Smart diagnosis of long bone tumor / Mazni Parimin

This project is about diagnosing the long bone tumor using backpropagation neural
network. The main purpose of this project is to construct artificial neural network
model that can be used to diagnose the long bone tumor and to implement the artificial
neural network model into the design of the prototype. The construction of this
prototype consist of neural network training and testing process, where the
backpropagation training algorithm are used to recognize the data sample provided via
a user-friendly interfaces. The data samples involve in this project are data that had
been normalize and become the inputs of the prototype. After gathering the data from
Hospital Universiti Sains Malaysia, these data samples will be processed through
normalization technique in order to extract useful information to make it ready for the
training process. Here, backpropagation training algorithm will be used and network
parameters will be set. To ensure the performance of the network, its parameter such as
momentum value, learning rate and number of hidden neuron will be adjusted and
observed in order to get the best weight for the network and enable it to diagnose the
bone tumor types.