Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

SegmentAnything helps microscopy images based automatic and quantitative organoid detection and analysis

Xiaodan Xing, Chunling Tang, Yunzhe Guo, Nicholas Kurniawan, Guang Yang

TL;DR

The paper addresses the bottleneck of manual, area-wide organoid morphology analysis by introducing a SegmentAnything–based pipeline to automatically detect individual organoids in bright-field microscopy. It couples SAM with targeted post-processing and computes five morphometrics, including non-smoothness and non-circularity, enabling fully automated quantification without task-specific annotations. In validation against StarDist on hiPSC-derived neural-epithelial organoids, SAM demonstrated superior segmentation performance and morphometric agreement with manual measurements, with results consistent across varying extracellular matrix conditions. The approach accelerates organoid morphology studies and is released as open-source tooling (SAM4organoid).

Abstract

Organoids are self-organized 3D cell clusters that closely mimic the architecture and function of in vivo tissues and organs. Quantification of organoid morphology helps in studying organ development, drug discovery, and toxicity assessment. Recent microscopy techniques provide a potent tool to acquire organoid morphology features, but manual image analysis remains a labor and time-intensive process. Thus, this paper proposes a comprehensive pipeline for microscopy analysis that leverages the SegmentAnything to precisely demarcate individual organoids. Additionally, we introduce a set of morphological properties, including perimeter, area, radius, non-smoothness, and non-circularity, allowing researchers to analyze the organoid structures quantitatively and automatically. To validate the effectiveness of our approach, we conducted tests on bright-field images of human induced pluripotent stem cells (iPSCs) derived neural-epithelial (NE) organoids. The results obtained from our automatic pipeline closely align with manual organoid detection and measurement, showcasing the capability of our proposed method in accelerating organoids morphology analysis.

SegmentAnything helps microscopy images based automatic and quantitative organoid detection and analysis

TL;DR

The paper addresses the bottleneck of manual, area-wide organoid morphology analysis by introducing a SegmentAnything–based pipeline to automatically detect individual organoids in bright-field microscopy. It couples SAM with targeted post-processing and computes five morphometrics, including non-smoothness and non-circularity, enabling fully automated quantification without task-specific annotations. In validation against StarDist on hiPSC-derived neural-epithelial organoids, SAM demonstrated superior segmentation performance and morphometric agreement with manual measurements, with results consistent across varying extracellular matrix conditions. The approach accelerates organoid morphology studies and is released as open-source tooling (SAM4organoid).

Abstract

Organoids are self-organized 3D cell clusters that closely mimic the architecture and function of in vivo tissues and organs. Quantification of organoid morphology helps in studying organ development, drug discovery, and toxicity assessment. Recent microscopy techniques provide a potent tool to acquire organoid morphology features, but manual image analysis remains a labor and time-intensive process. Thus, this paper proposes a comprehensive pipeline for microscopy analysis that leverages the SegmentAnything to precisely demarcate individual organoids. Additionally, we introduce a set of morphological properties, including perimeter, area, radius, non-smoothness, and non-circularity, allowing researchers to analyze the organoid structures quantitatively and automatically. To validate the effectiveness of our approach, we conducted tests on bright-field images of human induced pluripotent stem cells (iPSCs) derived neural-epithelial (NE) organoids. The results obtained from our automatic pipeline closely align with manual organoid detection and measurement, showcasing the capability of our proposed method in accelerating organoids morphology analysis.
Paper Structure (4 sections, 4 figures)

This paper contains 4 sections, 4 figures.

Table of Contents

  1. Description of purpose
  2. Method
  3. Results
  4. Conclusions

Figures (4)

  • Figure 1: Overview of the proposed method.
  • Figure 2: Challenges in directly applying SegmentAnything in real-world organoid morphology analysis workflow.
  • Figure 3: (a) The average detection scores comparison between our method and the StarDist. (c) and (d) are the segmentation results from our method and the StarDist algorithm, respectively.
  • Figure 4: The analysis results among four groups. In (a) we presented the representative image patches from four different groups. * in figure (b-f) represents a significant difference using Student t-test.