This is the second to last blog post in the series, (the first one here, second one here), where we will go into greater detail about how we can use Ray Serve to set up a server waiting to respond to our requests for processing. These last two are the most complex blogpost in the series and require some understanding of how HTTP, REST, and web services work. You can find relevant prereading here.
Ray Serve is a scalable model serving library for building online inference APIs. Serve is framework agnostic, so you can use a single toolkit to serve everything from deep learning models built with frameworks like PyTorch, Tensorflow, and Keras, to Scikit-Learn models, to arbitrary Python business logic.
In digital pathology, input data is often exceedingly too large for DL models to process directly, with Whole Slide Images (WSI) around 100k x 100k pixels. This post provides a quantitative and qualitative method, with code, to help optimize important digital pathology specific hyperparameters: patch size and magnification. Optimizing these variables can decrease training times, lowers hardware requirements, and reduces the amount of data required to effectively train a model.
This is an updated version of the previously described workflow on how to load and classify annotations/detections created in QuPath for usage in downstream machine learning workflows. The original post described how to use the Groovy programming language used by QuPath to export annotations/detections as GeoJSON from within QuPath, made use of a Python script to classify them, and lastly used another Groovy script to reimport them. If you are not familiar with QuPath and/or its annotations you should probably read the original post first to provide better context and understanding of the respective workflows, as well as being able to appreciate the more elegant approach taken here. If you are already using the described approach, you should be able to easily modify it to follow this newer approach.
The manual labeling of large numbers of objects is a frequent occurrence when training deep learning classifiers in the digital histopathology domain. Often this can become extremely tedious and potentially even insurmountable.
To aid people in this annotation process we have developed and released Quick Annotator (QA), a tool which employs a deep learning backend to simultaneously learn and aid the user in the annotation process. A pre-print explaining this tool in more detail is available [here].
Utilization of current GPUs is often limited by the ability to get the data onto and off the device quickly. More precisely, this means taking data from the host RAM, transferring it over the PCI-e bus to the GPU RAM is the bottleneck of many deep learning use cases.
Deep learning (DL) models have been performing exceptionally well on a number of challenging tasks lately. Unfortunately, given the current blackbox nature of these DL models, it is difficult to try and “understand” what the network is seeing and how it is making its decisions. Building upon our previous post discussing how to train a DenseNet for classification, we discuss here how to apply various visualization techniques to enable us to interrogate the network. The code here is designed as drop-in functionality for any network trained using the previous post, hopefully easing the burden of its implementation.
In this blog post, we discuss how to train a DenseNet style deep learning classifier, using Pytorch, for differentiating between different types of lymphoma cancer. This post and code are based on the post discussing segmentation using U-Net and is thus broken down into the same 4 components:
In this blog post, we discuss how to train a U-net style deep learning classifier, using Pytorch, for segmenting epithelium versus stroma regions. This post is broken down into 4 components following along other pipeline approaches we’ve discussed in the past:
This model focuses on using solely Python and freely available tools (i.e., no matlab).
This blog post assumes moderate knowledge of convolutional neural networks, depending on the readers background, our JPI paper may be sufficient, or a more thorough resource such as Andrew NG’s deep learning course.
Just wanted to take a moment and share some quick stain normalization type experimental results. We have a trained in-house nuclei segmentation model which works fairly well when the test images have similar stain presentation properties, but when new datasets arrive which are notably different we tend to see a decreased classifier performance.
Here we look at one of these images and ways of improving classifier robustness.