Automated FISH Signals Fusion Detection for Chronic Myeloid Leukemia Diagnosis

摘要

Cancer is considered as one of the undefeatable diseases around the world that occur due to the uncontrollable growth of abnormal cells [1]. In 2012, according to World Health Organization (WHO) and Cancer Research UK, 14.1 million adults in the world have been diagnosed by cancer, 58% of them died of it [2]. In Egypt and according to its national cancer registry institute, 108,600 have been diagnosed by cancer 66% of them died. Leukemia is cancer of the body's blood-forming tissues, including the bone marrow and the lymphatic system. Chronic Myeloid Leukemia (CML) is a type of blood cancer that causes the body to produce a large number of white blood cells. In 2012 according to WHO, 352,000 have been diagnosed with leukemia, 75% of them died of it. In 2013, 1125 diagnosed with Myeloid leukemia in Egypt [3]. In this research, FISH (Florescent In Situ Hybridization) images is the key factor of detecting abnormalities in genes. CML can be detected using FISH. CML diagnosis depends on the detection of fused red and green signals. Our approach first detects signals, their colors, and then fusion between these colors. Moreover, number of fusions need to be counted in order to specify the treatment. Our system achieved an accuracy of 98% which proofs the efficiency of the approach.