Quaternion Nondecimated Wavelet Descriptors for Multiclass Breast Histology Classification

2026-07-02

AI Summary

This study addresses the limitations of conventional approaches in analyzing H&E-stained breast histopathology images, which often discard color-vector information by separating RGB channels or converting to grayscale, thereby losing critical multiscale, directional, and structural cues needed to distinguish normal, benign, in situ carcinoma, and invasive carcinoma tissues. The work proposes an interpretable quaternion undecimated wavelet framework that encodes RGB images as pure quaternion fields and applies a two-dimensional quaternion undecimated wavelet transform (QNDWT2D) to extract spatially aligned, color-coupled multiscale directional coefficients. From these, geometric and phase-based features linked to pathological characteristics are constructed. Using radial kernel SVM with repeated nested cross-validation on the BACH dataset, the method achieves high classification performance—confusion primarily occurs between adjacent classes—and feature importance analysis confirms the pivotal roles of quaternion-derived attributes such as directionality, phase concentration, and anisotropy, attaining state-of-the-art results without pretraining or external data.

Abstract

Breast histology images carry diagnostic information in color, texture, orientation, and tissue architecture across a range of scales. In H&E microscopy this information is inherently chromatic and is not fully recovered when the red, green, and blue (RGB) channels are reduced to grayscale or transformed as independent scalar images. We propose an interpretable quaternion nondecimated wavelet framework for breast histology classification. Each RGB image is encoded as a pure quaternion field, and a quaternion nondecimated wavelet transform in two dimensions (QNDWT2D) produces multiscale, directional, color-coupled coefficient fields on the original image grid, keeping color as a single vector quantity rather than three separate channels. From these coefficients we build interpretable feature families summarizing stain balance, wavelet energy, amplitude heterogeneity, quaternion phase concentration, color-axis geometry, directional anisotropy, orientation entropy, and scale-dependent energy decay, each tied to a histopathological property such as nuclear density or glandular organization. We evaluate the descriptors on the BreAst Cancer Histology (BACH) challenge, a balanced four-class set of normal, benign, in situ, and invasive tissue, using a radial-kernel support vector machine (SVM) with repeated nested cross-validation. The descriptors yield balanced recognition across classes, with errors concentrated among adjacent categories while normal and invasive are rarely reversed. Permutation importance shows that directional, phase-concentration, anisotropy, scale, and amplitude-variability groups all contribute, indicating that the classifier draws on genuine quaternion and multiscale geometry rather than global color alone. The framework uses no pretrained networks, learned filters, or external databases, offering a reproducible, interpretable baseline for computational pathology.

Problem

• breast histology classification
• color-coupled representation
• multiscale texture analysis
• H&E microscopy
• quaternion representation

Innovation

• quaternion wavelet
• nondecimated wavelet transform
• color-coupled representation
• interpretable features
• multiscale histopathology analysis