README.md

CELL COUNTER

Deep learning for wbc classification and image processing for rbc counting

HOW TO USE IT :

1-install :

     git clone https://github.com/saeed5959/CellCounter
     pip install -r requirements.txt

2-RBC :

     python3 main.py   --mode rbc_count   --img_path ./test_data/RBC/1.jpg

2-WBC classify infer :

     python3 main.py   --mode wbc_classify_infer   --img_path ./test_data/WBC/classify/basophil.jpg  --model_path ./model.pth

2-WBC classify train :

     python3 main.py --mode wbc_classify_train --dataset_file ./dataset_file.txt  --model_path ./model.pth

2-WBC segmentation :

     python3 main.py --mode wbc_segment --img_path ./test_data/WBC/segment/main_image.jpg

segmentation and classification and counting the cells in blood :

1-red blood cell :

                  1- counting the numbers of RBC     RESULT : 99.25% in counting 

                  2- finding the radius of RBC       RESULT : mean = 99%  and variance = 90%  
                  
                  3- dataset : 322 images that averagely any image has 1000 RBC

2-white blood cell :

                  1- counting the numbers of WBC     RESULT : 100% in counting 

                  2- classification of WBC       RESULT : 92% in classification  
                  
                  3- dataset : 401 images that averagely any image has 3 WBC

blood cells

for medicl application , blood cells have a huge information about the diseases . so we can after taking a picture of these cells and processing and counting how many of these cells exist in the blood then we can detect a special desease

red blood cell after detection

esotrophil

lamphocyte

monocyte

neutrophile

basophill