GeneCard-BioImage provides:

• 1. Objective and automated analysis for H&E slide images.

• 2. Objective and automated analysis for PAS slide images.

• 3. Objective and automated analysis for immunofluorescence images.

• 4. Objective cell distance analysis for multiplex immunofluorescence/spatial proteomics slide images.

Objective image analysis can avoid the various shortcomings of subjective analysis，such as (1) arbitrary selection of image regions; (2) standard differences between different analysts; (3) standard differences in the same analyst at different times. Therefore, it can ensure the reproducibility of the results.

1. Objective and automated workflow for H&E slide images

1.1 Principle

The inflammation caused by various reasons is typically accompanied by leukocyte infiltration¹, or intuitively understood as an increase in cells at the lesion site. This is usually identified through H&E staining of tissue sections².

This workflow achieves objective and automated analysis of H&E slides by counting the relative total number of cells (nuclei) and the relative total area of nuclei in the H&E slides.

1.2 Workflow and example

Note：The tissue slide images used in this example were scanned using the Pannoramic 250 FLASH digital pathology scanner from 3DHISTECH. A 20× brightfield objective lens was used to obtain full-slide high-resolution digital images, which were saved in the .mrxs format for further analysis. The file size of the .mrxs format for this example slide is approximately 600 MB.

2. Objective and automated workflow for PAS slide images

2.1 Principle

PAS staining is mainly used to stain structures containing a high proportion of carbohydrates (including glycogen)³, such as the identification of mucus secretion in a mouse asthma model⁴.

This workflow achieves objective and automated analysis of PAS slides counting the relative total number of PAS-positive granules and the relative total area of PAS-positive granules in the PAS slides.

2.2 Workflow and example

Note：The tissue slide images used in this example were scanned using the Pannoramic 250 FLASH digital pathology scanner from 3DHISTECH. A 20× brightfield objective lens was used to obtain full-slide high-resolution digital images, which were saved in the .mrxs format for further analysis. The file size of the .mrxs format for this example slide is approximately 600 MB.

3. Objective and automated workflow for immunofluorescence image

3.1 Principle

Immunofluorescence (IF) is an important immunochemical technique that enables the detection and localization of various antigens in different types of tissues and cell preparations⁵, such as analyzing intracellular bacteria and host cell interactions⁶, or for cancer immunotherapy⁷.

This workflow involves unbiasedly selecting all intracellular bacteria in the field of view (image), followed by signal intensity analysis in specific light channel regions and colocalization analysis, to achieve objective and automated immunofluorescence image analysis.

3.2 Workflow and example

Note: The confocal images used in this example were obtained using a Leica confocal microscope with a 40× objective, and the images are saved in tif format.

4. Objective cell distance analysis workflow for multiplex immunofluorescence/spatial proteomics slide images

4.1 Principle

Multiplex immunofluorescence/spatial proteomics staining can be used to analyze protein spatial expression, localization, interactions, as well as the spatial distribution and interactions of cells⁸. For example, Franken et al. conducted a spatial proteomics comparative analysis using a 62-plex panel (Akoya Biosciences, USA) to assess the differences in the tumor microenvironment and cellular interactions in head and neck squamous cell carcinoma patients who received anti-PD-L1 + anti-CTLA4 or anti-PD-L1 treatment⁹.

This workflow involves unbiasedly selecting all cells from the entire slide (image), followed by cell type classification and cell localization, to achieve objective multiplex immunofluorescence/spatial proteomics cell distance analysis.

4.2 Workflow and example

Note: This workflow is adapted from¹⁰.

References

1.

Coussens, L. M. & Werb, Z. Inflammation and cancer. Nature 420, 860–867 (2002).

2.

Haan, K. de et al. Deep learning-based transformation of h&e stained tissues into special stains. Nature Communications 12, (2021).

3.

Singh, I., Weston, A., Kundur, A. & Dobie, G. Chapter 1 - introduction. in Haematology case studies with blood cell morphology and pathophysiology 1–8 (Elsevier, 2017). doi:10.1016/b978-0-12-811911-2.00001-5.

4.

Khumalo, J., Kirstein, F., Scibiorek, M., Hadebe, S. & Brombacher, F. Therapeutic and prophylactic deletion of IL‐4Ra‐signaling ameliorates established ovalbumin induced allergic asthma. Allergy 75, 1347–1360 (2020).

5.

Im, K., Mareninov, S., Diaz, M. F. P. & Yong, W. H. An introduction to performing immunofluorescence staining. in Biobanking 299–311 (Springer New York, 2018). doi:10.1007/978-1-4939-8935-5_26.

6.

Franco, L. H. et al. The ubiquitin ligase Smurf1 functions in selective autophagy of mycobacterium tuberculosis and anti-tuberculous host defense. Cell Host & Microbe 21, 59–72 (2017).

7.

Lin, J.-R. et al. High-plex immunofluorescence imaging and traditional histology of the same tissue section for discovering image-based biomarkers. Nature Cancer 4, 1036–1052 (2023).

8.

Lundberg, E. & Borner, G. H. H. Spatial proteomics: A powerful discovery tool for cell biology. Nature Reviews Molecular Cell Biology 20, 285–302 (2019).

9.

Franken, A. et al. CD4+ t cell activation distinguishes response to anti-PD-L1+anti-CTLA4 therapy from anti-PD-L1 monotherapy. Immunity 57, 541–558.e7 (2024).

10.

Franken, A., Bila, M. & Lambrechts, D. Protocol for whole-slide image analysis of human multiplexed tumor tissues using QuPath and r. STAR Protocols 5, 103270 (2024).