Recognition of leishmania parasite and macrophage infection rate using image processing techniques

Yazdanparast, Ehsan

http://hdl.handle.net/10400.1/7709

Use this identifier to reference this record.

Name:	Description:	Size:	Format:
[Master Thesis][Ehsan Yazdanparast].pdf		18.74 MB	Adobe PDF	Download

Send Feedback

Authors

Yazdanparast, Ehsan

Advisor(s)

Shahbazkia, Hamidreza

Abstract(s)

Leishmaniasis is an epidemic dangerous disease in tropical and sub tropical regions of the world and if it is not treated conveniently, it may cause several health infections and could even lead to the death of patients. Currently, there is no efficient vaccine for the disease and available treatments cause serious side effects such as toxicity and parasite resistance. Therefore, ongoing research in Drug Discovery and other related biological areas concentrate on finding adequate drug candidates for the disease. Drug discovery pipeline for Leishmaniasis disease facilitates so many biological techniques till now to understand level of effectiveness of different drug candidates for treating the disease. The accuracy and reliability of such techniques, however, highly depends on manual process of detecting, extracting, counting and analyzing components of interest such as cells, parasites and cytoplasm regions by expert biologists. Such kind of activities is subjected to so many human-made errors and is often considered to be too time and energy consuming. The existing computational based solutions, which are reduced, also suffer from limited level of analysis and in some cases inaccuracy of the results are evident. For instance, there is no package dedicated to Intracellular Parasites Counting, which is a central operation when researching drug candidates’ effects. The current research aims at addressing the urgent need to find a solution to the problem of investigating Infection Ratio of cells more accurately, which is in fact still lacking. A computational framework which fulfills all those mentioned tasks automatically, in an acceptable time range with maximum possible accuracy of the results is suggested. The proposed solution mainly uses Image Processing approaches to process clinical images and analyze the results. The visual and statistical results then could be used independently or as a complementary tool for laboratories to investigate infection ratio of drug candidates and even at a higher level introduce vaccines for the disease.