The suite is built around a , with optional C/C++ extensions for performance‑critical kernels. It follows the FAIR (Findable, Accessible, Interoperable, Re‑usable) principles and integrates seamlessly with other community tools such as Nilearn , MNE‑Python , FSL , SPM , and AFNI . 2. Historical Context | Year | Milestone | |------|-----------| | 2015 | Project conception at EPFL’s Laboratory for Cognitive Neuroimaging (LCN). | | 2016 | First public release (v0.1) on GitHub under the permissive BSD‑3‑Clause license. | | 2018 | Integration of a GPU‑accelerated diffusion‑tensor toolbox (via CUDA). | | 2020 | Introduction of the “Lausanne 2020 ” data‑standardisation layer, aligning with BIDS (Brain Imaging Data Structure). | | 2022 | Full support for containerised deployment (Docker, Singularity) and a cloud‑ready version for AWS/GCP. | | 2024 | Release of TWK Lausanne 2.0 , featuring a modular plugin architecture, a web‑based dashboard, and an extensive Python API. |
# ------------------------------------------------- # 4. Threshold and visualise the contrast # ------------------------------------------------- contrast = glm.contrast('2back > 0back') thresholded = tstat.threshold(contrast, p=0.05, method='fdr') tvis.plot_brain(thresholded, surface='fsaverage', cmap='cold_hot') The same pipeline can be that the web dashboard can execute without writing any code: twk lausanne download
1. What Is TWK Lausanne? TWK Lausanne (short for “Toolkit for Working Knowledge – Lausanne edition” ) is an open‑source, research‑oriented software suite that originated at the École Polytechnique Fédérale de Lausanne (EPFL) in Switzerland. Its primary purpose is to provide a flexible, modular environment for: The suite is built around a , with
pipeline_json = preproc.to_json() tvis.save_dashboard(pipeline_json, out="my_analysis.json") 6.1. GPU‑Accelerated Diffusion Tensor Imaging from twk.diffusion import DTI, cuda_enabled Historical Context | Year | Milestone | |------|-----------|