3D Solid Spherical Bispectrum CNNs for Biomedical Texture Analysis

TL;DR

This paper introduces a 3D spherical bispectrum CNN layer that enhances rotation invariance and texture analysis in biomedical images, outperforming standard CNNs especially with limited training data.

Contribution

The work proposes a novel 3D spherical bispectrum-based LRI layer for CNNs, improving rotation invariance and texture characterization in biomedical 3D image analysis.

Findings

Bispectrum CNNs outperform spectral and standard CNNs in texture classification.

Bispectrum CNNs require fewer training samples and parameters.

Enhanced detection of rotated patterns in synthetic and real CT datasets.

Abstract

Locally Rotation Invariant (LRI) operators have shown great potential in biomedical texture analysis where patterns appear at random positions and orientations. LRI operators can be obtained by computing the responses to the discrete rotation of local descriptors, such as Local Binary Patterns (LBP) or the Scale Invariant Feature Transform (SIFT). Other strategies achieve this invariance using Laplacian of Gaussian or steerable wavelets for instance, preventing the introduction of sampling errors during the discretization of the rotations. In this work, we obtain LRI operators via the local projection of the image on the spherical harmonics basis, followed by the computation of the bispectrum, which shares and extends the invariance properties of the spectrum. We investigate the benefits of using the bispectrum over the spectrum in the design of a LRI layer embedded in a shallow Convolutional Neural Network (CNN) for 3D image analysis. The performance of each design is evaluated on two datasets and compared against a standard 3D CNN. The first dataset is made of 3D volumes composed of synthetically generated rotated patterns, while the second contains malignant and benign pulmonary nodules in Computed Tomography (CT) images. The results indicate that bispectrum CNNs allows for a significantly better characterization of 3D textures than both the spectral and standard CNN. In addition, it can efficiently learn with fewer training examples and trainable parameters when compared to a standard convolutional layer.