Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Full segmentation annotations of 3D time-lapse microscopy images of MDA231 cells

Aleksandra Melnikova, Petr Matula

TL;DR

The paper addresses the lack of publicly available fully volumetric 3D time-lapse cell annotations by presenting fully volumetric annotations for two Fluo-C3DL-MDA231 sequences, created by three annotators on the first sequence and one on the second. It validates these annotations against gold tracking truth and 2D gold segmentation truth, demonstrating consistency with TRA and superior coverage compared to silver truth. The authors analyze segmentation performance using SEG and inter-annotator variability, showing that majority voting improves annotation quality and proposing future work to collect additional independent annotations for both sequences. The dataset provides a valuable resource for training and evaluating 3D cell segmentation and tracking, as well as for 3D shape analysis and fusion method development.

Abstract

High-quality, publicly available segmentation annotations of image and video datasets are critical for advancing the field of image processing. In particular, annotations of volumetric images of a large number of targets are time-consuming and challenging. In (Melnikova, A., & Matula, P., 2025), we presented the first publicly available full 3D time-lapse segmentation annotations of migrating cells with complex dynamic shapes. Concretely, three distinct humans annotated two sequences of MDA231 human breast carcinoma cells (Fluo-C3DL-MDA231) from the Cell Tracking Challenge (CTC). This paper aims to provide a comprehensive description of the dataset and accompanying experiments that were not included in (Melnikova, A., & Matula, P., 2025) due to limitations in publication space. Namely, we show that the created annotations are consistent with the previously published tracking markers provided by the CTC organizers and the segmentation accuracy measured based on the 2D gold truth of CTC is within the inter-annotator variability margins. We compared the created 3D annotations with automatically created silver truth provided by CTC. We have found the proposed annotations better represent the complexity of the input images. The presented annotations can be used for testing and training cell segmentation, or analyzing 3D shapes of highly dynamic objects.

Full segmentation annotations of 3D time-lapse microscopy images of MDA231 cells

TL;DR

The paper addresses the lack of publicly available fully volumetric 3D time-lapse cell annotations by presenting fully volumetric annotations for two Fluo-C3DL-MDA231 sequences, created by three annotators on the first sequence and one on the second. It validates these annotations against gold tracking truth and 2D gold segmentation truth, demonstrating consistency with TRA and superior coverage compared to silver truth. The authors analyze segmentation performance using SEG and inter-annotator variability, showing that majority voting improves annotation quality and proposing future work to collect additional independent annotations for both sequences. The dataset provides a valuable resource for training and evaluating 3D cell segmentation and tracking, as well as for 3D shape analysis and fusion method development.

Abstract

High-quality, publicly available segmentation annotations of image and video datasets are critical for advancing the field of image processing. In particular, annotations of volumetric images of a large number of targets are time-consuming and challenging. In (Melnikova, A., & Matula, P., 2025), we presented the first publicly available full 3D time-lapse segmentation annotations of migrating cells with complex dynamic shapes. Concretely, three distinct humans annotated two sequences of MDA231 human breast carcinoma cells (Fluo-C3DL-MDA231) from the Cell Tracking Challenge (CTC). This paper aims to provide a comprehensive description of the dataset and accompanying experiments that were not included in (Melnikova, A., & Matula, P., 2025) due to limitations in publication space. Namely, we show that the created annotations are consistent with the previously published tracking markers provided by the CTC organizers and the segmentation accuracy measured based on the 2D gold truth of CTC is within the inter-annotator variability margins. We compared the created 3D annotations with automatically created silver truth provided by CTC. We have found the proposed annotations better represent the complexity of the input images. The presented annotations can be used for testing and training cell segmentation, or analyzing 3D shapes of highly dynamic objects.

Paper Structure

This paper contains 9 sections, 2 figures, 4 tables.

Table of Contents

  1. Introduction
  2. Related work
  3. Terminology
  4. Data collection
  5. Dataset Characteristics
  6. Segmentation score
  7. Inter-annotator variability
  8. Discussion and future work
  9. Conclusion

Figures (2)

  • Figure 1: A 3D visualization of one cell mask that consists of 5 slices (w.r.t. the z-axis). The first row shows a 3D visualization of the 3 annotations: (A) Gold reference tracking annotations (TRA), (B) Silver reference segmentation annotations (ST), and (C) the proposed full annotations (FA). Image (D) shows all five 2D slices of the cell. Image (D) shows a 3D visualization of the slices (D). It is visible that thin protrusions are absent in images (A) and (B).
  • Figure 2: Overlay of 3 annotations (S01_TRA and S01_ST, and S01_FA_MV) over a raw image from the Fluo-C3DL-MDA231 dataset for a random 2D slice. The Figure shows that silver truth annotations have problems with the segmentation of thin protrusions (a yellow dashed bounding box denotes problematic regions).