Toward Generalizable Whole Brain Representations with High-Resolution Light-Sheet Data

2026-03-31

AI Summary

Existing vision models struggle to generalize to whole-brain light-sheet microscopy data at subcellular resolution and lack scalable petabyte-level analysis pipelines. To address this, this work introduces CANVAS—a benchmark comprising the first large-scale, subcellular-resolution whole-mouse-brain light-sheet imaging dataset, featuring six neural and immune cell markers, comprehensive whole-brain cellular annotations, and a standardized evaluation framework. The benchmark exposes critical generalization limitations of current models across diverse phenotypes and anatomical locations, while establishing a high-resolution foundational platform for whole-brain cell detection and classification tasks. CANVAS thus catalyzes the development of foundation models tailored to the unique challenges of such large-scale, high-resolution neuroimaging data.

Technology Category

• Computer Vision: Large Vision Models
• Machine Learning: Multimodal Learning
• Humans and AI: Brain-Sensing and Analysis

Application Category

• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Web Mining and Content Analysis: Large pretrained models with web data
• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets

Abstract

Unprecedented visual details of biological structures are being revealed by subcellular-resolution whole-brain 3D microscopy data, enabled by recent advances in intact tissue processing and light-sheet fluorescence microscopy (LSFM). These volumetric data offer rich morphological and spatial cellular information, however, the lack of scalable data processing and analysis methods tailored to these petabyte-scale data poses a substantial challenge for accurate interpretation. Further, existing models for visual tasks such as object detection and classification struggle to generalize to this type of data. To accelerate the development of suitable methods and foundational models, we present CANVAS, a comprehensive set of high-resolution whole mouse brain LSFM benchmark data, encompassing six neuronal and immune cell-type markers, along with cell annotations and a leaderboard. We also demonstrate challenges in generalization of baseline models built on existing architectures, especially due to the heterogeneity in cellular morphology across phenotypes and anatomical locations in the brain. To the best of our knowledge, CANVAS is the first and largest LSFM benchmark that captures intact mouse brain tissue at subcellular level, and includes extensive annotations of cells throughout the brain.

Problem

• whole-brain imaging
• light-sheet microscopy
• generalization
• subcellular resolution
• large-scale data analysis

Innovation

• light-sheet fluorescence microscopy
• whole-brain imaging
• subcellular resolution
• benchmark dataset
• model generalization