Disease Prediction using Image Processing

Machine Learning is a technology that gives computers the power to learn from their past mistakes and experiences. This approach is finding large-scale applications in many fields around the world. One of the most significant uses of this technology is in the medical field. Every year, diseases like diabetes and cancer take the lives of millions of people. What if there was some way we could predict the early onset of such diseases? What if there was some way we could understand we have the disease before it was too late? Well, doctors and medical researchers around the world are working together to build a big-data backed machine learning algorithm that will help with disease prediction. In this Machine Learning project, we are going to be looking at something similar- an application that will help us predict the occurrence of a disease based on patient history.

Looking to build projects on Computer Vision?:

Computer Vision Kit will be shipped to you and you can learn and build using tutorials. You can start for free today!

1. Surveillance Robot

2. Sixth Sense Robot

3. 3 Computer Vision Projects (Combo Course)

4. Computer Vision - Text Scanner

5. Computer Vision Based Mouse

6. Computer Vision Based Smart Selfie

7. Computer Vision Training & Internship

Project Description

Big data analysis in the field of biomedical engineering and medicine has led to an accurate study of medical data, and this has resulted in the early detection of diseases. However the system is far from perfect, as incomplete medical records, and lack of proper patient history makes it hard for the system to correctly analyze the disease pattern, affecting its overall efficiency. Also, the presence of regional diseases makes things more challenging as the algorithm has to take in several unique factors to be efficient at predicting diseases. In this project, we will try to build a system that correctly predicts the occurrence or probability of disease early on, so that the treatment can be started as soon as possible.

Concepts Used

• Fundamentals of Machine Learning
• Basics of Neural Networking
• Big Data Analysis
• Data Segmentation
• Clustering and Feature Extraction

Did you know

Skyfi Labs helps students learn practical skills by building real-world projects.

You can enrol with friends and receive kits at your doorstep

You can learn from experts, build working projects, showcase skills to the world and grab the best jobs.
Get started today!

Project Objectives

1. Prevent preventable diseases and reduce suffering due to the same
2. Minimize the cost of treatment and the financial burden due to it
3. Provide first-hand knowledge regarding the outbreak of diseases
4. Identify outbreak zones and possible high-risk patients
5. Identify risk factors for diseases

Project Implementation

• In this project, we will be building a system that predicts the chance of an outbreak for a virus and the areas of high caution.
• We shall be taking the example of Zika Virus as there is a lot of datasets available for this particular pathogen.
• We will be making use of the Zika Data Repository created by the Center for Disease Control and Prevention to build and test the model.
• Such epidemics are caused due to the congregation of several factors such as ecology, host, population and pathogen behavior.
• Hence, these will serve as our main features around which we will base our model.
• The datasets are integrated into the system and then pre-processed so that they become ready for data extraction.
• The preprocessing gets rid of redundant data making it easier for the model to weed out unnecessary details, helping saving time and creating a more efficient system.
• The feature extraction may be accomplished by using either Random Forest or an Xgboost algorithm.
• Next, the data is split into a train and test sets, wherein 70% of the data goes in for training and the rest is used to test the system. To split the data, we make use of the StratifiedShuffleSplit function in the scikit-learn library.
• Stratified Splitting is employed as it helps in dealing with any class imbalance that may arise within the dataset.
• Finally, the classifiers are set and the model is trained.
• The trained models can then be used to predict the occurrence of diseases and to find localized zones of high outbreak probability.

• No kit required

• Image Processing
• Computer vision

Disease Prediction using Image Processing