Diagnostic Image Analysis Group's repositories
StreamingCNN
To train deep convolutional neural networks, the input data and the activations need to be kept in memory. Given the limited memory available in current GPUs, this limits the maximum dimensions of the input data. Here we demonstrate a method to train convolutional neural networks while holding only parts of the image in memory.
pathology-he-auto-augment
H&E tailored Randaugment: automatic data augmentation policy selection for H&E-stained histopathology.
Report-Guided-Annotation
Report Guided Annotation
pathology-tiger-algorithm-example
Example algorithm and docker for the TIGER challenge
pathology-whole-slide-packer
extracts tissue sections from one or multiple whole slide images and combines them into a single new slide removing excess white space
AbdomenMRUS-csPCa-CAD-nnUNet
Clinically Significant Prostate Cancer Detection in bpMRI using models trained with Report Guided Annotations
bodyct-dsb2017-grt123
Repository which contains the code of the grt123 solution from the Kaggle DSB 2017 challenge on lung cancer detection
prostateMR-USSL
Uncertainty Aware Self-Supervised Prostate Zonal Segmentation in 3D MRI scans using Probabilistic Attention U-Net..
bodyct-NegBio
:newspaper: High-performance tool for negation and uncertainty detection in radiology reports
cobra
This repo will contain the documentation and splits that belongs to the COBRA dataset hosted on AWS
MultiStainDeepLearning
Code from Foersch et al. (Under Construction / Development)
open-data-registry
A registry of publicly available datasets on AWS
panorama_eval
Evaluation of 3D detection and diagnosis performance —geared towards pancreatic ductal adenocarcinoma detection in CT
pathology-whole-slide-learning
library for weak whole slide learning with attention to compress and classify whole slide images
SPIDER-template
Template for algorithm submission for the SPIDER Challenge
tiger-nnUNet-algorithm-CG
Example algorithm and docker for the TIGER challenge