A Brain-Inspired Deep Separation Network for Single Channel Raman Spectra Unmixing

TL;DR

This paper introduces RSSNet, a deep neural network inspired by speech separation, capable of unmixing single-channel noisy Raman spectra into individual components, outperforming existing methods and generalizing well to real-world data.

Contribution

The paper presents a novel deep separation neural network for single-channel Raman spectra unmixing, addressing noise tolerance and underdetermined systems, with demonstrated superiority over existing methods.

Findings

RSSNet outperforms competing methods by over 4dB on synthetic datasets.

RSSNet trained on synthetic data successfully unmixes real-world spectra.

The approach enables fast detection of Raman mixtures in practical applications.

Abstract

Raman spectra obtained in real world applications are often a noisy combination of several spectra of various substances in a tested sample. Unmixing such spectra into individual components corresponding to each of the substances is of great value and has been a longstanding challenge in Raman spectroscopy. Existing unmixing methods are predominantly designed to invert an overdetermined mixed model and therefore require multiple mixed spectra as input. However, open domain and/or non-cooperative detection applications in Raman spectroscopy such as controlled substance detection, call for single-channel solutions which can identify individual components from thousands of candidates by analyzing only a single noisy mixed spectrum. To our knowledge, sparse regression is the only existing solution which can cope with this scenario, yet it has very low tolerance to noises and can hardly be applicable in practice. To address these limitations, we introduce a novel neural approach for single-channel Raman spectrum unmixing inspired by speech separation. It aims at solving underdetermined systems and can decompose a noisy mixed spectrum from a library of thousands of components (substances). The core of our method is a deep separation neural network (RSSNet) which takes a mixed spectrum as input and outputs spectra of pure components. We created two synthetic datasets of single-channel Raman spectra unmixing and demonstrated feasibility and superiority of RSSNet on these datasets (outperform competing methods by >4dB). Furthermore, we verified that RSSNet, trained solely on synthetic data, can successfully unmix real-world mixed spectra of mixtures of mineral powders, exhibiting strong generalization. Our approach represents a new paradigm for Raman unmixing and enables new possibilities for fast detection of Raman mixtures.