Our most recently published paper #Nanopyx has been featured in @EutopiaUni's "Breakthroughs: A new series of remarkable stories from our network of universities" In this edition, our research joins the health innovations of the partner universities @VUBrussel and @UPFBarcelona

Today we highlight the recent coverage from @OeirasValley regarding our #Nanopyx paper! The unwavering support from the Oeiras municipality for scientific research and technology plays a crucial role in advancing our work at @itqbnova. oeirasvalley.com/ferramenta-…

#NanoPyx and #LiquidEngine hitting the national news, highlighting our research @itqbnova rtp.pt/noticias/pais/investi…

Efficiently accelerated bioimage analysis with NanoPyx, a Liquid Engine-powered Python framework (relevance: 65%) nature.com/articles/s41592-0… #biophotonics #biomedicaloptics

顕微鏡画像データの高速な分析ワークフローNanoPyxがnature methods誌に発表されました。ノイズ除去、レジストレーション、超解像等が高速にできます。githubでもcolabで動くチュートリアルあり。 nature.com/articles/s41592-0…

Our dream team of researchers that made it possible!💪🏅 #Nanopyx paper in @naturemethods Know more: doi.org/10.1038/s41592-024-0…

Happy new year!! Out today from the @HenriquesLab comes NanoPyx, which optimizes bioimage analysis by dynamically selecting code variations based on input data and hardware to ensures maximum computational efficiency. nature.com/articles/s41592-0…

Accelerating bioimaging analysis with NanoPyx: A Python framework enhanced by Liquid Engine Modern microscopy often produces massive, complex image datasets—where even routine tasks such as drift correction or denoising can become computationally overwhelming. A recent paper by Saraiva et al. introduces NanoPyx, a Python framework tailored to address these challenges efficiently. At its core is the Liquid Engine, a system that adaptively switches among CPU- or GPU-based implementations for each task. By learning from past run times and hardware configurations, it provides data-driven selection of the most efficient parallelization strategy (e.g., OpenCL, CUDA, or CPU threading). This adaptive approach has enabled NanoPyx to achieve up to 24-fold faster processing for methods like super-resolution radial fluctuations (eSRRF) and nonlocal means (NLM) denoising. Beyond low-level code optimizations, NanoPyx’s runtime monitoring ensures it stays consistently high-performing across various hardware setups, from consumer laptops to workstation clusters. Moreover, the framework offers user-friendly Jupyter notebooks and a napari plugin, facilitating integration into existing bioimaging pipelines. Although primarily geared toward advanced microscopy, NanoPyx’s design philosophy could extend to other fields requiring accelerated, scalable data processing. As an open-source solution, it stands out as a flexible and robust platform for contemporary imaging challenges—underscoring the broader relevance and potential of adaptive, data-driven runtime optimization. Paper: nature.com/articles/s41592-0…

What an incredible journey with @Bruno_MSaraiva and @Antonio_DBrito ❤️ I learned so much with these guys and had a blast working with them. This was truly the dream team ❤️🥰 #NanoPyx

#NanoPyx is powered by the #LiquidEngine - a novel machine learning system that dynamically benchmarks and selects the fastest algorithm implementations for each dataset. Whether working on a laptop or a high-end workstation, it adapts to your hardware for maximum efficiency

🚨🔬🤖📜 Paper is out! Introducing the #LiquidEngine and #NanoPyx - novel #AI strategies for accelerating #microscopy that explore how to maximise performance 🚀. Brainchild of @Bruno_MSaraiva, @inesmcunha and @Antonio_DBrito. Faster #SRRF and #eSRRF!! 📰: nature.com/articles/s41592-0…

Great news to start a great year for @Henriques Lab. The @naturemethods paper on #Nanopyx is out!🚀 A game-changing bioimaging tool for super-resolution imaging analysis. Congratulations to @Bruno_MSaraiva, @inesmcunha and @Antonio_DBrito for this remarkable achievement!🎉🎓

HOW CAN YOU USE #NanoPyx ? We recorded a series of 1-minute video tutorials that showcase how you can use #NanoPyx, and the capabilities of the #LiquidEngine on your own image analysis workflows! Video 1: youtu.be/Dx2lHoRB044 (5/n)

What methods does #NanoPyx offer? We have Image Registration, Denoising, Super-Resolution Image Reconstruction🔬 and Quality Control metrics! (4/n)

And to start 2025 strong… Our work #NanoPyx is out in @naturemethods 😍❤️ DOI: doi.org/10.1038/s41592-024-0… #NanoPyx is a Python framework that uses the #LiquidEngine to efficiently accelerate bioimage analysis for any user! (1/n) ⬇️

Look who I found 🥺♥️ @Bruno_MSaraiva !! #NanoPyx team assembled (missing @Antonio_DBrito )

@Bruno_MSaraiva is 🤘🔥 at #SMLMS2024 with #NanoPyx and how to boost any computer power!! It sounds perfect after having heard so much these days about the limitations of sooooo many little points and clusters atlas 👀 look at these great slides!! 😍

Now at #SMLMS2024, @Bruno_MSaraiva showcasing #NanoPyx !! 😍 How we can accelerate image analysis for everyone 💻🔬

I’m very proud to showcase our next speaker @Bruno_MSaraiva presenting NanoPyx, a library aimed to accelerate image analysis and successor to #NanoJ @inesmcunha @Antonio_DBrito @HenriquesLab github.com/HenriquesLab/Nano… @SMLM_Symposium #SMLMS2024

We've implemented Decorrelation and FRC in nanopyx, in case its useful. They also run on Google Colab if you don't want to spend much time setting up a py-env. These are obviously just analytical predictions and nothing beats a real reference structure github.com/HenriquesLab/Nano…