Acute Lymphoblastic Leukaemia (ALL) is the most frequent leukaemia entity in children and adolescents. Despite continued progress and refinement of therapeutic approaches, disease relapse due to insufficient extinction of leukaemic Blasts still remains the number one cause of treatment failure. Flow cytometry (FCM) is one of the methodologies most useful in this respect, because it is widely available and applicable to most patients.
New methods for AutoGating of FCM data are developed in the coarse of the AutoFLOW project. The AutoGating uses GMMs in order to distinguish between Blasts and healthy blood cells. Therefore, the GMMs need to be trained on manually annotated samples. Training GMMs utilizes algorithms such as the NMF which are currently implemented in Matlab and R. Since this renders training slow and impractical, the training should be ported to C++.
For more information see the AutoFLOW project website.
Literature Review – getting to know the algorithms
Code Review – review the current implementation
Porting and Integrating to the existing C++ project
Evaluation – comparison with previous implementation
Good knowledge of C++ and Pattern Recognition, basic knowledge of OpenCV and Qt.