🎉Congratulations to our researchers Mariana Ferreira and
@gomez-mariscal.bsky.social for their #ReScale4DL work available on pre-print at @biorxivpreprint.bsky.social🧐 Find out how down-sampling data can improve your high-resolution imaging at biorxiv.org/content/10.1...

#BioImageAnalysis #DL #AI

ReScale4DL: Balancing Pixel and Contextual Information for Enhanced Bioimage Segmentation [new]
Lower res bioimages boost DL seg, balancing detail & context for better efficiency.

In #ReScale4DL optimisation, we show that high resolution (low pixel-size) can lead to suboptimal DL performance🧐 (surprised?)
Good news: downsampling images, or going to lower resolutions in acquisition, is so easy (and cheaper!) and can indeed improve imaging throughput, yay!!!😉

Following the release of @gomez-mariscal.bsky.social and Mariana Ferreira #ReScale4DL work, Estibaliz shared her view on "FAIR and high-quality AI-ready data for bioimage analysis" on the 4-part @ai4life.bsky.social webinar series in collab with #GloBIAS 📅 Don´t miss the next one at: t.co/9TdywIGrgy

🚨Alô #bsky and #microscopy community, we got #ReScale4DL, our new work in
@aiopticalbiolab.bsky.social
, out today!😍 🥳
We present the #imageresolutionparadox phenomenon for #DL-results 🤓
👇Check it out!!

ReScale4DL: Balancing Pixel and Contextual Information for Enhanced Bioimage Segmentation https://www.biorxiv.org/content/10.1101/2025.04.09.647871v1

ReScale4DL: Balancing Pixel and Contextual Information for Enhanced Bioimage Segmentation https://www.biorxiv.org/content/10.1101/2025.04.09.647871v1

Why do lower-resolution images sometimes yield better results in deep learning for bioimaging analyses? 🤔📉

Mariana Ferreira's new preprint on #ReScale4DL explores this paradox and introduces optimal resolution design! @gomez-mariscal.bsky.social brainchild🧑‍🔬✨

Check: www.biorxiv.org/content/10.1...