Michał Januszewski

👀 For more than 15 years, Janelia researchers have been on a mission to catalogue the neurons of the fly optic lobes & develop genetic tools to study them. This effort has culminated in a complete inventory of all the visual neurons in the fruit fly ➡️ www.nature.com/articles/d41...

🐁🧠 Results are for a mouse cortex IBEAM-mSEM volume. Work is a collaboration between @GoogleAI & the Hess lab at @HHMIJanelia, who pioneered this EM technique. More soon!

We find an OOM reduction in splits while maintaining mergers low enough that the traced axons should be useful for automated analysis or serve as an excellent starting point for further manual corrections.

We focus on the most challenging type of reconstruction (axon tracing) and evaluate PATHFINDER on a scale that ensures statistical significance and lack of bias (over 4 m total path length, all axons in the volume).

FFN v1.5 segments the volume EM images. Then SENSE defines the agglomeration space. Finally, SHAPE guides an efficient combinatorial search for optimal, morphologically plausible reconstructions.

PATHFINDER is an AI system designed to address this. It uses specialized models (FFN v1.5, SENSE, SHAPE) to process data across progressively larger spatial scales, currently up to the range of tens of microns.

Connectomics has made huge strides in recent years. But we can acquire data far faster than we can analyze it. With even insect brains requiring tens of years of manual corrections, proofreading remains a critical and expensive bottleneck, and a mouse brain seems out of reach.

Wouldn't it be great if we could not only image large connectomic volumes but also completely reconstruct them? And if a whole mouse brain project didn't cost billions?

With the PATHFINDER preprint (www.biorxiv.org/content/10.1...), we preview a future where it doesn't have to.

We'll present our #ICLR2025 spotlight on ZAPBench this afternoon: 📍 Hall 3 #61!

ZAPBench ZAPBench evaluates how well different models can predict the activity of over 70,000 neurons in a novel larval zebrafish dataset.

⚡️ Excited to introduce ZAPBench, our #ICLR2025 spotlight: The Zebrafish Activity Prediction Benchmark measures progress in predicting neural activity within an entire vertebrate brain (70k+ neurons!)

Explore interactive visualizations, datasets, code + paper: google-research.github.io/zapbench

🧠🧪