Convolutional neural networks for automated targeted analysis of raw gas chromatography-mass spectrometry data

  title={Convolutional neural networks for automated targeted analysis of raw gas chromatography-mass spectrometry data},
  author={Angelika Skarysz and Yaser Alkhalifah and Kareen Darnley and Michael Eddleston and Yang Hu and Duncan B. McLaren and William Henry Nailon and Dahlia Salman and Martin D. Sykora and C. L. Paul Thomas and Andrea Soltoggio},
  journal={2018 International Joint Conference on Neural Networks (IJCNN)},
Through their breath, humans exhale hundreds of volatile organic compounds (VOCs) that can reveal pathologies, including many types of cancer at early stages. Gas chromatography-mass spectrometry (GC-MS) is an analytical method used to separate and detect compounds in the mixture contained in breath samples. The identification of VOCs is based on the recognition of their specific ion patterns in GC-MS data, which requires labour-intensive and time-consuming preprocessing and analysis by domain… 

Figures and Tables from this paper

Multiple Compounds Recognition from The Tandem Mass Spectral Data Using Convolutional Neural Network
A model based on convolutional neural network (CNN) which can analyze the chemical peak information in the tandem mass spectrometry (MS/MS) data and extract features of different compounds and classify multi-label mass spectral data is proposed.
Employing and Interpreting a Machine Learning Target-Cognizant Technique for Analysis of Unknown Signals in Multiple Reaction Monitoring
This work evaluates three machine learning algorithms abilities to autonomously associate raw signal peaks based on accuracy in training and testing to associate well-known as well as any potentially unknown peaks present within gas chromatography-mass spectrometry (GC/MS/MS) raw instrument signal.
Qualitative detection of pesticide residues using mass spectral data based on convolutional neural network
The problem transformation method is used to convert pesticide residues prediction into multi-label classification problem and a new convolutional neural network structure Pesticide Residues Neural Network (PRNet) was proposed to solve the problem of multi- label organophosphate pesticide residue prediction.
Cancer Diagnosis by Neural Network Analysis of Data from Semiconductor Sensors
The research results of breath samples analysis from 59 people, including patients with a confirmed diagnosis of respiratory tract cancer, show the accuracy of determining oncological disease was 81.85%, the sensitivity was 90.73%, and the specificity was 61.39%.
Deep Learning for Type 1 Diabetes Mellitus Diagnosis Using Infrared Quantum Cascade Laser Spectroscopy
A diagnostic method based on infrared laser spectroscopy of human exhaled breath for type 1 diabetes mellitus (T1DM) with high accuracy that can be implemented in routine medical practice is described.
Fast and automated biomarker detection in breath samples with machine learning
A machine learning-based system to perform GC-MS data analysis that exploits deep learning pattern recognition ability to learn and automatically detect VOCs directly from raw data, thus bypassing expert-led processing is proposed.


Current breathomics--a review on data pre-processing techniques and machine learning in metabolomics breath analysis.
The current state of the art in data pre-processing and multivariate analysis of breathomics data is described and the community is made aware of the existing data fusion methods, as yet unresearched in breathomics.
Volatile biomarkers in the breath of women with breast cancer.
We sought biomarkers of breast cancer in the breath because the disease is accompanied by increased oxidative stress and induction of cytochrome P450 enzymes, both of which generate volatile organic
Deep learning application trial to lung cancer diagnosis for medical sensor systems
This paper tried to build a diagnosis system of lung cancer based on Deep Learning with 90% accuracy in judging whether the patient had lung cancer or not and it will be useful for pre- and personal diagnosis.
Spectral Deconvolution for Gas Chromatography Mass Spectrometry-Based Metabolomics: Current Status and Future Perspectives
This review focuses on progress that has been made in the development of deconvolution algorithms and provides thoughts on future developments that will expand the application of GC-MS in metabolomics.
MetaboliteDetector: comprehensive analysis tool for targeted and nontargeted GC/MS based metabolome analysis.
The intuitive graphical user interface of MetaboliteDetector additionally allows for a detailed examination of a single GC/MS chromatogram including single ion chromatograms, recorded mass spectra, and identified metabolite spectra in combination with the corresponding reference spectra obtained from a reference library.
Breath gas aldehydes as biomarkers of lung cancer
Sensitivity and specificity of this method were comparable to the diagnostic certitude of conventional serum markers and CT imaging and noninvasive recognition of lung malignancies may be realized if analytical skills, biochemical knowledge and medical expertise are combined into a joint effort.
A profile of volatile organic compounds in breath discriminates COPD patients from controls.
Metabolomics-based methods for early disease diagnostics
The current trends and recent advances in NMR- and MS-based metabolomics are described with a focus on the development of advanced NMR and MS methods, improved multivariate statistical data analysis and recent applications in the area of cancer, diabetes, inborn errors of metabolism and cardiovascular diseases.
Pre-Processing Mass Spectrometry Data
Mass spectrometry can use these proteomic patterns for early diagnosis, to predict prognosis, to monitor disease progression or response to treatment, or even to identify which patients are most likely to benefit from particular treatments.