Developing A Machine Learning-Powered Web Application To Enhance Heart Disease Prediction And Diagnosis
Nguyen, Hoang (2023)
2023
All rights reserved. This publication is copyrighted. You may download, display and print it for Your own personal use. Commercial use is prohibited.
Julkaisun pysyvä osoite on
https://urn.fi/URN:NBN:fi:amk-2023052212621
https://urn.fi/URN:NBN:fi:amk-2023052212621
Tiivistelmä
Modern people tend to live busy lives and do not have time to check their health. That is why the number of serious diseases has risen recently, and heart disease is one of them. The thesis aims to create a web application driven by machine learning to quickly predict heart disease based on user information and allow customers to decide whether to go to the hospital.
First, the application has a core machine-learning model, which was built, trained and tested with given datasets using several libraries, such as pandas or Tensorflow, to diagnose the status of the heart. Secondly, utilizing the Django - Python framework a server was developed to get requests, generate the prediction, and then send the result to clients. Lastly, a MERN stack web application was a crucial part that will collect user information, request the diagnosis from the Django server, and display the final result on the user's screen.
The research discovers the importance of data, which means training with a better dataset will produce a more efficient output model and vice versa. Moreover, the thesis proves the flexibility of REST APIs that transfer data between clients and servers. Finally, the project also demonstrates the concepts of Redux flow on state management on the front-end side.
First, the application has a core machine-learning model, which was built, trained and tested with given datasets using several libraries, such as pandas or Tensorflow, to diagnose the status of the heart. Secondly, utilizing the Django - Python framework a server was developed to get requests, generate the prediction, and then send the result to clients. Lastly, a MERN stack web application was a crucial part that will collect user information, request the diagnosis from the Django server, and display the final result on the user's screen.
The research discovers the importance of data, which means training with a better dataset will produce a more efficient output model and vice versa. Moreover, the thesis proves the flexibility of REST APIs that transfer data between clients and servers. Finally, the project also demonstrates the concepts of Redux flow on state management on the front-end side.