Week 1: Blood Cell Classification
Theme
Our project is about Image Classification.In this project, our aim is to classify the blood cells and to predict the possible disease according to the blood cell we detected, and to ensure that necessary precautions are taken after diagnosis.The diagnosis of blood-based diseases often involves identifying and characterizing patient blood samples. Automated methods to detect and classify blood cell subtypes have important medical applications.
Team Members: Emre Tunç, Muhammed Sezer Berker, Tolga Furkan Güler
Brief information about white blood cells:
White blood cells, are produced in the bone marrow. White blood cells, which protect the body against infectious diseases and foreign substances, form an important part of the immune system.
In our project we will work on 4 different kinds of these cells.These are:
Eosinophils are a variety of white blood cells and one of the immune system components responsible for combating multicellular parasites and certain infections in vertebrates.
Lymphocytes, known as LYM, are a type of white blood cell found in the blood and serve to protect the body against diseases. LYM (lymphocyte), which strengthens the immune system, makes the body more resistant to infections.
Monocyte is a type of white blood cell that fights bacteria, viruses and fungi. Monocytes are the largest type of white blood cells in the immune system.
Neutrophil, also known as NEU, is a white blood cell that fights against bacteria that attack the immune system. Neutrophils make up 55 to 70 percent of your white blood cells.
Cases with elevated White Blood Cells:
- Various skin diseases (eczema, psoriasis, exfoliative dermatitis)
- Parasitic diseases
- Some malignant neoplastic diseases (some types of cancer)
- Collagen tissue diseases
- Lung diseases and infections
- Tuberculosis
Low eosinophil conditions:
- Being overly stressed
- Cortisol secreting tumors
- Low blood sugar
- Increase in steroid
Dataset
We will use Blood Cell Images dataset. This dataset contains 12,500 augmented images of blood cells.There are approximately 3,000 images for each of 4 different cell types grouped into 4 different folders.Each of these classes represents a white blood cell, and each class contains photos of these white blood cells.We will make image classification by adapting this dataset to CNN algorithm.
