Image processing based measurement system for quantification of the target leukemia cells

Abstract

Acute Lymphocyte Leukemia (ALL) is the most common type of cancer diagnosed in childhood. The progress of this cancer is quite fast, so time is crucial for the treatment. There are some treatments for patients such as drug therapy (chemotherapy), bone marrow transplantation, radiation treatment, and immunotherapy. Among these treatments chemotherapy is a preferred method, however, its result differs from patient to patient. It is very important to measure the effect of chemotherapy during the treatment in order to adjust the dosage of the drug. Flow cytometry (FC), polymerase chain reaction (PCR) and microscopic examination techniques are used to measure the effect of chemotherapy. However, these methods are time consuming, expensive and require experts. In this thesis, the problem is addressed by an automated cell detection and quantification method based on image processing techniques. The images acquired by optical microscopy were processed and the quantification of ALL cells, which were captured by immunomagnetic beads, was provided. Compared to the conventional methods, this technique is time-efficient, low-cost and easy to use. In addition to the quantification of ALL cells, the image processing algorithms were also developed for the quantification of the magnetic beads that were used in the initial part of the project for signal amplification. The images were acquired from a cell phone microscope for both microscope slide and diffraction grating-based biosensors. In both methods, magnetic bead accumulations were observed. Thus, the accumulation of magnetic beads was detected automatically