Label-Free Observation and Quantitative Analysis of Lipid Droplets in Senescent Cells Using Holotomography

2025 ASBMB Deuel poster_LD in Senescent cells_Maeve thumb nail.png

Extends senescence-imaging work beyond lipid droplets to whole-cell and vesicle-level organelle quantification in Hs68 fibroblasts, introducing a custom "SE AI model" that corrects over-segmentation errors seen with existing AI models on large, flattened senescent cells.

What's inside

- Comparison of an existing AI segmentation model (which erroneously splits large senescent cells/nuclei into two objects) vs. a custom-trained "SE AI model" that segments them correctly

- Volume and concentration quantification for normal vs. senescent cells (both highly significant, P < 0.001)

- Vesicle-like structure segmentation pipeline and per-cell quantification (volume ratio, area ratio) comparing passage 2 (normal) vs. passage 27 (senescent) cells

- Discussion of applications to mitochondrial dynamics, lysosomal function, autophagy, and ferroptosis research

Keywords

cellular senescence organelle quantification custom AI segmentation model holotomography Hs68 fibroblast vesicle analysis label-free imaging TomoAnalysis refractive index tomography aging research over-segmentation correction

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.