Lunit Logo

Product

Explore our Products

AI analysis for Chest x-ray

Insight CXR

AI analysis for Mammography

Insight MMG

AI analysis for Tissue Slides

Insight SCOPE

Background

With the rapid development and spread of digital pathology, the field of pathology is going through a disruptive period of innovation. By AI-based comprehensive analysis of digitized pathology slides, large-scale data that has never been properly utilized will become available.
These data allows AI analysis of diverse cancer types and also enables AI-powered identification and quantification of cancer tissues based on histology images, which lay pivotal groundwork for developing AI biomarkers in cancer therapy.
TILs stand for tumor infiltrating lymphocytes which are identified to be associated with improved outcomes and predictive for response to treatment across various types of tumors including colon, lung, ovarian, and breast. Presence of TILs is now recognized as a sign that the immune system is trying to attack the tumor and has the potential to function as a b iomarker for immune response1-3.

Product description

Developed using Lunit’s cutting-edge deep learning technology4, Lunit SCOPE accurately detects cancer stroma, cancer epithelium and lymphocytes in H&E whole slide images. The AI solution generates (1) intratumoral TIL density information (2) stromal TIL density information, and (3) cancer stroma-epithelium ratio. The solution can also classify tissues in terms of immune phenotype.

Primary value proposition

  • Assist pathologists or researchers to objectively quantify the amount of tumor infiltrating lymphocytes.
  • Enable researchers to discover novel biomarkers through immune phenotyping.

Training & Validation

  • Trained with dataset from extensive annotations by 10+ expert pathologists.
  • Demonstrated to perform with 90% accuracy for breast cancer pathology images.
  • Validated to be highly associated with immune-related RNA expression levels in various cancer types.

Example cases

example case1example case1

example case2example case2

Journals & Conference
Abstracts

  • Deep learning-based predictive biomarker for adjuvant chemotherapy in early-stage hormone receptor-positive breast cancer, AACR 2019
  • Pan-cancer analysis of tumor microenvironment using deep learning-based cancer stroma and immune profiling in H&E images, AACR 2019

References

1 Pages F, Kirilovsky A, Mlecnik B, Asslaber M, Tosolini M, Bindea G,et al. In situ cytotoxic and memory T cells predict outcome in patients with early-stage colorectal cancer. J Clin Oncol. 2009;27(35):5944–51

2 Hwang WT, Adams SF, Tahirovic E, Hagemann IS, Coukos G. Prognostic significance of tumor-infiltrating T cells in ovarian cancer: a meta-analysis. Gynecol Oncol. 2012;124(2):192–8. doi:10.1016/j.ygyno.2011.09.039.3.

3 Dieu-Nosjean MC, Antoine M, Danel C, Heudes D, Wislez M, Poulot V,et al. Long-term survival for patients with non-small-cell lung cancer with intratumoral lymphoid structures. J Clin Oncol. 2008;26(27):4410–7

4 Lunit’s high-end deep learning technology has been demonstrated in various international competitions - won World #1 in MICCAI TUPAC 2016, and CAMELYON 2017; Recognized as one of the world's top 100 AI startups by CB Insights in 2017.