NeuralNetwork_Model

This development explores in detail various types of feature measurements that are suitable for the automated scoring of human epidermal growth factor receptor 2 (HER2) in histology slides This model implemeted on intensity features known as characteristic curves, texture features in the form of uniform local binary patterns (ULBPs), morphological features specifying connectivity of regions, and first-order statistical features of the overall intensity distribution. This considers important properties of the above features and outlines methods for reducing information redundancy, maximizing inter-class separability, and improving classification accuracy in the combined feature set. Detail description of the work is available in the link: https://ieeexplore.ieee.org/document/8547043