DAB-quant

DAB-quant quantifies 3,3'-diaminobenzidine (DAB) immunohistochemical (IHC) staining in scanned tissue slides to provide objective, reproducible measurements of stained tissue fraction.

Key Features:

• Objective Quantitation: Objectively measures IHC signal to minimize subjective bias from manual assessment.
• Automated Thresholding: Applies Otsu's method to negative control slides to establish thresholds that distinguish tissue from background.
• Normalized Red Minus Blue (NRMB) Intensity Scoring: Scores each tissue pixel by NRMB color intensity to differentiate stained and unstained areas.
• User-Defined Tolerance for Error: Allows specification of an error tolerance to refine the NRMB threshold separating stained versus unstained tissue.
• Fraction Calculation: Calculates the fraction of stained tissue pixels on test slides using the established NRMB threshold.
• Data Presentation and Documentation: Records results in spreadsheet format and generates pseudocolor images to document pixel categorization.
• Flexible Analysis Options: Supports full-slide analysis or sampling using small boxes randomly scattered across the tissue area to assess heterogeneity or exclude problematic regions.

Scientific Applications:

• Histopathology IHC quantification: Quantification of DAB IHC staining in tissue sections for comparative and quantitative histopathological analysis.
• High-throughput quantitative analysis: Processing of large slide datasets to enable high-throughput, quantitative comparisons across samples.

Methodology:

Scanned slides are organized into designated folders for negative controls and test slides; Otsu's method is applied to control slides to define a tissue/background threshold; each tissue pixel is scored by NRMB intensity; a user-defined error tolerance refines the NRMB cutoff to classify stained versus unstained pixels; the fraction of stained pixels on test slides is calculated, results are recorded in a spreadsheet, pseudocolor images are generated, and analysis can be performed on full slides or by random small-box sampling.