with our device. Because we designed a glucose monitoring device that releases the appropriate amount of insulin based on glucose levels, knowing more information on specifically how they function helped us present and explain the devices function accurately. Additionally, more knowledge on the biosensor’s functionality verified our incorporation of the sensors in the device and ultimately its performance. For flow simulation, because our design has tubes incorporated that pump blood, insulin, and glucose into the patient, this simulation helped us to gain more knowledge on how these substances would flow through our device. Additional knowledge contributed to the creation of an accurate device that incorporates fluid flow correctly. The simulation gave us experience with SolidWorks flow simulations, aiding us in our SolidWorks skills and ultimately the creation of an accurate flow simulation that pertains to our device.
Potential limitations that could be improved through research and additional testing include battery life, malfunctions, and insulin refill. While this device will dramatically decrease patient involvement in treatment, the insulin still must be refilled every two weeks, creating possible inconvenience for the patient or caregiver. Additionally, although a change of implant is an improvement from a possible alternative such as daily injections, implanting the device is a painful process for patients. Lastly, while the app does communicate device health with the patient, possible malfunctions have potential to be dangerous and are not easily fixed as they would require the removal of the implant.
Potential new work that would improve the functionality of the device would be testing on patients. Because the device only works in theory, animal testing and eventual patient testing are beneficial in verifying its function. Additional research could also be done to improve battery life and increase the time in between insulin refills.
VI. Conclusions and Future Work
The device was created to alleviate the pains associated with current Type I diabetes treatments and to address the most important user needs: painless, cheap, long-term, comprehensive, reliable, and convenient. The implant addresses these needs as the device is placed into the user’s arm and insulin is delivered as needed, so users can rely solely on the device, and need to get the device refilled at the doctor’s office every two weeks. Also, the battery level and other device information can be accessed through an app on the user’s mobile device (Figure B3). The device is made from titanium outer shell and insulin reservoir, PVDF coating (a type of polyelectrolyte hydrogel), poly(ethylene glycol) (PEG) hydrogel, two batteries, insulin, a refilling port [4], glucose sensing probe [5], wires, a board and a Bluetooth transmitting device [8]. A model of the device was on SolidWorks and was tested through a flow simulation, where we were able to obtain data on the velocity of the insulin. The model included two batteries, the insulin reservoir, and a tube to release insulin that was all housed in a tube-like container. The device criteria were tested, such as small size, biocompatible, refillable, battery stability, insulin delivery rate, and device communication. The results and research showed that the device met all the criteria except the size requirement, due to the design. The device has some limitations such as battery life, since the batteries need to be replaced every year. In the future, more testing could be done on mammals and humans to determine the safety of the implant. More research could be done to modify the insulin to make it last longer in the device and when implanted. Also, a limitation of the device is the small size of the insulin reservoir, since only a small amount of insulin could be stored and used. All in all, the implant provides many advantages over current Type I diabetes treatment options and provides a convenient way of receiving insulin.
VII. References:
[1] D. Freeborn, T. Dyches, S. O Ropper, B. Mendleco, “Identifying Challenges of Living with
Type 1 diabetes: Child and Youth Perspective,” Journal of Clinical Nursing, vol.22, no. 13, pp. 1890-1898, Jan. 2013, doi: 10.1111/jocn.12046
[2] American Diabetes Association, “Economic Costs of Diabetes in the U.S. in 2017,” Diabetes Care, vol 41, no. 5, pp. 917-928, Mar. 2018, doi: 10.2337/dci18-0007
[3] C.R. Gordijo et al, “Nanotechnology-Enabled Closed Loop Insulin Delivery Device: In Vitro and In Vivo Evaluation of Glucose-Regulated Insulin Release for Diabetes Control,” Advanced Functional Materials, vol. 21, no. 1, Jan. 2011, doi: 10.1002/adfm.201001762
[4] M. Zaki AJ et al, “Implantable Drug Delivery System: A Review,” International Journal of PharmTech Research, vol. 4, no. 1, pp. 280-292, ISSN: 0974-4304
[5] H. Yoon et al, “Wearable, robust, non-enzymatic continuous glucose monitoring system and its in vivo investigation,” Biosensors and Bioelectronics, vol. 117, pp. 267-275, Oct. 2018, doi: 10.1016/j.bios.2018.06.008
[6] A. Kumar, J. Pillai, “Implantable Drug Delivery Systems: An Overview.” Nanostructures for
the Engineering of Cells, Tissues and Organs, William Andrew Publishing, 23 Feb. 2018, https://www.sciencedirect.com/science/article/pii/B9780128136652000132.
[7] Claus, Johanna, Andreas Brietzke, Celina Lehnert, Stefan Oschatz, Niels Grabow, and Udo
Kragl. “Swelling Characteristics and Biocompatibility of Ionic Liquid Based Hydrogels for Biomedical Applications.” PLOS ONE15, no. 4, April 2020. https://doi.org/10.1371/journal.pone.0231421.
[8] “Bluetooth Audio Transmitter Module KCX-BT003-V1.1,” electronicscomp.com.
https://www.electronicscomp.com/bluetooth-audio-transmitter-module-kcx-bt003-v1.1.