Sketchpose: Learning to Segment Cells with Partial Annotations

TL;DR

Sketchpose introduces a novel cell segmentation method that effectively handles partial annotations, enabling resource-efficient training and transfer learning without compromising accuracy.

Contribution

It presents a new approach that leverages distance maps for cell segmentation with partial annotations, addressing a key limitation of existing fully supervised methods.

Findings

Substantial reduction in annotation time and resources.

Maintains high segmentation accuracy with partial annotations.

Effective in transfer and regular learning scenarios.

Abstract

The most popular networks used for cell segmentation (e.g. Cellpose, Stardist, HoverNet,...) rely on a prediction of a distance map. It yields unprecedented accuracy but hinges on fully annotated datasets. This is a serious limitation to generate training sets and perform transfer learning. In this paper, we propose a method that still relies on the distance map and handles partially annotated objects. We evaluate the performance of the proposed approach in the contexts of frugal learning, transfer learning and regular learning on regular databases. Our experiments show that it can lead to substantial savings in time and resources without sacrificing segmentation quality. The proposed algorithm is embedded in a user-friendly Napari plugin.