Machine Learning based decipherment of Research CBC parameters in early discernment of common types of anemias

Project Title

Machine Learning based decipherment of Research CBC parameters in early discernment of common types of anemias

Project Area of Specialization Artificial IntelligenceProject Summary Background of project:

We are developing this project with Dr Rana Zeeshan Haider as an external who is associated with NCIBD. He gives us the dataset of different patients which contains all CBC parameters. This dataset will be used for training the model in order to find Anemia and its common type. For deep understanding we researched through different websites and research papers. Detailed overview of anemia i.e., it is a common nutritional deficiency disorder and there are several types and classification of anemia.

Common types of anemia include:

1. Aplastic anemia
2. Iron deficiency anemia
3. Sickle cell anemia
4. Thalassemia
5. Vitamin deficiency anemia and many others

Significance/Importance of project:

To help the health sector in the detection and discernment of common types of anemia in early stages to avoid severity of the disease in patients. 

Description of project:

Using Machine Learning algorithm to identify and detect the common types of anemia at the early stage using Research CBC parameters instead of routine CBC to help the health sector to control this disease among the patients before it reaches a severe level

Project Objectives Objectives:

To assist the health sector to detect and identify anemia in early stages, along with its type using research CBC parameters instead of routine CBC. Doctors and research will have probabilistic approach about the presence/absence of disease along with its type without spending many time on research to obtain any conclusion.

Project Implementation Method Methodogy:

Methodology used for the development of this software project is the Waterfall model. As this is a safetycritical system that addresses the health disease/disorder, its requirement must be clearly specified and gathered in great detail without any loophole left for the understanding of any point. Also, as this is a research-based project, its requirements and documentation must be clearly specified.

Benefits of the Project Benefits of Project:

• This project mainly aims to solve a health sector problem that is related to a blood disease known as Amenia.
• The system will be beneficial for lab assisstants, doctors and researchers to detect Anemia along with its type at an early stage using their research CBC parameters of their report.
• The system will further facilitate the creation of a substitution for the unavailability of disease specialists so that early decisions may be taken to secure the life of patients.
• This will reduce the dependency of patients over specialists FYP anemia detection however for its treatment and further info the concerned specialist must be contacted

Technical Details of Final Deliverable Final Deliverable details:

The final product that will be delivered will consisit of the following parts:

FRONTEND: 

• Flask
• Heroku

BACKEND: 

• Core Python
• Machine Learning
• Supervised algorithm

API:

• For ML model deployment

More about final deliverable:

The finalized product is web based application that allow end-user to upload the CBC parameters in CSV file. The data is then send to the trained Machine Learning model, which draw some conclusion from the given data. After that it provide the results in the form of visuals, accuracy of output, Fscore of output, comparison of possibilities of different kind of anemia as predicted according to the given data.

Final Deliverable of the Project Software SystemCore Industry MedicalOther Industries IT , Health Core Technology Artificial Intelligence(AI)Other Technologies OthersSustainable Development Goals Good Health and Well-Being for People, Industry, Innovation and InfrastructureRequired Resources

Item Name	Type	No. of Units	Per Unit Cost (in Rs)	Total (in Rs)
			Total in (Rs)	65000
Nvidia RTX 2080	Equipment	1	65000	65000