Breast IHC AI

Breast IHC AI is an AI-powered image analysis tool developed by Indica Labs for the standardized quantification of key breast cancer biomarkers. Designed for pathologists and researchers working with whole slide images of primary invasive breast cancer, it accurately detects regions of invasive tumor and automates the scoring of HER2 (including low and ultralow expression), ER, PR, and Ki67. By reducing interobserver variability and streamlining biomarker evaluation, Breast IHC AI enables more consistent, efficient, and reproducible results across research workflows.

Breast IHC AI is intended for research use only and is not intended for clinical diagnostic use. It is accessed via the HALO AP® enterprise digital pathology platform from Indica Labs, which is CE-IVDR marked for in-vitro diagnostic use in Europe, the UK, and Switzerland, and provides built-in compliance with FDA 21 CFR Part 11, HIPAA, and GDPR.

Key Outputs and Biomarker Scoring

• HER2: HER2 Grade (Negative, Ultralow, Low, Grade 2, Grade 3) and HER2 Score (0, 1+, 2+, 3+), along with percent positive cells
• ER and PR: Intensity Score including percent of weak, moderate, and strong positive cells; Allred Score
• Ki67: Percent of positive cells

HER2 Low and Ultralow Detection

• The HER2 algorithm detects all levels of HER2 expression, including low and ultralow levels
• Outputs a comprehensive grade and score at the slide level, covering the full spectrum from Negative and Ultralow through to Grade 3 / 3+
• Automatically reports percentage of positive cells and biomarker scores, accompanied by image analysis masks viewable within the HALO AP® platform

Comprehensive Analysis Capabilities

• AI-based analysis assays for automated scoring of HER2, ER, PR, and Ki67
• Built-in artifact exclusion to ensure accurate and consistent results
• Benign epithelial region exclusion to focus analysis on relevant tissue
• Automated tumor detection steps performed prior to biomarker analysis
• Comprehensive results and markups generated for each image, including clinical score and percentage positivity

Workflow and Efficiency Benefits

• Standardizes biomarker scoring and reduces interobserver variability in IHC evaluation without compromising accuracy
• Automated biomarker evaluation reduces workload for pathologists and researchers
• Provides consistent, standardized measurements, freeing pathologists to focus their expertise on interpretation and informed decision-making
• Validated using Hamamatsu NanoZoomer S360 and Leica Aperio GT 450 scanners

Supported Clones

• HER2: 4B5; CB11; BMA
• ER: EP1; SP1
• PR: PgR636; Pgr A/B: 16+SAN27
• Ki67: MIB1; SP6

Supported File Formats

• Non-proprietary: JPG, TIF, OME.TIFF, DICOM (DCM)
• Leica: SVS, AFI, SCN, LIF
• Hamamatsu: NDPI, NDPIS
• Philips: iSyntax, i2Syntax
• 3DHistech: MRXS
• Nikon: ND2
• Akoya: QPTIFF, component TIFF
• Olympus / Evident: VSI
• Zeiss: CZI
• Ventana: BIF
• KFBIO: KFB, KFBF

Breast IHC AI is fully integrated and deployed within the HALO AP® platform, supporting seamless anatomic pathology workflows. Real-world implementations include deployment at MidAtlantic Permanente Medical Group (MAPMG), where it was used to standardize prognostic scores in newly diagnosed breast cancers and increase diagnostic efficiency, and at Kaiser Permanente, where it was deployed as a laboratory developed test. Additional clones beyond those listed may be compatible; contact Indica Labs directly for further information.