karanmehra7107 / medical_diagnosis

Medical Diagnosis A Machine Learning Based Web Application

Medical_Diagnosis | A Machine Learning Based Web Application

Capstone-2: LPU | CAP347 CARGC0019

Table of Content

• Problem statment / Why this topic?
• Flow Chart / Archeticture
• Directory Tree
• Quick start
• Screenshots
• Technical Aspect
• Team
• License

• This repository consists of files required to deploy an WEB PAGE created with HTML, CSS, BOOTSTRAP, ML, DL on github.io platform.

Problem Statment

The proposed project would be very useful in the medical field. In the proposed project a machine learning- based web application would be created for medical diagnosis. For a medical diagnosis, a machine learning model would be developed and integrated with the created web application. The user would be able to upload his medical data on the web application. The web application would pass this data to a developed machine learning model for health disease detection. After detection of health disease, if the person wants to take advice from a doctor then he can fix the appointment on the web application. A chat(Email) option would be provided on the web application to provide the communication between the patient and the doctor.

Why this Project?

Although, we know that humans can do the mistakes but machines doesnt. Plus we can check the predicted outcome accuracy with machine learning. So we go for Machine learning, Keeping this in mind we researched alot in the allopathic, homeopathy and ayurvedic data. Due to less research paper for the data set of patients in homeopathy and ayurvedic we go for allopathic data set that are avalible in Kaggle and UCI machine learning portals.

Flow chart

Front-end UX/UI, Back-end Machine learning, Deep learning flow chart

Directory Tree

├── Pyhon notebooks code files
├── trained models.pkl file
├── static logos
├── Templates
│   ├── Home.html
│   ├── contact.html
│   ├── about us.html
│   ├── services.html
│   ├── css folder
│   ├── js folder
│   ├── images folder
│   └── fonts folder
│         ├── Diabetes
│         ├── Breast Cancer
│         ├── Heart Disease
│         ├── Kidney Disease
│         ├── Liver Disease
│         ├── Malaria
│         └── Pneumonia
├── app.py
├── readme.md
├── runtime.txt
└── requirements.txt

Quick start

Step-1: Download the files in the repository.
Step-2: Get into the downloaded folder, open command prompt in that directory and install all the dependencies using following command

pip install -r requirements.txt

Step-3: After successfull installation of all the dependencies, run the following command

python app.py

or
flask run

Step-4: Go to the New command prompt of root folder, run the following commands in new cmd terminal

cd templates
index.html

Screenshots

Technical aspect

This webapp was developed using Flask Web Framework. The models used to predict the diseases were trained on large Datasets. All the links for datasets and the python notebooks used for model creation are mentioned below in this readme. The webapp can predict following Diseases:

• Diabetes
• Breast Cancer
• Heart Disease
• Kidney Disease
• Liver Disease
• Malaria
• Pneumonia

Models with their Accuracy of Prediction

Disease	Type of Model	Accuracy
Diabetes	Machine Learning Model	98.25%
Breast Cancer	Machine Learning Model	98.25%
Heart Disease	Machine Learning Model	85.25%
Kidney Disease	Machine Learning Model	99%
Liver Disease	Machine Learning Model	78%
Malaria	Deep Learning Model(CNN)	96%
Pneumonia	Deep Learning Model(CNN)	95%

NOTE
==> Python version 3.6.8 was used for the whole project.

Dataset Links All the datasets were used from kaggle.

• Diabetes Dataset
• Breast Cancer Dataset
• Heart Disease Dataset
• Kidney Disease Dataset
• Liver Disease Dataset
• Malaria Dataset
• Pneumonia Dataset

Links for Python Notebooks used for model creation

• Diabetes Notebook
• Breast Cancer Notebook
• Heart Disease Notebook
• Kidney Disease Notebook
• Liver Disease Notebook

Team

Karan Mehra (Data modeling, model integration, Front-end)
Surbhi (Exploratory Data cleaning, Data gathering)
Navdeep Nijjar (Quality assurance, content writter)

Special thanks to: Dr. Amar Singh (Assoicate professr) AI in data science & Machine learning.

License

Copyright 2021 Karan Mehra | Surbhi | Navdeep

Licensed under the Apache License, Version 2.0 (the "License"); you may not use this file except in compliance with the License. You may obtain a copy of the License at

http://www.apache.org/licenses/LICENSE-2.0

Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the specific language governing permissions and limitations under the License.

medical_diagnosis's Issues

Error in version conflict between dependencies.

I tried installing the required packages via pip3 install -r requirement.txt but it returned an error as.

ERROR: Cannot install -r requirements.txt (line 14), -r requirements.txt (line 20), -r requirements.txt (line 21), -r requirements.txt (line 26), -r requirements.txt (line 35), -r requirements.txt (line 36), numpy==1.19.0 and tensorflow==2.3.1 because these package versions have conflicting dependencies.

The conflict is caused by:
    The user requested numpy==1.19.0
    h5py 2.10.0 depends on numpy>=1.7
    keras-applications 1.0.8 depends on numpy>=1.9.1
    keras-preprocessing 1.1.2 depends on numpy>=1.9.1
    opt-einsum 3.2.1 depends on numpy>=1.7
    scikit-learn 0.23.1 depends on numpy>=1.13.3
    scipy 1.5.1 depends on numpy>=1.14.5
    tensorflow 2.3.1 depends on numpy<1.19.0 and >=1.16.0

To fix this you could try to:
1. loosen the range of package versions you've specified
2. remove package versions to allow pip attempt to solve the dependency conflict

ERROR: ResolutionImpossible: for help visit https://pip.pypa.io/en/latest/topics/dependency-resolution/#dealing-with-dependency-conflicts

Please help me in resolving this conflict, almost every package is having a version issue.