Segmentation and Analysis Method for Mitochondria Enhanced with Open-Source AI Tools in Holotomography Images

2025.01.12 Keystone Symposia_Mitochondria length analysis thumb nail.png

Presents an AI-integrated TomoAnalysis pipeline (StarDist for cell segmentation, ilastik for mitochondria prediction) for per-cell mitochondrial length/morphology analysis, applied across 8 different cell lines and to drug- and time-dependent fission/fusion dynamics in Hep3B cells.

What's inside

- Three-stage pipeline: StarDist-based single-cell segmentation, ilastik-based mitochondria segmentation, and per-cell mitochondria assignment/measurement

- Drug-induced (Mdivi-1/MPP+) and 32-minute time-lapse mitochondrial length-class quantification (fragmented/normal/elongated/outlier)

- Cross-cell-line validation of 8 customized ilastik models (MCF-7, Hep3B, HeLa, Fibroblast, NIH-3T3, Huh-7, COS-7, HepG2) with representative segmentation images

- Violin-plot comparison of mitochondrial length distribution between Hep3B and Fibroblast cell lines

Keywords

mitochondria segmentation ilastik StarDist open-source AI tools holotomography mitochondrial length analysis multi-cell-line validation label-free imaging refractive index tomography TomoAnalysis

Download the Application Note

First name*

Last name*

Title

Company*

Region*

Country*

Interested Product*

E-mail*

Comment

I have read and agree to the Terms and Policies.