[Paper Review] Breath analysis by ultra-sensitive broadband laser spectroscopy detects SARS-CoV-2 infection

Qizhong Liang, Ya-Chu Chan+7 more|arXiv (Cornell University)|Feb 4, 2022

Advanced Chemical Sensor Technologies4 citations

TL;DR

This study demonstrates that cavity-enhanced direct frequency comb spectroscopy (CE-DFCS) enables non-invasive, rapid, and highly sensitive detection of SARS-CoV-2 infection in human breath, achieving an area under the ROC curve of 0.849(4) using 170 breath samples, with excellent agreement to RT-PCR. The technique detects molecular absorption features at parts-per-trillion sensitivity without chemical processing, offering a promising alternative for real-time, chemistry-free breath-based diagnostics.

ABSTRACT

Rapid testing is essential to fighting pandemics such as COVID-19, the disease caused by the SARS-CoV-2 virus. Exhaled human breath contains multiple volatile molecules providing powerful potential for non-invasive diagnosis of diverse medical conditions. We investigated breath detection of SARS-CoV-2 infection using cavity-enhanced direct frequency comb spectroscopy (CE-DFCS), a state-of-the-art laser spectroscopic technique capable of a real-time massive collection of broadband molecular absorption features at ro-vibrational quantum state resolution and at parts-per-trillion volume detection sensitivity. Using a total of 170 individual breath samples (83 positive and 87 negative with SARS-CoV-2 based on Reverse Transcription Polymerase Chain Reaction tests), we report excellent discrimination capability for SARS-CoV-2 infection with an area under the Receiver-Operating-Characteristics curve of 0.849(4). Our results support the development of CE-DFCS as an alternative, rapid, non-invasive test for COVID-19 and highlight its remarkable potential for optical diagnoses of diverse biological conditions and disease states.

Motivation & Objective

• To evaluate the feasibility of ultra-sensitive broadband laser spectroscopy for non-invasive detection of SARS-CoV-2 infection in exhaled breath.
• To compare the performance of cavity-enhanced direct frequency comb spectroscopy (CE-DFCS) against RT-PCR, the gold standard for SARS-CoV-2 diagnosis.
• To assess CE-DFCS’s ability to detect subtle molecular changes in breath associated with viral infection, independent of sample preparation or thermal processing.
• To explore the potential of CE-DFCS as a scalable, non-invasive platform for diagnosing diverse respiratory, metabolic, and gastrointestinal conditions through breath analysis.
• To establish a foundation for future development of portable, real-time diagnostic tools based on high-resolution molecular spectroscopy.

Proposed method

• CE-DFCS was employed to measure broadband molecular absorption spectra in exhaled breath across the 2810–2945 cm⁻¹ range, corresponding to ro-vibrational transitions of volatile organic compounds.
• A high-finesse optical cavity enhanced the light-matter interaction, enabling parts-per-trillion sensitivity for trace gas detection.
• Frequency comb lasers provided simultaneous, coherent measurement across thousands of optical frequencies, enabling rapid acquisition of molecular fingerprints.
• Data were collected from 170 individuals (83 RT-PCR positive, 87 negative) with breath samples collected non-invasively via a breath collection device.
• Multivariate statistical analysis, including partial least squares-discriminant analysis (PLS-DA), was used to classify infection status based on spectral patterns.
• The method required no sample pre-treatment, derivatization, or thermal processing, preserving native breath composition.

Experimental results

Research questions

• RQ1 Can CE-DFCS detect SARS-CoV-2 infection in exhaled breath with sensitivity and specificity comparable to RT-PCR?
• RQ2 How does CE-DFCS perform in distinguishing infected from uninfected individuals in a vaccinated, young cohort with low viral load?
• RQ3 To what extent can molecular absorption patterns in breath serve as a reliable biomarker for viral infection without chemical manipulation?
• RQ4 Can CE-DFCS detect isomeric, isobaric, or isotopologue-specific signals that distinguish host response from viral components?
• RQ5 What is the potential of CE-DFCS for broader application in diagnosing other respiratory, metabolic, or gastrointestinal conditions via breath analysis?

Key findings

• CE-DFCS achieved an area under the ROC curve (AUC) of 0.849(4) in distinguishing SARS-CoV-2 positive from negative individuals, indicating strong diagnostic performance.
• The technique showed excellent agreement with RT-PCR results, supporting its potential as a non-invasive alternative to swab-based testing.
• CE-DFCS successfully discriminated between individuals based on smoking history, biological sex, and gastrointestinal symptoms, indicating sensitivity to physiological and lifestyle factors.
• The method detected molecular signatures associated with SARS-CoV-2 infection without requiring sample heating, purification, or chemical derivatization.
• The spectral range of CE-DFCS covers ro-vibrational bands of key biomolecules, including C-H stretches in viral proteins such as the spike protein, suggesting potential for direct viral load quantification.
• Future integration with deep learning models and chip-scale miniaturization could enable portable, low-cost, and widely deployable breath diagnostic devices.