An efficient DICOM image based brain lesion segmentation model

Project Title

An efficient DICOM image based brain lesion segmentation model

Project Area of Specialization NeuroTechProject Summary

To develop a console application for effcient segmentation and volume calculation of brain and tumor region from the MRI data.The final output of this project will be a Desktop software to solve advanced image computing challenges with a focus on clinical and biomedical applications. Using this project the Community of knowledgeable users and developers working together to improve medical computing.

Project Objectives

A workable simulation will be developed to achieve the following objectives. A system that can perform different pre-processing ( Contrast adjustment noise removal ) and post-processing  ( Segmentation, area,  and volume calculation) operation on images. As a series of images at the same time can be processed, hence a mechanism to handle parallel processing will be developed.  A simulation, where medical image analyst can perform different operations on DICOM  image to easily and accurately diagnose. This simulation based application will processes data from different types of machine, if the data is in DICOM format.

Project Implementation Method

A brain tumor segmentation system will be developed using the MATLAB 2021 software. As MATLAB has its own image processing toolbox where lot of perdefine functions are avaialble to support various operation on the DICOM data. The proposed segmentation model will be implmented in MATLAB and after successful implmentation and system validation, the application can coded into C language mex file using the MATLAB coder application which convert the MATLAB to C Language which can be further supported on various hardwares and software platforms.

Benefits of the Project

All teaching hospitals, but initially it will be deployed in the lab  where researcher and developers can use it to segment the tumor from the volumetric MRI data and also the estimate the volume of brain and tumor regions which will assist in giving the idea of tumor size and location.  Once this simulation based desktop application got successful deployment, giving satisfactory insignia from the experts, then it will be proceeded toward complete deployable project. Successful completion of deployment phase will lead to detail project preparation and commencement. Software development in Medical field is an evolutionary process, usually following spiral model of SDLC. Because human life is more precious and highly cautious and vigilant measures should take into account. This project can surface the potential of Pakistanis researchers in the field of Medical Industry

Technical Details of Final Deliverable

A MATLAB console based medical image analysis application which will take MRI data of head region, the proposed system will search for the tumor and brain voxels according the given criteria to segment the regions of interest. The applications will also have the following feautres :

• Statistics of the MRI Data
• Image quality enhancment
• File type conversion between DICOM and nifti data
• Exporting the segmented DICOM images which can be further used by advance medical data processing systems.  

Final Deliverable of the Project Software SystemCore Industry HealthOther IndustriesCore Technology NeuroTechOther TechnologiesSustainable Development Goals Good Health and Well-Being for PeopleRequired Resources

Item Name	Type	No. of Units	Per Unit Cost (in Rs)	Total (in Rs)
			Total in (Rs)	70000
GPU Based System	Equipment	1	70000	70000
Printing, Documentation and traveling	Miscellaneous	0	10000	0