Group project details

Objective

The purpose of this project is for each team of students to learn more about modeling and simulation of signals and systems using software tools, particularly MATLAB and Simulink. These projects will provide everyone with an opportunity to apply techniques learned in class to real physiological data.

Datasets

Each team should examine and study a specific physiological data set from those listed below and available via PhysioNet. The below list has been curated for size (not too small, not too big) and ability to visualize the waveforms without downloading them (look for a link "Visualize Waveform" on the page for each dataset) but not much else.

Questions

In your project you should try to answer the following:

• Review the background material about the instrument that obtained the data.

  • How was the data collected? What instrument was used? Are there limitations to the data? Has any signal processing been performed?

  • What is the most important information in the signal? What characteristics does this information have (frequency range, magnitude, etc.)?

• Plot the data in both the time and frequency domains.

  • What insights can you gain from these plots about the data?

• Design and apply one filter using simulation techniques in MATLAB.

  • After filtering, plot the output in the time and frequency domains and compare it to the plots from raw data. How well did the filter work?

• Design and apply a second filter using simulation techniques in MATLAB and/or Simulink.

  • After filtering, plot the output in the time and frequency domains and compare it to the plots from raw data. How well did the filter work?

• Compare the two filters.

  • Which one presents important information better? How do you know?

• Compare the filters on at least 5 trials/participants/runs.

  • Within the dataset, select four additional trials or subjects where at least one variable of the experiment changed. How do the outputs compare?

Deliverables

There are three main deliverables. These are a project proposal, the final poster/video and your code along with its documentation.

Project proposal

The first is a project proposal due before Spring Break (i.e. March 10, 5.00 pm). This is a one page PDF document listing the group members (more than one, less than four), the chosen dataset along with a short summary of it & why it is interesting to the group and the intended type of final deliverable (poster/video). In this document, mention how often the team plans to meet, what software the team intends to use and the mutual expectations and responsibilities between team members. This will serve as a contract amongst the team.

Summary Poster/Video

Each team will create a poster or video presentation summarizing the background information about the dataset, the rationale behind the analysis approach as well as key takeaways including results of the modeling and simulation. The results should include a description of each filter and why it was chosen and the comparison of the results. This will be the primary final deliverable.

The poster should be no larger than 36 inches by 48 inches and in a vector PDF format. It can be a traditional research poster or infographic as long as it covers the above information. The video summary should be 8-10 minutes in length (hard upper limit). The video can be created via a service such as Powtoon, recorded with slides on a screen in a conference room, or with slides via Zoom.

Supplemental report

In addition to the summary presentation, each team will submit a supplemental text document that includes references used in the presentation and a list of the contributions of each team member to the project. The references should include all sources used from journals, the internet, discussions, etc. The references should be documented using the IEEE citation format. See this guide or our library reference and the official documentation

All MATLAB and Simulink files should be well documented using text (in Live Scripts or Simulink annotations) and comments. Another group in the course should be able to open your LiveScript and understand what you did after reading the document. The dataset used should be documented within the files. If someone (e.g., a grader, TA or Instructor) has the dataset and the team’s MATLAB/Simulink files, they should be able to run the files and recreate the plots. Therefore, all data loading, processing, and simulation should be contained in one of the files that are submitted.

Deadlines

• March 10: Project proposal due by 5.00 pm

• May 10: Final deliverable due by 5.00 pm

Suggested Datasets

Note: Projects in bold have been used in previous semesters.