Proposal

pdf

School

Clemson University *

*We aren’t endorsed by this school

Course

343

Subject

Mechanical Engineering

Date

Oct 30, 2023

Type

pdf

Pages

Uploaded by MinisterRainGrasshopper26

[Letter of transmittal] Cardiovascular Disease Predictor 28 November 2021 Team # 48 Mentor: Arindam Sanyal (Arindam.Sanyal@asu.edu) Anirudh Kumar (akuma282@asu.edu) Hassan Ouro-Gneni (hourogne@asu.edu) Gregory Romero (gromer14@asu.edu) Stephanie Wesh (swesh@asu.edu) The attached proposal contains information on the feasibility of undertaking the project of creating a cardiovascular disease prediction device, which is a full solution to the end user containing hardware, software, and machine learning components. Our research indicates that while factors and algorithms have been identified for the prediction of cardiovascular diseases, a machine learning solution has not yet been implemented. Given the available medical data, off-the-shelf machine learning hardware available, and accessibility of cell phone apps, we have identified a path to utilize all of these to deliver an all- inclusive patient solution for the prediction of cardiovascular diseases. Thank you for your consideration, and if you have any questions feel free to contact us using the above information. We look forward to working with you on this project.

Executive Summary Cardiovascular Disease Predictor The project will consist of three components: · Hardware · Software · Machine Learning Summary: The deliverable of this project will be a complete portable solution for an end user (medical patient) to test themselves for the onset of cardiovascular disease outside of a medical office. The hardware component will consist of 1. An off-the-shelf machine learning development board (NVIDIA Jetson Nano), and 2. A custom circuit board that will interface the connection between an ECG lead connected to the patient, and the development board. The software component will consist of a mobile app that will allow the user to enter their medical profile and will communicate this profile with the machine learning board to facilitate the prediction process. The machine learning component will read in cardiovascular disease data, the user’s profile, and the ECG signal to perform the prediction analysis. The project’s budget is estimated to be less than $1000: NVIDIA Jetson Nano: $59 Custom PCB: $100 ECG Cable: $200 Total: $359 The project is to have a timeline of three and a half months. With research completed, the hours will be budgeted as follows: PCB Design: 40 hours Machine Learning Development: 40 hours Mobile App Development: 40 hours MATLAB/Python Data Processing: 40 hours

Introduction This project involves creating an ECG application in order to help detect cardiovascular diseases early so it can help save millions of lives. At the end of the project, the computer or telephone used by the person will be able to tell the difference between a healthy and unhealthy heart by using a machine learning program and data collected. Cardiovascular diseases are one of the major leading causes of death for men, women and people of most racial and ethnic groups in the United States according to the CDC. And the ECG remains one of the best ways to diagnose a heart disease. So with our application, the overall cardiovascular diseases will be detected at least four hours before onset. This will help save countless lives. The app will be accessible to everybody, since nowadays at least every household in the US has a smartphone or a computer. No need to stay in lines for hours at the ER in order to see a doctor. At the end, the outcome for our project is to allow countless life saving, time and of course money saving. Our project will benefit every social class. Literature Review Previous work in this specific area of health detection has not been conducted, but similar technology has been proposed before commencing this project that involved audio detection and use of audio processing for voice identification and for a music tool for DAW software. These propositions are far removed from the health industry and our device we have chosen now deals with specific health related situations. The development of this device will utilize both components of electronics as well as programming and machine learning to develop a detection device that will be able to predict cardiovascular diseases in at-risk patients. All of this will be synthesized into an app for more communicable access to the average patient. What makes this project so unique is that it can all be done at home, no nervous trips to a doctor’s office for an expensive consultation and no fees, it can all be done in the safety and comfort of a patient’s own home. It is a very desirable device e specially in times where there are health outbreaks such as COVID- 19 nowadays most people don’t feel comfortable making visits to the doctor’s and with good reason, this would at least provide a safe alternative where one doesn’t need to risk their health over something that should be able to be done quickly. Technical Description

Your preview ends here

Eager to read complete document? Join bartleby learn and gain access to the full version

Access to all documents
Unlimited textbook solutions
24/7 expert homework help

Some defined features of this device would be its ability to read ECG data and be able to give a prediction through an app. It's also free, simple to use, and quick and affordable. The way it functions is that the PCB is accepting ECG data and modifying, then sending it to an ML model that will be programmed to detect the illness. As far as technical work that has been completed this semester, since we are still in the planning and designing stage so far we have been able to collect the ECG data as well as the EMR data which we will be using as a reference for the PCB. Our proposed implementation and drafted prototype can be seen in this figure below. Fig. 1: Drafted prototype for PCB Figure 2: Final idealized prototype for PCB Future Plans For technical work to come to fruition next semester we have already assigned two members who will be working on the electronics portion and two other members who will be focusing on the programming and the machine learning portion of the project.

Fig 3: Demonstrated above is a future schedule for divided tasks. Team Capacities and Facilities Since all four of us are online students, the only facilities we could use is our home with the necessary components which we will be making plans to have on hand. Processing abilities/needs to conduct this product are all feasible as long as all the team members have access to a computer once an ML model is created it can be shared with the electronics team. The abilities/needs to successfully conduct this project are all controlled remotely on a computer, we have all of our data for ECG’s on a driv e which we will then connect to a PCB once the ML model is completed it will be shared with the electronics subteam in order to send the signals to the ML model. Budget ASU will be an aid when making decisions on purchases, knowing the devices that we will need in order to purchase them, we will be using most likely a PCB manufacturing company JLPCB to create a designed PCB from our electronics sub-team. The range for the entire budget should be below $1000. For hours in terms of next semester each team member will most likely be expected to yield 25- 30 hours a week on this project. Conclusion

All in all, the project in turn will end up finding itself very serendipitous in times of the pandemic such as this, this device will not only prove to be safe but also reliable and free. The point of the entire project is to offer an alternative and to overall develop a method that would put patients at ease when it comes to hospital visits and because cardiovascular diseases are so common detecting the chances of it happening and turning into an even more fatal issue. Discussions According to the EC2000 Criterion 4, when it comes to engineering aside from its mathematics and science background a lot of it involves application and sometimes innovative application, skills that would be acquired through either iterative practice or related relevant coursework pertaining to the matter. As for our project it will be more creatively based and would require more technical skills specifically with hardware and programming with of course elements of the computer sciences that were obviously acquired either through coursework or iterative practices. The whole aspect of our device being partially mechanical and partially related to programming makes this a truly integrative project that requires all skills discussed above. References F. S. C. M. R. G. F. FC; “Home blood pressure monitoring is better predictor of cardiovascular disease and target organ damage than office blood pressure: A systematic review and meta- analysis,” Current cardiology reports . [Online]. Available: https://pubmed.ncbi.nlm.nih.gov/24057836/ EM SCORE a. For our specific project, there is a market for such a device, particularly because it affects such a large amount of people just dealing with it alone in the US. Also, with times where COVID is a rampant issue and consultation fees tend to deter potential victims of fatal disease, a detection device such as E-Cardio works as a perfect solution. b. There are plenty of solution paths that are being implemented into the device for example as part of the detection is still in question how much information will be needed besides ECG will be needed to make a prediction. There could be multiple different categories needed but we are still in the process of making that decision. c. Again, as stated before there was data collected that helped bolster our findings that an at home device would not only be safer but also efficient in financial and health terms for anyone who thinks they may be at risk for developing cardiovascular diseases.

Your preview ends here

Eager to read complete document? Join bartleby learn and gain access to the full version

Access to all documents
Unlimited textbook solutions
24/7 expert homework help

d. Any new ideas are encouraged when it comes to this production. Right now, things are still in simple terms because we are in the literary and planning semester so anything new or precocious is encouraged for the sake of the project. f. There has been some data collected and applied against the proposition that our design team has, for example there was a study conducted in which it stated that blood pressure monitoring is a better predictor than office blood pressure monitoring. This indicated that there was at least some good response to a similar at home device except this time blood pressure was used. g. The team for E-cardio is diverse in terms of Electrical engineering, we have engineers who have skills in hardware and device design as well as more software-oriented engineers who deal more with audio or signal processing which would also help encompass a vital aspect of this project. k. As discussed before, with cardiovascular diseases being so common especially in the US there exists a large market that would find such a device to their interest depending on their risks for heart issues. l. There could be many hindrances in the process of coming up with a system for this project when it comes to the technical, financial, or even the testing of it. As well there could be many new beneficial discoveries that could help aid in the development in the device. m. Since cardiovascular diseases know no creed, color, or orientation it is a universal issue that could be addressed for anyone who believes they are at risk especially people who have had a history of heart attacks and/or strokes or if such happenings are hereditary. n. A discovery like E-Cardio adds value to society for anyone who is at risk because it saves not only time but also money. It also gives the subject a state of mind without judgment by giving them an opportunity to give them detection results in their own home. q. Knowing how cardiovascular diseases work and who is more at risk and more about the disease itself will lend itself to improve this device for better detecting such disease in at-risk patients. Not having a medical background isn’t a problem but basic kno wledge about early detection, symptoms, and how cardiovascular diseases affect the body would be beneficial to improving the device. p. The team is made up of different members who have experience from powers and robotics to machine learning and programming experience to develop a device that will be able to mimic the

detection that has been tested with a dataset that has been provided with real patients who could be at risk for this disease. Acknowledgment: Arindam Sanyal Sudarsan Sadasivuni Michael Kozicki