Skip to content

mayankag30/HealthCare-v2

Repository files navigation

Healthcare-Project

About

This website facilitates 8 end-to-end Machine Learning and Deep Learning model built upon Healthcare datasets that can be used to predict whether a person is suffering from a particular disease or not

Significance/Impact

This website is a simple demonstration of how the power of AI can be leveraged in the field of Healthcare. Whenever a patient undergoes testing for a certain disease, he had to get his reports analysed by the doctor

This idea can help us:

  1. Reduce dependancy on the doctors, and help patients get diagnosed in minimal time and at his/her own convenience
  2. Help people avoid paying huge amount to doctors for consultation and diagnosis
  3. Extend the role of technology in healthcare
  4. To run this project on local machine - Clone the repository, download it, and run the following command on your terminal
  5. cd "Directory of main folder"
  6. python app.py
  7. Copy the Local URL onto the web browser

Tech used for Project Development

  1. Python 3
  2. Flask
  3. Jupyter Notebook
  4. HTML, CSS and Javascript
  5. Python libraries - Numpy, Pandas, Sklearn, Matplotlib, Seaborn, etc
  6. Machine Learning algorithms - SVM, Random Forest, Xgboost, etc
  7. Deep Learning - CNN, ResNet50, VGG19
  8. Heroku Cloud for deployment

Sources for dataset

  1. Pneumonia - https://www.kaggle.com/paultimothymooney/chest-xray-pneumonia
  2. Brain Tumor - https://www.kaggle.com/navoneel/brain-mri-images-for-brain-tumor-detection
  3. Malaria - https://www.kaggle.com/iarunava/cell-images-for-detecting-malaria
  4. Heart Disease - https://www.kaggle.com/ronitf/heart-disease-uci
  5. Kidney - https://www.kaggle.com/mansoordaku/ckdisease
  6. Liver - https://www.kaggle.com/uciml/indian-liver-patient-records
  7. Breast Cancer - https://www.kaggle.com/uciml/breast-cancer-wisconsin-data
  8. Diabetes - https://www.kaggle.com/uciml/pima-indians-diabetes-database

About

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors