攻读博士学位研究生学习计划

Study Plan for Doctoral Degree Program

研究课题名称/ Research title

AUTOMATIC DETECTION AND COUNTING OF BLOOD CELLS FROM MICROSCOPIC

姓/Surname：NDIKUMANA ABOU
名/First name: Salama Al wajid

申请院（系）/Application School/Department: Medical Imaging and Nuclear Medicine

申请专业/ Application Major: Biomedical Engineering

申请导师/Expected supervisor: none

申请研究领域/ Expected Research area: Biomedical Engineering

This research proposes a method for automatic detection and count of Red blood
cells (RBC), white blood cells ( WBC) and platelets from microscopic images of blood
smear . The goal is to provide a low cost method using image processing techniques, which
is accurate and trustable as well as suitable for low income countries like Cameroon.
The cell detection method should be able to detect regular blood cells as well as
irregular blood cells , in order to be reliable and accurate .
To achieve the objectives of the specifications , it was necessary to review the different
methods of segmentation and counting of blood cells from microscopic images . Knowing
that blood cells are approximately circular shape [1], circle detection algorithms can still
handle the challenge of blood cells detection. Hough Transform is considered as one of
the most known algorithms for line and circle detection , but helpful mostly if the detected
cells are normal and consist of small number of overlapping cells with regular shape[1].

In this research , we concentrate on circle detection algorithm. The separation of blood cells
will be first carried out , before performing the count. The separation of WBCs from RBCs
will be achieved by thresholding, and specific preprocessing steps will be developed for
each cell type. The extraction of platelets , will be done using segmentation method
developed by Lipi B. Mahanta et al . [2] . For the counting of the blood cells , we opted for
the iterative structured circle detection algorithm , developed by Alomari et al. [1] .

二、 研究目的及意义 Research Purpose and Significance

The blood cells analysis is an important step for early diagnosis of pathologies.
Identification and counting of blood cells give crucial informations on health conditions of
individuals. It helps to evaluate , diagnose , and monitor various health conditions such as
anemia , infections etc.
On the market , there are various systems for the automatic quantification of blood cells that
allow to count the number of different types of cells within the blood smear. These counters
usually make use of the electronic impedance method to detect and measure changes in
electrical resistance between two electrodes as cells pass through a sensing aperture. The
measurable voltage changes are plotted on frequency distribution graphs, or histograms, that
allow the evaluation of cell populations based on cell volume[3]. Although the results of
quantification are very precise , morphological abnormalities of the cells are not detected by
the machine and , it is seen necessary a subsequent analysis of blood under the microscope .
Microscopy images are also still used for early diagnosis , analysis and count of some blood
disorders such as sickle cell anemia and leukemia , before confirming it with other laboratory
tests [1] .
In low income countries such as Cameroon , due to the high costs for getting automatic
hematology analyzers , there are still a lot of hospitals that carry out manual count of blood
cells from blood smears using microscope. The manual counting of blood cells in microscopic
images is an extremely tedious, time consuming , and inaccurate process ( because it depends
on the skills of the manipulator ).

2
In front of this situation , taking into account the importance of blood cells analysis , our goal
is to propose a strategic alternative solution to help laboratory experts for counting blood cells
using microscopy images in order to take care of patients in low income countries like
Cameroon. The purpose of this research is to set-up an universal reliable low cost automatic
system to count white blood cells (WBC) , Red blood cells (RBC) , and platelets on microscopy
images of blood smears . The system would be easily installable on computers , the laboratories
would just need to own a microscope equipped with camera , to capture microscopic images.

三、 文献综述 Literature Review

The use of image processing techniques has grown rapidly in the recent years. These
techniques help to count the cells in the human blood and, at the same time, provide
information on the cells morphology. Unfortunately, there are not standard techniques for
the analysis and processing of the images valid for each type of them, but the processing
must be adapted to the context. Regarding the microscopic images, processing techniques
vary depending on the type of blood cells to be analyzed [ 4]. Segmentation and counting
of blood cells are considered as an important step that helps to extract features to diagnose
some specific diseases like malaria or leukemia [1] .
Many researchers have investigated blood cell segmentation and counting. The
important issue is to take into account clumped and overlapping cells . Some researchers
[5-8] used morphological operations and thresholding to do the segmentation and counting.
Berge et al. [5] presented an approach based on a morphological method and iterative
threshold techniques. Segmentation was performed on red blood cells, Their method
tolerated a degree of overlapping, but in cases with a high degree of overlapping cells, the
cells were unable to be detected. Additionally, the iterative threshold method was unable
to detect faint red cells [ alamri] . Khan et al. [8] proposed a method to count WBCs, RBCs,
and platelets. Several preprocessing steps were performed before converting the image to
binary. Segmentation and cell counting were performed based on the optimal threshold
value, which was determined from a histogram. They achieved 95% accuracy with their
proposed method compared to manual counting and a hematology analyzer. However, this
method is unable to detect overlapping cells.
Since blood cells are approximately circular shape, circle detection algorithms can
still handle the challenge of blood cells detection. Hough Transform is considered as one
of the most known algorithms for line and circle detection. Since the Hough Transform
presented a good performance in different fields, many researchers [9–12] used Hough
transform method for detecting circle when performing RBC’s and WBC’s calculation task
in the microscopic images. Mahmood et al. [9] applied Hough transform method for
counting the RBC’s and WBC’s for the microscopic images . However , the ability to
tolerate a high degree of overlapping and irregular cells was limited which that method;
therefore, accuracy was not high [ 1].

Aware that not all blood cells have regular cell shapes , and this especially true if the blood
cells are diseased , we should implement advanced counting algorithm . Alomari et al. [1]
developed an iterative structured circle detection algorithm for automatic detection and
quantification of WBCs and RBCs , which performed an average accuracy of 95.3% for
RBCs and 98.4% forWBCs.

3
Works of Alomari et al. constitute the base of our project, with of course different stakes.
Because, except counting the WBC and RBC , we are also going to extract the platelets and
later counting them using the proposed algorithm.

五、科研方法 Research Methods

The counting of the RBCs , WBCs , and platelets will be based on the iterative
structured algorithm , developed by alomari et al. The original image will be first separated
into three images which exclusively contain for each of them , one type of blood cells that are
going to be analyzed . The separation of WBCs and RBCs , will be done by thresholding
segmentation and morphological operations . After the segmentation , each image is
preprocessed separately using morphology operators to obtain the edge image using canny
operator [1 ] . The extraction of platelets will be done from the original image using the method
developed by Lipi B. Mahanta et al . [2 ] . Overall , they applied Gaussian filter to remove
random noise before converting the image to the LAB color space . The transformed image is
then binarized by the method of masking the pixels having values under a certain threshold
(determined experimentally by hit and trial method). The platelets are extracted by
morphological transformation on the masked image using opening and dilution , and with the
help of the opencv function ‘’findcontor’’ . The counting will finally be done on the well
extracted blood cells , using iterative structured circle algorithm. Our system will be impleted
with C++ programing language . The tests will be carried out on real blood smears images ,
and the results will be assessed by a laboratory expert , and be compared to the result given by
automatic hematology analyzer .
The below figure shows , the general methodology for the proposed system :
Read image

Extract WBCs by Extract RBCs by Extract platelets by

Subtract WBCs
image

WBCs RBCs
preprocessing preprocessing

Counting using iterative structured

Show results

Figure 1 : Overall methodology for detection and count of WBCs, RBCs, and platelets

4
 References
[ 1] Yazan M. Alomari, Siti Norul Huda Sheikh Abdullah , Raja Zaharatul Azma and
Khairuddin Omar ‘’ Automatic Detection and Quantification of WBCs and RBCs Using
Iterative Structured Circle Detection Algorithm ‘’ , Computational and Mathematical
Methods in Medicine Volume 2014, Article ID 979302, 17 pages , pp 2-10 , 2014.
[ 2] Lipi B. Mahanta1 , Kangkana Bora, Sourav Jyoti Kalita, and Priyangshu Yogi , ‘’
Automated Counting of Platelets and White Blood Cells from Blood Smear Images’’ ,
[Link] , pp 14-17 , november 2019.
[ 3] Jo Ann Molnar, ‘’ Automated blood cells analysis ‘’in Rodak's Hematology (Sixth
Edition), at [Link] 2020
[ 4] Lorenzo Putzu, and Cecilia Di Ruberto, ‘’ White Blood Cells Identification and Counting
from Microscopic Blood Image’’ , International Journal of Medical, Health, Biomedical and
Pharmaceutical Engineering Vol:7, No:1, pp 15-18 , 2013.
[5] H. Berge, D. Taylor, S. Krishnan, and T. S. Douglas, “Improved red blood cell counting
in thin blood smears,” in Proceedings of the 8th IEEE International Symposium on
Biomedical Imaging: From Nano to Macro (ISBI ’11), pp. 204–207, Chicago, Ill, USA, April
2011.
[6] L. B. Damahe, P. G. Student, R. G. College, and R. G. C. Engg, “Segmentation based
approach to detect parasites and RBCs in blood cell images,” International Journal of
Computer Science and Applications, vol. 4, no. 2, 2011.
[7] L. B. D. Panchbhai and V. Vishal, “RBCs and parasites segmentation fromthin smear
blood cell images,” International Journal of Image, Graphics and Signal Processing, vol. 10,
no. 10, pp. 54–60, 2012.
[8] S. Khan, A. Khan, and A. Naseem, “An accurate and cost effective approach to blood cell
count,” International Journal of Computer Applications, vol. 50, no. 1, pp. 18–24, 2012.
[9] N. H. Mahmood, P. C. Lim, S. M. Mazalan, and M. A. A. Razak, “Blood cells extraction
using color based segmentation technique,” International Journal of Life Sciences
Biotechnology and Pharma Research, vol. 2, no. 2, 2013.
[10] N. [Link] andM. [Link], “Red blood cells estimation using Hough transform
technique,” Signal & Image Processing, vol. 3, no. 2, pp. 53–64, 2012.
[11] B. Venkatalakshmi and K. Thilagavathi, “Automatic red blood cell counting using
Hough transform,” in Proceedings of the IEEE Conference on Information and
Communication Technologies (ICT ’13), pp. 267–271, JeJu Island, Korea, April 2013.
[12] M. Maitra, R. K. Gupta, and M. Mukherjee, “Detection and counting of red blood cells
in blood cell images using Hough transform,” International Journal of Computer
Applications, vol.53, no. 16, pp. 18–22, 2012.