Classification based on annotation information
Systems and techniques for classification based on annotation information are presented. In one example, a system trains a convolutional neural network based on training data and a plurality of images. The training data is associated with a plurality of patients from at least one imaging device. The plurality of images is associated with a plurality of masks from a plurality of objects. The system also generates a loss function based on the plurality of masks, where the loss function is iteratively back propagated to tune parameters of the convolutional neural network. The system also predicts a classification label for an input image based on the convolutional neural network.
1. A machine learning system that employs decoder blocks for improved performance, comprising:
a memory that stores computer executable components;
a processor that executes computer executable components stored in the memory, wherein the computer executable components comprise:
a training component that trains a convolutional neural network based on training data and a plurality of images, wherein the training data is associated with a plurality of patients from at least one imaging device, wherein the plurality of images is associated with a plurality of masks from a plurality of objects, and wherein the convolutional neural network comprises repeating decoder blocks comprising up-sampling layers and convolutional layers;
a loss function component that generates a loss function based on the plurality of masks, wherein the loss function is iteratively back propagated to tune parameters of the convolutional neural network;
a classification component that predicts a classification label for an input image based on the convolutional neural network; and
a visualization component that generates a multi-dimensional visualization associated with the classification label for the input data.
2. The machine learning system of claim 1 , wherein the convolutional neural network comprises a pretrained classifier network that outputs convolutional feature maps.
3. The machine learning system of claim 2 , wherein the convolutional neural network comprises a classification/localization network that outputs corresponding feature maps based on the convolutional feature maps.
4. The machine learning system of claim 2 , wherein a size of a mask from the plurality of masks is matched with a size of a convolutional feature map from the convolutional feature maps.
5. The machine learning system of claim 2 , wherein a size of a mask from the plurality of masks is matched with a size of a convolutional feature map from the convolutional feature maps based on a max pooling process.
6. The machine learning system of claim 1 , wherein the loss function component employs a set of convolutional neural networks to generate a localization map.
7. The machine learning system of claim 1 , wherein the loss function component employs a set of convolutional neural networks.
8. The machine learning system of claim 1 , wherein the loss function component generates the loss function based on a probability for a class associated with the plurality of masks.
9. A method that employs decoder blocks for improved performance, comprising using a processor operatively coupled to memory to execute computer executable components to perform the following acts:
receiving a plurality of images associated with a plurality of patients from at least one imaging device;
receiving a plurality of masks from a plurality of objects, wherein each image comprises at least one mask associating an object of interest with a corresponding class label;
training a convolutional neural network based on the plurality of images and the plurality of masks, wherein the convolutional neural network comprises a pretrained classifier network that outputs convolutional feature maps, and a classification/localization network that outputs corresponding feature maps, wherein the classification/localization network comprises repeating decoder blocks comprising up-sampling layers and convolutional layers;
generating a loss function based on the plurality of masks;
iteratively back propagating the loss function to tune parameters of the convolutional neural network;
predicting a classification label for an input image based on the convolutional neural network; and
generating a multi-dimensional visualization associated with the classification label for the input data.
10. The method of claim 9 , further comprising matching a size of a mask from the plurality of masks with a size of a convolutional feature map from the convolutional feature maps.
11. The method of claim 9 , further comprising matching a size of a mask from the plurality of masks with a size of a convolutional feature map from the convolutional feature maps based on a max pooling process.
12. The method of claim 9 , wherein the generating the loss function comprises employing a set of convolutional neural networks to generate a localization map.
13. The method of claim 12 , wherein the generating the loss function comprises modifying a portion of the convolutional neural network.
14. The method of claim 9 , wherein the generating the loss function comprises generating the loss function based on a probability for a class associated with the plurality of masks.
15. A computer readable storage device comprising instructions that, in response to execution, cause a system comprising a processor to perform operations that employ decoder blocks for improved performance, comprising:
receiving a plurality of images associated with a plurality of patients from at least one imaging device;
receiving a plurality of masks from a plurality of objects, wherein each image comprises at least one mask associating an object of interest with a corresponding class label;
training a convolutional neural network based on the plurality of images and the plurality of masks, wherein the convolutional neural network comprises a pretrained classifier network that outputs convolutional feature maps, and a classification/localization network that outputs corresponding feature maps, and wherein the classification/localization network comprises repeating decoder blocks comprising up-sampling layers and convolutional layers;
generating a loss function based on the plurality of masks;
iteratively back propagating the loss function to tune parameters of the convolutional neural network;
predicting a classification label for an input image based on the convolutional neural network; and
generating a multi-dimensional visualization associated with the classification label for the input data.
16. The computer readable storage device of claim 15 , wherein the generating the loss function comprises employing a set of convolutional neural networks to generate a localization map.
17. The computer readable storage device of claim 15 , wherein the generating the loss function comprises generating the loss function based on a probability for a class associated with the plurality of masks.