Multi-contrast x-ray identification of inhomogeneous materials and their discrimination through deep learning approaches

TL;DR

This study demonstrates that combining multiple contrast channels in advanced x-ray imaging, along with machine learning, significantly improves material discrimination in complex datasets, with broad potential applications.

Contribution

The paper introduces an optimized multi-contrast x-ray approach combined with machine learning for enhanced material identification, validated on large, complex datasets.

Findings

Enhanced material discrimination using combined contrast channels.

Effective machine learning-based classification of threat materials.

Validation on large, complex imaging datasets.

Abstract

Recent innovations in x-ray technology (namely phase-based and energy-resolved imaging) offer unprecedented opportunities for material discrimination, however they are often used in isolation or in limited combinations. Here we show that the optimized combination of contrast channels (attenuation at three x-ray energies, ultra-small angle scattering at two, standard deviation of refraction) significantly enhances material identification abilities compared to dual-energy x-ray imaging alone, and that a combination of off-the-shelf machine learning approaches can effectively discriminate e.g., threat materials in complex datasets. The methodology is validated on a range of materials and image dataset that are both an order of magnitude larger than those used in previous studies. Our results can provide an effective methodology to discriminate, and in some cases identify, different materials in complex imaging scenarios, with prospective applications across the life and physical sciences. While the detection of threat materials is used as a demonstrator here, the methodology could be equally applied to e.g., the distinction between diseased and healthy tissues or degraded vs. pristine materials.