Florida Tech Computer Science Senior Design
Automated COVID-19 Detection Using Machine Learning
Team Members: Rodrigo Alarcon - ralarcon2019@my.fit.edu Emma Conti - econti2020@my.fit.edu Lamine Deen - ldeen2016@my.fit.edu Audrey Eley - aeley2020@my.fit.edu
Faculty Advisor: Zahra Nematzadeh, znematzadeh@fit.edu
Client: Zahra Nematzadeh, znematzadeh@fit.edu
Date of Meeting: 08/27 - 3:30 PM 09/05 - 3:30 PM
Goal and Motivation: The COVID-19 pandemic has underscored the need for effective and innovative diagnostic tools. A web-based application that analyzes cough audio to predict COVID-19 infection can offer a convenient and non-invasive screening method, potentially aiding early detection and reducing healthcare burdens.
Approach (Key Features of the System):
The user can record their coughs and receive predictions on their COVID-19 infection status. This feature not only provides real-time feedback but also aids in maintaining a history of the user’s infection status, making it a non-invasive and cost-effective tool for early screening. By tracking this data over time, the user can monitor their health status without the immediate need for a healthcare provider.
The web app design prioritizes ease of use, ensuring anyone can navigate it effortlessly and check their COVID-19 status at any time. Other aspects of the website will include details about the research and development of the ML model
The user can view a week-long progress chart, making it easier to visualize changes in their infection status over time. This feature helps determine when a user is recovering and no longer symptomatic, or still infected. The user-friendly layout ensures that navigating the web app is straightforward.
Users can access their data at any time, providing continuous access to their COVID-19 status history. This eliminates the need to wait for a healthcare provider for early detection and offers users a convenient and effective way to monitor their health
Novel Features/Functionalities: The web app not only predicts COVID-19 infections based on cough recordings but can also track the progression of the uses condition over time. By analyzing daily recordings, the user can observe trends in your symptoms, making it easier to determine when medical intervention is necessary. This continuous monitoring feature offers a more personalized health-tracking experience.
Algorithms and tools (Potentially useful algorithms and software tools): Convoluted Neural Network (CNN) A neural network architecture with the potential for good performance for this classification task. This will take in cough audio as input and output a COVID-19 infection prediction. Flask, Django, React Common web development frameworks. To be researched and further redefined for selection. This will handle the back end of the website and data transference. TensorFlow, Numpy, Pandas Common python frameworks and tools that will help in manipulating data and creating the neural network model.
Technical Challenges: All team members have limited experience in website application development, which is a necessary component of the final product.
The dataset should be developed using a machine learning algorithm, a convoluted neural network (CNN) was suggested. Research must be done to ensure that a CNN can be developed appropriately.
Different frameworks/architectures may be better suited for this task which would require additional research to learn and understand.
| Plan (Sep 4) | Presentation |
| Milestone 1 (Sep 30) | Presentation, Progress Evaluation |
| Milestone 2 (Oct 28) | Presentation, Progress Evaluation |
| Milestone 3 (Nov 25) | Presentation, Progress Evaluation |
| Plan (Sep 4) | Plan, Presentation |
| Milestone 4 (Sep 30) | Presentation, Progress Evaluation |
| Milestone 5 (Oct 28) | Poster, e-book page, Presentation, Progress Evaluation |
| Milestone 6 (Nov 25) | User and/or Developer Manual, Demo Video, Presentation, Progress Evaluation |