Applied Biomedical Engineering Using Artificial Intelligence and Cognitive Models - Chapter 5 - dataset - Deep Learning Models Principles Applied to Biomedical Engineering
Description
The underlying principle of “Deep Learning” is that of the compositional nature of “neural networks” inspired by the biological elements that forms the “human brain,” such as a collection of “nodes” emulating “brain neurons” and their “neuron synapses connections as primary elements, that combine to form midlevel elements identified as “Artificial Neural Networks (ANNs),” which in turn are combined with different architectures to form more complex networks. These “ANNs*” are organized based on their architectural type, and the way their different components are connected to one another, defining the specific learning goal in different types*. They are classified in families as “Feed Forward Neural Network,” “Backpropagation Neural Networks,” “Recurrent Neural Networks,” “Memory Augmented Neural Networks,” “Modular Neural Networks,” and “Evolutionary Neural Networks.” The first two “ANN types” are studied in depth in this chapter and the others are covered in the Chapter 6. Please read this chapter at Science Direct: https://www.sciencedirect.com/science/article/pii/B9780128207185000015, or the book - eBook available at Elsevier, Amazon or bookstores worldwide Note*: There is no standard set of rules to classify the different types of “ANN,” Section 5.2.3 “RBF” to analyze “heart rhythm" (folder MATLAB_RBF) Section 5.2.4 “Probabilistic neural network (PNN)” to classify the “types of FLU: A, B, and C.” (folder MATLAB_PNN) Section 5.2.5 “Extreme Learning Machine (ELM)” for diabetes values prediction (folder MATLAB_ELM) Section 5.3.1 5.3.1 Research 5.1 Feed Forward Neural Network to Analyze “Human Body Fat” (folder MATLAB_BFE) Section 5.3.2 Research 5.2 Neural Network for clustering based on “Self-Organizing MAP* through a Shallow Neural Network” to analyze Surface Electromyography signals (folder MATLAB_CL) Section 5.3.3 Research 5.3 Neural Network for Dynamic Time series based on a “NARX is a nonlinear autoregressive exogenous model” to analyze vertical Ground Reaction Forces signals (folder MATLAB_DTS) Section 5.4 Back propagation neural network Auto Encoder Neural Network to reconstruct chest X-rays (folder MATLAB_AE) Section 5.4.4.1 Research 5.4 Backpropagation Neural Network for Patterns Recognition and classification of “Breast Cancer” (folder MATLAB_PR) Section 5.5.1 Research 5.5 “Pretrained Deep Convolutional Neural Network to obtain an AI model to classify Mammograms standard views types” (folder MATLAB_IMG) Section 5.5.2 Research 5.6 modify a “Pretrained Deep Convolutional Neural Network” to obtain AI model to “classify Mammograms view type and suggest breast abnormalities as possible breast tumor” (folder MATLAB_DND) Section 5.5.3 Research 5.7 “custom Deep Convolutional Neural Network” to obtain an AI model to “classify Cervical X-rays view types” (folder MATLAB_Build_DCN)