AI-Powered Biomarker
for Immuno-Oncology

Online Demo
  • 94%

    Demonstrated to perform with
    94% accuracy for pan-cancer
    tissue slide image analysis.

  • 1 Giga pixel

    Size of one tissue slide image,
    digitized for AI analysis.
    (c.f. An HD video is about 1,280 pixels)

  • 15+

    Trained to cover pan-cancer tissue data,
    including lung, breast,
    colorectal cancer and more.

  • Background
  • Our Vision
  • How It Works
Free Online Demo


What if
AI can predict
cancer treatment outcomes?

  • Predicting immunotherapy outcome with AI analysis

    Lunit lab in Samsung Medical Center, Seoul, Korea

  • Finding more responders to
    cancer treatment

  • Tissue Data is
    huge and complex

  • Complex is
    what AI does best

Our Vision

Pioneer and Conquer
"Tissue Phenomics"

  • Accurate
    analysis of tissue slides using
    deep learning

  • Lunit is one of the few companies with high-level tissue data analysis technology. We provide not only detection of features, but link it to clinical outcomes, developing biomarkers specific to drugs.

    • H&E
    • IHC
    • Multiplex IHC

How It Works

Enabling Better Immunotherapy
based on
Tissue Phenomics

  • Accurate
    analysis of tissue
    slides using
    deep learning

    Detects cancer stroma, cancer epithelium, lymphocytes, and etc.

  • Generated by
    Lunit SCOPE IO

    • Detection & Quantification

      Detects intratumoral TILs and stromal TILs from whole slide images.

    • 3 Immune Phenotype Map Generation

      Classifies three immune phenotypes—inflamed, excluded, desert.

    • AI Score & Cutoff

      Calculates AI biomarker score based on the phenotype map. If the score is higher than the cut-off value, it indicates that the patient would be more likely to respond to immunotherapy.

  • The solution covers
    all major types of cancer including

  • How Our AI Works

    • STEP 1

      Tissue Slide (FFPE)
      Pre-treatment & H&E staining

    • STEP 2

      Digitization using a scanner
      Supports various scanner vendors

    • STEP 3

      File upload to server
      (both cloud or on-prem)

    • STEP 4

      Lunit SCOPE IO analysis

  • Lunit SCOPE IO

    • Detection results for visualization
    • Quantification results of key features detected
    • Final score for response prediction
    • Turnaround time: <10 min

    *Lunit SCOPE IO is currently available for
    research use only

  • Presented in

Studies featuring

Our solutions