Bernhard Kainz's Projects
Use of Attention Gates in a Convolutional Neural Network / Medical Image Classification and Segmentation
depth estimation using tensorflow
docker argo fork
see here http://freegroup.github.io/draw2d_js.app.brainbox
drop2 - intensity-based image registration
FastFlowNet: A Lightweight Network for Fast Optical Flow Estimation (ICRA 2021)
fast non-rigid medical image registration via accelerated optimisation on the manifold of diffeomorphisms
GPU accelerated source code for motion compensation of multi-stack MRI data
A lightweight web dashboard for monitoring GPU usage
The Image Registration Toolkit
Automated lung segmentation in CT
The Medical Image Registration ToolKit (MIRTK), the successor of the IRTK, contains common CMake build configuration files, core libraries, and basic command-line tools. Extension packages are hosted by the MIRTK GitHub group at
The Medical Imaging Interaction Toolkit.
Revision notes for machine learning
AI Toolkit for Healthcare Imaging
GPU accelerated approach to compensate motion from multiple stacks of MRI slices.
These are the lecture slides used at FAU Erlangen-Nuremberg, Germany for the lecture "Medical Engineering". This class gives a complete and comprehensive introduction to the fields of medical imaging systems, as designed for a broad range of applications. The authors of the book first explain the foundations of system theory and image processing, before highlighting several modalities in a dedicated chapter. The initial focus is on modalities that are closely related to traditional camera systems such as endoscopy and microscopy. This is followed by more complex image formation processes: magnetic resonance imaging, X-ray projection imaging, computed tomography, X-ray phase-contrast imaging, nuclear imaging, ultrasound, and optical coherence tomography. Open Access Link to the Text Book: https://link.springer.com/book/10.1007/978-3-319-96520-8#about Link to Video Recordings on YouTube: https://www.youtube.com/watch?v=vvftvjnXzsY&list=PLpOGQvPCDQzsgK1XuhUXO8r9M4WRqhvDf
A Code Release for Mip-NeRF 360, Ref-NeRF, and RawNeRF
VTK-based Data Analysis and Visualization Application
Deep universal probabilistic programming with Python and PyTorch
RAdiological Text Captioning for Human Examined Thoraxes
Raytracer in our favourite spreadsheet application!
The Shader Lab Framework is a teaching tool to solidify the fundamentals of Computer Graphics with OpenGL 4.0 and GLSL.
Model descriptions and weights for all the variations of the SonoNet for real-time fetal standard scan plane detection.