reading notes, deep learning in biomedical image analysis

A Survey on Deep Learning in Medical Image Analysis
Geert Litjens, Thijs Kooi, Babak Ehteshami Bejnordi, Arnaud Arindra Adiyoso Setio, Francesco Ciompi, Mohsen Ghafoorian, Jeroen A.W.M. van der Laak, Bram van Ginneken, Clara I. Sa ́nchez
Diagnostic Image Analysis Group Radboud University Medical Center Nijmegen, The Netherlands 

https://www.sciencedirect.com/science/article/pii/S1361841517301135?via%3Dihub

"Currently, the most popular models are trained end- to-end in a supervised fashion, greatly simplifying the training process. The most popular architectures are convolutional neural networks (CNNs) and recur- rent neural networks (RNNs). CNNs are currently most widely used in (medical) image analysis, although RNNs are gaining popularity. "

The second key difference between CNNs and MLPs, is the typical incorporation of pooling layers in CNNs, where pixel values of neighborhoods are aggregated using a permutation invariant function, typically the max or mean operation. This induces a certain amount of translation invariance and again reduces the amount of parameters in the network. At the end of the convo- lutional stream of the network, fully-connected layers (i.e. regular neural network layers) are usually added, where weights are no longer shared. Similar to MLPs, a distribution over classes is generated by feeding the activations in the final layer through a softmax function and the network is trained using maximum likelihood.

https://en.wikipedia.org/wiki/Softmax_function
In mathematics, the softmax function, or normalized exponential function,^[1]:198 is a generalization of the logistic function that "squashes" a K-dimensional vector ${\displaystyle \mathbf {z} }$ of arbitrary real values to a K-dimensional vector ${\displaystyle \sigma (\mathbf {z} )}$ of real values, where each entry is in the range (0, 1], and all the entries add up to 1.