Is it Possible to Extract Metabolic Pathway Information from In Vivo H Nuclear Magnetic Resonance Spectroscopy Data?

TL;DR

This paper introduces a machine learning framework combined with chaos theory analysis to extract metabolic pathway information from in vivo H NMR spectroscopy data, revealing dynamic signals that may improve brain tumor diagnosis.

Contribution

It presents a novel approach integrating machine learning and chaos theory to analyze NMR signals, uncovering dynamic features linked to metabolic pathways.

Findings

NMR signals exhibit rich chaotic dynamics.

Dynamic aspects of NMR signals contain valuable metabolic information.

Proposed methods outperform traditional static analysis.

Abstract

In vivo H nuclear magnetic resonance (NMR) spectroscopy is an important tool for performing non-invasive quantitative assessments of brain tumour glucose metabolism. Brain tumours are considered fast-growth tumours because of their high rate of proliferation. In addition, tumour cells exhibit profound genetic, biochemical and histological differences with respect to the original non-transformed cell types. Therefore, there is strong interest from the clinical investigator's point of view in understanding the role of brain metabolites under normal and pathological conditions and especially in the development of early tumour detection techniques. Unfortunately, current diagnosis techniques ignore the dynamic aspects of these signals. It is largely believed that temporal variations of NMR Spectra are simply due to noise or do not carry enough information to be exploited by any reliable diagnosis procedure. Thus, current diagnosis procedures are mainly based on empirical observations extracted from single averaged spectra. In this paper, firstly a machine learning framework for the analysis of NMR spectroscopy signals which can exploit both static and dynamic aspects of these signals is introduced. Secondly, the dynamics of the signals are further analyzed using elements from chaos theory in order to understand their underlying structure. Furthermore, we show that they exhibit rich chaotic dynamics suggesting the encoding of metabolic pathway information.