Cxr8 Explorer
Mapping the Latent Space of Patient Geometry
The CXR8 dataset is a collection of chest x-ray images that has been produced by the NIH Clinical Center. It includes a total of 220,000 grayscale images, with a resolution of 1024x1024 pixels. The images have been labeled as either normal or abnormal, with abnormalities ranging from common respiratory diseases such as pneumonia and tuberculosis, to other conditions such as cancer and cardiac abnormalities.
Cxr8Explorer is an educational application that allows users to view individual chest x-ray images, as well as to browse and filter the images based on various criteria such as image quality, patient age, and the presence of specific abnormalities. It includes a model that allows navigation of the latent space of patient geometries, using our "anat-o-mixer" control.
Theory of Operation of the Model
The model of interest arises from an analogy between a locally-connected variational auto-encoder and the mammalian visual system.
Shifter dynamic routing circuit
Van Essen's Distributed Hierarchy
Mumford's dynamic blackboard model of the thalamus
AnatoMixer
The first interactive visualization is the slider-based anatomixer control
***BUT WAIT*** There is also the merged Cxr8Explorer for different platforms.
***IN THE MEANTIME*** the streamlit app shows a generative model
Visualization vs. Quantitative Imaging
As an example of how visualization of latent spaces can be used to assist clinical judgement, trend graphs for motion management.
The issue with QI is that clinical data is needed...