Computer-assisted Acute Lymphoblastic Leukemia detection and diagnosis

Project Title

Computer-assisted Acute Lymphoblastic Leukemia detection and diagnosis

Project Area of Specialization Artificial IntelligenceProject Summary

Leukemia is a cancer of white blood cells (WBCs)which damages blood and bone marrow of human body. It can befatal disease if not diagnose at earlier stage. Generally completeblood count (CBC) or morphological image analysis is used tomanually diagnose the leukemia cells. These methods are timeconsuming and less accurate which needs to be fixed. In this project we have proposed an automated technique for the detectionof acute lymphoblastic leukemia by microscopic blood imageanalysis. This approach first segment out the different types ofcells from the image i-e. white blood cells, red blood cells andplatelets. After that Lymphocytes are separated from the whiteblood cells. Then shape and color features are extracted fromthese lymphocytes which are given to SVM classifier to classifythe cells into normal and blast. This automated leukemia detection system will be more effective,fast and accurate as compare to manual diagnosing methods

Leukemia is a cancer of white blood cells (WBCs)which damages blood and bone marrow of human body. It can befatal disease if not diagnose at earlier stage. Generally completeblood count (CBC) or morphological image analysis is used tomanually diagnose the leukemia cells. These methods are timeconsuming and less accurate which needs to be fixed. In this project we have proposed an automated technique for the detectionof acute lymphoblastic leukemia by microscopic blood imageanalysis. This approach first segment out the different types ofcells from the image i-e. white blood cells, red blood cells andplatelets. After that Lymphocytes are separated from the whiteblood cells. Then shape and color features are extracted fromthese lymphocytes which are given to SVM classifier to classifythe cells into normal and blast. This automated leukemia detection system will be more effective,fast and accurate as compare to manual diagnosing methods

Project Objectives

The following are the objectives:

To classify the lymphocytes into normal or blast

Cells.

To diagnose the patient with leukemia.

To overcome the de?ciency of manual diagnosing.

Moreover, it will lessen the

load of medical professionals and will provide the higher

accuracy and speed as compare to manual diagnosing methods.

Project Implementation Method

The process for automated leukemia detection consists

of 5 major modules including preprocessing, segmentation,

identi?cation and separation of grouped lymphocytes, feature

extraction and classi?cation.

Our methodology overview is as follows:

 A microscopic blood image comprises

of white blood cells, red blood cells and platelets. The

proposed architecture ?rst perform preprocessing over these

blood smears. Then segmentation is carried out to identify the

lymphocytes. After that grouped lymphocytes are identi?ed

and separated. Then different features are extracted from the

cells and classi?cation is carried out to classify normal and

blast cells

Benefits of the Project

Following are the benifits of this project:

It's much less time consuming then other methods of testing.

Result Accuracy is better then other methods.

It is more effective then manual diagnosing methods.

It is automated method so not much manual labor.

Technical Details of Final Deliverable

Inthis project, efforts have been made for the detection

of acute lymphoblastic leukemia from microscopic blood

images by using image processing techniques. Preprocessing

was applied over the images to remove any noise, then

segmentation is performed to detect lymphocytes from the

image. Watershed is used to separate the grouped lymphocytes,after extracting shape and color features, SVM will be used to classify normal and blast cells.

In future, we can further improve this

system to detect different types of leukemia and other blood

related diseases. Also we can further classify the leukemia into its subtypes from the FAB classi?cation which are L1,

L2 and L3. Another direction for future work is to apply

deep learning algorithm for detection which can automatically

detect different visual feature from the images with no need

of segmentation. But for this purpose we have to increase the

dataset to train deep neural networks

Final Deliverable of the Project HW/SW integrated systemCore Industry MedicalOther Industries IT Core Technology Artificial Intelligence(AI)Other Technologies Internet of Things (IoT), Augmented & Virtual RealitySustainable Development Goals Good Health and Well-Being for PeopleRequired Resources

Elapsed time in (days or weeks or month or quarter) since start of the project	Milestone	Deliverable
Month 1	Problem Identification	Identified status of problem
Month 2	data collection	Data at head
Month 3	Analysis of data	Data analysis
Month 4	training and testing	trained and tested data
Month 5	Thesis	Thesis Submission