Identification Of Influential Biomarkers For Human Leukemia – An Artificial Neural Network Approach
Leukemia is a blood cancer. Children and adults under the age of 20 are most affected in leukemia. If it is
detected and treated at the early stage cancer can be cured. This paper represents the classification of leukemia blood cells
inside the B lymphocytes and plasma cells by using multi layer feed-forward Artificial Neural Network. Here, the sample
consists of B lymphocytes, plasma cells, chronic lymphocytic leukemia, and multiple myeloma. The Back Propagation
algorithm has been employed to train the network. There are ten input node, five hidden node and one output node are
created. However, the network trained by the Back Propagation algorithm and validated using t- test and F- score.
Keywords— Back propagation; t-test; True positive; False positive; False negative; F-score.