Introduction to artificial neural networks

What are deep learning neural networks, how do they process information and analyze images with examples from different medical applications of AI.
Written by Aiforia

Introduction to artificial neural networks

Deep learning (DL) AI is powering healthcare into the future, but what powers deep learning? In this text we will dive deep into the architecture of DL and discuss convolutional neural networks (CNNs), which Aiforia’s AI models are built from. If you want to learn more about the basics of deep learning AI you can read our introductory article here.

Artificial neural networks (ANNs) are what form and drive deep learning. They are computing systems designed to find patterns that are too complex to be manually taught to machines to recognize. Hence why deep learning is so adept at image analysis and in some regards more powerful than machine learning.

What are neural networks and how do they work?

ANNs are made up of a collection of connected units of mathematical functions, referred to as artificial neurons. In deep learning models the neurons can range in amount from dozens to millions of units always arranged in a series of layers. The neurons are connected to each other between the different layers with a series of connectors, called weighted connections.

A very simplified explanation of how the layers work together when processing information through deep learning is broken down here:

  • The input layer is fed information or data to process and learn from 
  • It transmits its output to the subsequent layers called the hidden layers
  • The output layer is ultimately responsible for producing the final result 

The information being processed in neural networks can travel through the layers, neurons, and connections in different ways: only in one direction from input to output, or in multiple directions, and more permutations of these. These ways differ between different types of ANNs. We will now dive into the type of neural network deployed by Aiforia: convolutional neural networks.

What are CNNs and how do they analyze images?

Convolutional neural networks are incredibly powerful, enabling image recognition and analysis to reach new heights, far surpassing human capability in most cases. CNNs are named after convolutions, a type of mathematical operation which they use to assess input data to extract information.

So how do CNNs see? Or more specifically, how do they analyze images? CNNs decipher sensory data, such as images or sound, through a sort of machine perception by recognizing patterns as numerical. This basically means that sensory information is translated into numbers.

For a rudimentary idea of how CNNs learn to analyze images let’s use an analogy of a factory line. After the raw materials are input they move down the line and stop at different stages of production for processing. With each stop a different set of features or information is extracted, the line processes the raw material further and further until an end-product is created.

CNNs power Aiforia and image analysis

This is a simplified explanation of how CNNs process data or images. There are different methods they can use, such as supervised, unsupervised, or reinforcement learning. You can learn more about these in our next article in the series. 

Convolutional neural networks are superhuman when it comes to image analysis and recognition. They are powerful and efficient models of artificial neural networks allowing Aiforia’s AI and its users to translate images into new discoveries and better informed decision making.