1. Pipeline: (3D) Lipid Droplet per Cell

This pipeline segments and quantifies lipid droplets (LDs) on a per-cell basis using a rule-based algorithm optimized for holotomography (HT) data. It first identifies individual cells and then analyzes the lipid droplets contained within each cell, enabling cell-specific quantification of LD count, volume, and spatial distribution.

By assigning lipid droplets to individual cells, this pipeline provides detailed insight into cell-to-cell variability in lipid accumulation. Compared to AI-based methods, the rule-based approach allows for faster analysis and is particularly effective when cells are well-separated and not overly confluent.

Due to its reliance on cell segmentation, the pipeline is best suited for low-confluency datasets where individual cell boundaries can be clearly distinguished. In densely packed or tightly clustered samples, segmentation accuracy may be reduced, and the use of global LD analysis pipelines is recommended instead.

Output: Single-cell masks and measurements, Labeled lipid droplets with individual measurements, Per-cell LD metrics




2. Pipeline: (2D) Lipid Droplet per Cell 

This pipeline performs per-cell lipid droplet (LD) analysis in 2D projections using a fully rule-based approach. Lipid droplets are segmented based on their high refractive index (RI) values and round morphology, while cell segmentation is also carried out using rule-based algorithms without AI assistance.

The absence of AI enables faster processing, making the pipeline ideal for high-throughput applications where speed and efficiency are critical. It is best suited for low-confluency datasets with well-separated cells, where it delivers accurate and consistent results. For more complex or crowded cell images, complementary AI-based pipelines are available to ensure optimal segmentation performance.

Output: Single-cell masks, Labeled lipid droplets with individual measurements, Per-cell LD measurements

 


Access to analysis pipelines is available upon request. Please send your inquiry to info@tomocube.com to explore the options best suited to your needs.