Sometimes it’s necessary to downgrade packages in google colab to restore compatibility with an older version (e.g. pytorch). One way of doing this is using conda within colab and a few tricks make this work. Here is the colab notebook: https://colab.research.google.com/drive/1W5UnflB8m1zo6lP1y11V78QczMRBI0FE?usp=sharing Here a quick code gist:
Our new preprint is online. NeXtQSM is a deep learning pipeline for data-consistent quantitative susceptibility mapping trained with simulated data build by our CIBIT MPhil student Francesco Cognolato: [2107.07752] NeXtQSM — A complete deep learning pipeline for data-consistent quantitative susceptibility mapping trained with hybrid data (arxiv.org)
Our latest preprint describes the QSMxT pipeline developed by Ashley Stewart – a CIBIT PhD student in our team: QSMxT enables robust & automated quantitative susceptibility mapping and extraction of quantitative information from regions of interest. The code can be found here: https://github.com/QSMxT/QSMxT The preprint is here: QSMxT: Robust Masking Read more…
I was recently invited to join the panel for an @OSFramework webinar on Leveraging Open Ecosystems to Enhance Reproducible Workflows: The webinar covers OSF integration with @protocolsIO, R @aaronwolen, command line interfaces and jupyter notebooks – thank you for the great discussions!
For this year’s AICI artificial intelligence in clinical imaging forum I prepared a tutorial for applying deep learning models to medical imaging data. It was very interesting to prepare the tutorial and see what models exist and how difficult it still is to make them work with real data. Here Read more…
For the 2020 OHBM Brainhack we made a little tutorial explaining git and github. We hope it’s useful and thanks to the organizers for putting this great event together 🙂 Video: Slides:
For a paper we recently submitted (Improving FLAIR SAR efficiency at 7T by adaptive tailoring of adiabatic pulse power using deep convolutional neural networks – pre-print here: https://arxiv.org/abs/1911.08118 and published article here: https://doi.org/10.1002/mrm.28590) I was wondering if we can do more than just provide the source code. One problem I Read more…
Motion dominates the contribution to variance in fMRI time series and it is therefore important to account for this variability correctly. Currently, most correction schemes use a rigid body realignment procedure, but interactions with magnetic field inhomogeneities and physiological fluctuations lead to non-linear deformations. Non-linear realignment increased spatial resolution by Read more…