Progress Report EEE488-converted

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

Project Report For: The Prediction of Cardiovascular diseases from ECG Team 42 Team Awesome Team Members Anirudh Kumar Gregory Romero Hassan Ourogneni Stephanie Wesh 31 st October 2021 Faculty Advisor Arindam Sanyal Assistant Professor School of Electrical, Computer, and Energy Engineering Arizona State University

Abstract/Executive Summary: The overall substance of this report is to discuss the findings that have been researched for a portable at-home cardiovascular disease predictor that uses technology such as the mechanisms of a PCB to communicate signals to an ECG after filtering the signals and sending it to an ML model. The prediction results should be able to communicate results to a mobile app. The purpose of this device is to aid people who have a predisposition to cardiovascular disease to better bolster them to track their risk. The methodology of this project would be to first purchase any relevant equipment. Second divide this process into two teams, one team would work particularly on the design and the maneuvers of the PCB while the other team would work on the software portion and the machine learning of the device, the second team will also complete the process of condensing the detection device into a mobile app. Introduction: The report is about the architecture, planning, and design for the substance of the entire project. The overall problem that needs to be tackled is being able to facilitate the risks of the illness, in which this case is cardiovascular diseases in patients who have had a history of related illness or fall under certain risk categories to develop such illness. Such device would be able to detect the illness in the span of four hours before onset. This would all be communicated onto an app that would be comprehensible for the average user to decipher. • Upon research we have found the following materials that would help facilitate the process for our proposed ML model for this project. We will be using the NVIDIA Jetson Nano developer kit. We were also provided with a cardiovascular dataset to use as reference or data to incorporate into an ML model. We also were able to find cables that would be able to accommodate the transfer of the ECG from the computer into a PCB for modifying. • As far as shortcomings, so far, we have not run into any as we have not started to build this yet. In short, we will be able to discuss further the main pathways of our device including its ability to communicate to a PCB built from scratch to a model, the methodology of how it would work in a step-by-step process which can be considered our approach, what also can be discussed were former ideas that were also contemplated that bare some similarity or complete dissimilarity before taking on this final project. All in all, our main findings had mostly to with how the device would be able to communicate with each aspect to finally be condensed into something that can be comprehensible for the average user. Approach: • Hassan Ouro-Gneni – Hassan’s skills pertain more to Power Related he also had an interest in signal processing related projects as well. He has great experience and is

skilled in hardware and circuitry which is also a great and vital aspect required for this project Gregory Romero – Gregory’s skil ls are more aligned to robotics and hardware design as well which is also a great skill needed for this project as far as the physical part of it. Hassan and Greg both have experience with PCB’s which would be the design that initiates the device for its communicative design Anirudh Kumar – Anirudh’s skills are related to anything to with audio, image, and signal processing. He has taken an ASIC design class and UGTA for EEE241, 120, and 304. He is very skilled in signals and systems as well and his skills are very useful for the ML model and the coding for the second half of the project. Stephanie Wesh – Anirudh and Stephanie have very similar interests. Specifically audio processing and dsp. Initially the project was supposed to be related to audio processing but because of such diverse skills we were able to find a project that would be able to encompass all our skills equally. • The final design for this device would ultimately be readable to the reader but in commencement would first start off by building a PCB once all materials for such has been required whether it be custom, or hand made. Once that aspect will be completed the AI part of this device will be designed by the second half of the team which deals primarily in software and machine learning. Figure 1: Depicts visual methodology of communication between PCB and ML model • Project requirements for this would be partially hardware which will account for about 50 percent and the other half would account for more software related completion. Things that will be needed to successfully ensure its successful execution would be cable wires that are both suitable for a computer and the PCB, an ML model which will be implemented through. Another requirement we will need will also be a coding program which can be Xcode, Swift, ect.

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

• The cables that are both suitable for a computer and the PCB will be used for transferring data for modification and filtering on the PCB after the PCB has been built. MATLAB would then receive this data from the PCB for its design to be able to detect said illness, this will all then be programmed into an app. Figure 2: Depicts ECG electrode connection point on a PCB that would be able to transfer data from computer to PCB

Figure 3: Depiction of cable that would be able to accommodate both computer and ECG data • It meets project requirements because it is integrative with all aspects from hardware and software being used in order to create a detection device. It also has a design and a methodology that is easily followed by the subgroups of the entire team in order to finalize a final mechanism for it. • It is better than existing solutions because it considers people who have not yet developed the disease yet but could develop it either due to diet, current lifestyle, or genetic predisposition. • Relevant research that somewhat pertains to our approach is recent studies that have been conducted that allege home blood pressure monitoring is a better predictor of cardiovascular disease than office blood pressure the article reads “The available, albeit rare, evidence indicates the superiority of home – over office blood pressure monitoring (HBPM vs OBP) to predict cardiovascular (CV) outcomes. We performed a systematic review to update the efficacy of HBPM vs OBP as predictors of all – cause mortality, CV death, and tar get organ damage…… ……. A meta-analysis with a fixed-effect model was conducted, and the heterogeneity and inconsistency indices were assessed. The search identified 291 articles, of which 10 were eligible for inclusion in the study, and five

articles published in 2012 were included in the meta-analysis. A previous meta-analysis showed the superiority of HBPM over OBP to predict all-cause mortality, CV mortality, and CV events.” • Some limitations that we might run into could be pertaining to required facilities because we are all online students. Time constraints might come into play as well because some are in different time zones. Results: So far what we have accomplished is we have assigned responsibilities to each member of the team, further subjugating us into sub teams of two according to skill. Also, we have discussed and planned models and have also discussed materials and requirements that will be needed for the future of this project. Discussion: • As far as our results, it is cohesive with two people separated in two working and focusing on respective aspects of the project to ensure success. This is relevant to problem we are solving is that machine learning mostly involves interpreting results and being able to come up with a reasonable answer which involves detection as well which is what this device will be able to interpret with the data that we have. • Additional work that is required is that we must come up with a system level design and a detailed design and that also requires more research. • Figure 4: Depicts laid out plan for semester. Conclusion

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

Cardiovascular diseases are one of the major leading causes of death for men, women and people of most racial and ethnic group in the United States according to the CDC. And the ECG remains one of the best ways to diagnosis a heart disease. That is why we as a group decided to take on this project because it will help save thousands of lives. With our project, Doctors will be able to interpret the waveforms collected and decide whether the heartbeat belong to a normal heart or to an abnormal one which can be used to identify the heart disease at the earliest stage and save the patient life. Since we are using an ECG data from a database we received for our project, the main issue was how to carry that data to our computer, and we finally found and 3 lead ECG cable with USB connector. We also decided on which board to use for the project, since we will be using a machine learning programing to teach the computer how to read the wave forms. All the achievement mentioned are very crucial for our project especially the 3 lead ECG cable. Because that is the main part that will help us read the heartbeat. With all these major milestones achieved, our next goal is to come up with the architecture of our PCB board for our project and we will be set and ready for the realization. And each member of the team is looking forward to that. References: Fuchs SC, Mello RG, Fuchs 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. Curr Cardiol Rep. 2013 Nov;15(11):413. doi: 10.1007/s11886-013-0413-z. PMID: 24057836.