A Multi-Scale Inception-UNet with Structure-Aware Evaluation for Branch-Preserving Segmentation of Organoids

Sandra H. Andrusca, Christopher D. Kießling, Andreas R. Bausch

Keywords: medical image analysis, organoid segmentation, branched organoids, multi-scale features, structure-aware loss, topology preservation, microscopy imaging

TL;DR: We propose a multi-scale Inception-UNet for segmenting complex branched organoids and demonstrate stable gains over classical UNet models.

Abstract:

Branched organoids exhibit increasingly complex morphologies as they progress from simple spheroid states to highly ramified structures, making topology-preserving segmentation essential for quantitative biological analysis. Capturing thin protrusions and maintaining branch continuity remains challenging for classical UNet-based architectures, particularly in brightfield imaging where fine structures are easily blurred or disconnected.

In this work, we present a multi-scale Inception-UNet designed to capture the heterogeneous spatial scales of branched organoids through parallel convolutional paths with complementary receptive fields. As a model system, we analyze brightfield pancreatic ductal adenocarcinoma (PDAC) organoids, a system known for strong morphological heterogeneity and invasive branching behavior, cultured using high-throughput Patternoid assays that enable standardized imaging and robust quantitative analysis.

To assess segmentation quality beyond region overlap, we combine Dice with a structure-aware skeleton-based Dice score that directly probes branch integrity and topological continuity. Across deterministic seeds and strictly separated organoid positions, the Inception-UNet achieves the highest region-based Dice ($0.8035 \pm 0.0076$) and skeleton-based Dice ($0.2513 \pm 0.0156$), and most importantly, the strongest preservation of branch continuity compared to UNet and UNet++. These improvements become increasingly pronounced with growing morphological complexity.

Overall, our results demonstrate that multi-scale feature extraction combined with topology-aware evaluation substantially improves segmentation of branched organoids and provides a robust foundation for downstream morphological and invasion-related analyses.

Primary Subject Area: Segmentation

Secondary Subject Area: Learning with Noisy Labels and Limited Data

Registration Requirement: Yes

Visa & Travel: No

Read CFP & Author Instructions: Yes

Originality Policy: Yes

Single-blind & Not Under Review Elsewhere: Yes

LLM Policy: Yes

Submission Number: 394