Deep Learning in Bio-Medical Imaging
Download as PPTX, PDF1 like948 views
This document provides an introduction to deep learning in medical imaging. It explains that artificial neural networks are modeled after biological neurons and use multiple hidden layers to approximate complex functions. Convolutional neural networks are commonly used for image data, applying filters over images to extract features. Modern deep learning platforms perform cross-correlation instead of convolution for efficiency. The key process for improving deep learning models is backpropagation, which calculates the gradient of the loss function to update weights and biases in a direction that reduces loss. Deep learning has applications in medical imaging modalities like MRI, ultrasound, CT, and PET.
Deep Learning in Bio-Medical Imaging
- 1. Deep Learning in Medical Imaging, An Introduction Joonhyung Lee, Joonyoung Song, Gyutaek Oh, Junghyun Lee, Deborah Martin, Gemechu Bekele Tolossa
- 2. What is Deep Learning?: Artificial Neurons Artificial Neural Networks (ANNs) are very loosely modeled on biological neurons. They take inputs from other neurons, multiply a weight, add a bias, and then apply an activation function.
- 3. What is Deep Learning?: Hidden Layers Neural Networks are structured in multiple layers so that artificial neurons performing simple calculations can approximate complex functions which are difficult to state explicitly.
- 4. Convolutional Neural Networks Convolutional Neural Networks are used for image data. These use ‘filters’ or ‘kernels’ to pass over the image. Each location on the kernel has its own weight and each kernel has its own bias. Each kernel performs a convolution over al the data, for each channel and each item in the mini-batch.
- 5. Side Note Technically, most modern deep leaning platforms perform cross-correlation, not convolution, due to ease of implementation. The results are the same, except that the filters are transposed.
- 6. ▪ The method in use in modern deep learning is called backpropagation (for backward propagation of errors) ▪ However, in order for a deep learning model to work, there must be a process by which it improves.
- 7. Backpropagation: How to improve the model? ▪ Select a loss function (or cost function) “ℒ” which can be differentiated by its parameters (weights and biases). ▪ Then change the parameters in the direction of the gradient that reduces the loss.