Identification of different types of leucocytes in dried blood smears using neural networks

A new method based on artificial neural networks is developed for the identification of white blood corpuscle images. It uses a two-dimensional Self Organizing Feature Map (SOFM) and a single-layer perceptron. The SOFM algorithm imitates the ordering of sensory pathways and the high level of organization created during learning in the human brain. It is used to reduce the dimension of the input vector corresponding to the digitized image. This reduced dimensional 'feature' vector is applied to a single-layer perceptron.<
>