Detect Mitosis Using Neural Nets
If you have enough data, you can train a neural network from scratch, a network that automatically learns the relevant features and simultaneously acts as a subsequent classifier.
As an example, take detecting cells that undergo mitosis. Here is a simple convolutional neural network that can do the job.
The data for training and testing has been extracted from the Tumor Proliferation Assessment Challenge 2016. The data has been preprocessed into 97×97 images, centered around the actual cells in question.
Use roughly three-quarters of the data for training and the rest for testing.
A subset of mitosis samples looks like this.
A subset of non-mitosis samples looks like this.
Again, to increase the training set, perform image mirroring and rotation.
Calculate the classifier metrics and verify the effectiveness of the neural network.
Considering the challenging task, an error rate of less than 10% is comparable to what a pathologist would achieve.