INSM1

Decision Support & AI

Insulinoma-associated protein 1 (INSM1) is a zinc-finger transcriptional factor which is abundantly expressed in fetal neuroendocrine developmental tissues. However there is also a high occurrence of INSM1 found in neuroendocrine tumors (NETs), such as small cell lung cancer (SCLC), pituitary tumor, medullary thyroid carcinoma, and pheochromocytoma [1]. This analysis module has been created to investigate the clinical utility of INSM1 as a new diagnostic marker for NETs.

Back to Other Samples

Keywords

Neuroendocrine Tumors, INSM1, Image Analysis.

Methods

The techniques of nuclear classification algorithm is used to detect the nuclei in morphological form and size and classify them as positive or negative according to pixel colorization and intensity. Also, tile extraction, color deconvolution, morphological filters, threshold value method,and cell segmentation algorithms are used to develop INSM1 analysis. The analysis algorithm contains H-DAB colour deconvolution method. To detect negative nuclei H channel, and to detect and classify brown nuclei DAB channel are used. The positive nuclei are classified into three groups according to given threshold values by the user before running the analysis.

Quantitative output variables

Negative Nuclei Number
Dark Brown Positive Nuclei Number
Medium Brown Nuclei Number
Light Brown Nuclei Number
Stromal Nuclei Number
Positivity index (%)

Workflow

View the INSM1 stained whole slide digital image with ViraPath.
Outline tumor either manually or automatically using Virapath Tissue Segmentation algorithm.
Select INSM1 analysis and calibrate the parameters.
Run the analysis.

References

[1]Fujino, K., Yasufuku, K., Kudoh, S., Motooka, Y., Sato, Y., Wakimoto, J., ... & Ito, T. (2017). INSM1 is the best marker for the diagnosis of neuroendocrine tumors: comparison with CGA, SYP and CD56. Int J Clin Exp Pathol, 10(5), 5393-5405.