ImPartial: Multi-channel Whole-Cell Segmentation using Partial Annotations
Quick Take
ImPartial is a deep learning framework that achieves state-of-the-art cell segmentation with minimal annotations, matching fully supervised models while requiring significantly fewer expert labels. It employs self-supervised multi-channel quantized imputation, demonstrating consistent performance improvements on multiplexed cellular imaging datasets. The framework is available on GitHub.
Key Points
- ImPartial uses sparse scribbles and limited supervision for cell segmentation.
- Achieves performance comparable to fully supervised models with fewer annotations.
- Demonstrates consistent improvements on benchmark multiplexed cellular imaging datasets.
- Employs self-supervised classification to align with segmentation goals.
- All datasets and code are publicly available on GitHub.
Article Content
From source RSS / original summaryarXiv:2605. 24128v1 Announce Type: new Abstract: Accurate cell segmentation in pathology images typically requires dense pixel-wise annotations, which are costly and time-consuming to obtain. This challenge is especially important for emerging biological imaging modalities and multiplexed datasets with variable channel configurations, where expert-labeled data are scarce.
In this work, we introduce ImPartial, a deep learning framework designed to achieve state-of-the-art segmentation performance in low-annotation regimes using sparse scribbles and limited supervision. ImPartial augments the segmentation objective via self-supervised multi-channel quantized imputation.
This approach leverages the observation that perfect pixel-wise reconstruction or denoising of the image is not needed for accurate segmentation, and thus, introduces a self-supervised classification objective that better aligns with the overall segmentation goal. We demonstrate that ImPartial achieves performance at par with fully supervised models while requiring substantially fewer annotations.
Extensive experiments on benchmark multiplexed cellular imaging and single-plex clinical brightfield immunohistochemistry datasets show consistent improvements over strong baselines with only partial annotations. All benchmark datasets and code are available via our Github: https://github. com/nadeemlab/ImPartial.
Reader Mode unavailable (could not extract clean content).
Want this in your inbox every morning?
Daily brief at your local 8am — bilingual EN/中文, free.
More from arXiv cs.CV
See more →Evi-Steer: Learning to Steer Biomedical Vision-Language Models through Efficient and Generalizable Evidential Tuning
Evi-Steer introduces a novel evidential tuning framework for BiomedCLIP, achieving 0.11% parameter updates while enhancing uncertainty-aware fine-tuning. It outperforms state-of-the-art methods across 15 biomedical imaging datasets, proving effective in few-shot learning and domain shifts for clinical applications.