Advancing ICK Assay: Real-time, Label-free Imaging of Lymphocyte Subsets

2024 WBC_Poster_ICK Assay_HYLEE thumb nail.png

Applies a deep-learning classifier (DenseNet121) to 3D refractive index images of human PBMCs to enable label-free classification of lymphocyte subsets (monocytes, granulocytes, B/T/NK cells, CD4/CD8 T cells) for use in Immune Cell Killing (ICK) assays, avoiding the need for antibody staining during dynamic effector–target interaction studies.

What's inside

- Three-stage PBMC classification pipeline (Stage 1: monocyte/granulocyte/others; Stage 2: T/B/NK cells; Stage 3: CD4 vs. CD8) with confusion matrices for each stage

- Per-subtype cell morphology panels (3D raw, MIP, overall RI, nucleus) for CD4, CD8, CD19, CD16&56, CD14, CD15

- Accuracy comparison of 2D digital holographic microscopy (76.2%) vs. 3D holotomography (90.2%) for CD4/CD8 discrimination

- HeLa–PBMC co-culture 3D rendering demonstrating simultaneous PBMC classification and effector–target interaction visualization in one field of view

Keywords

ICK assay immune cell killing assay lymphocyte subset classification label-free PBMC imaging holotomography deep learning classification CD4 CD8 T cells NK cell imaging immunology refractive index tomography DenseNet

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.