Spectroscopy Instruments Gain Momentum in Research, Yet Face Three Major Hurdles in Clinical Translation

For a long time, China’s laboratory instruments and equipment industry, having started relatively late, has lagged behind international giants in R&D capabilities, product quality, and scale, with a particularly high reliance on imports in the field of high-end analytical instruments. In recent years, however, supported by favorable national policies, domestic manufacturers of laboratory instruments and equipment are seizing unprecedented development opportunities, driving rapid growth in the sector.

In clinical practice, spectroscopic instruments have demonstrated extensive application potential and value. They not only enable rapid and accurate detection of specific components in biological samples but also monitor disease progression and treatment efficacy, thereby providing decision support to clinicians and facilitating the formulation of more precise treatment plans.

On May 10, 2024, at the Laboratory Instruments and Equipment Sub-forum of the VBEF Future Healthcare Ecosystem Expo hosted by VCBeat,Song Yizhi, a researcher at the Suzhou Institute of Biomedical Engineering and Technology, Chinese Academy of Sciences, and Chief Scientist of the National Key R&D Program, delivered a keynote presentation titled “Clinical Application Progress and Challenges of Spectroscopic Instruments.”Below is a transcript of the speech.

 Song Yizhi, Researcher at the Suzhou Institute of Biomedical Engineering and Technology, Chinese Academy of Sciences, and Chief Scientist of the National Key R&D Program

A spectrum is a pattern formed by monochromatic light, dispersed from polychromatic light through a dispersive system (such as a prism or grating), arranged sequentially by wavelength or frequency; its full name is optical spectrum. The spectral wavelength range of interest typically spans from the ultraviolet to the visible and then to the infrared regions.

Based on the differences in material components interacting with radiation, spectra are classified into two major categories: molecular spectra and atomic spectra. Molecular spectra include absorption spectra, emission spectra, fluorescence spectra, and scattering spectra.

In recent years, research on spectroscopic techniques in the clinical field has been flourishing, with the number of related research articles growing exponentially. Among various techniques, Raman spectroscopy and near-infrared spectroscopy have garnered particular attention.

In recent years, the application of infrared spectroscopy in clinical diagnosis has advanced rapidly. The integrated visible and NIR-I/II multispectral imaging instrument developed by Tian Jie’s team at the Chinese Academy of Sciences achieved in vivo NIR-II imaging of liver cancer for the first time. NIR-II offers advantages in multiple aspects, including imaging clarity, ease of operation, and tumor detection rate.

As a type of scattering spectroscopy, Raman spectroscopy can reveal numerous chemical fingerprint information in cells, such as sugars, lipids, proteins, and nucleic acids, and reflect metabolic states and functions when combined with isotope labeling. Meanwhile, Raman spectroscopy is suitable for non-destructive or low-damage detection, enabling in situ and real-time analysis, which holds significant value in clinical diagnostics.

The micro-confocal Raman spectrometer developed by our team, which integrates a CT-structure monochromator with a confocal system, enables the acquisition of fluorescence spectra with high spectral resolution and detection with layer-scanning spatial resolution. This technology has demonstrated significant potential in tumor tissue detection and non-invasive blood glucose monitoring. Our team is also closely focusing on its application in the detection of pathogenic microorganisms.

Antimicrobial resistance is a major challenge currently facing global public health. According to WHO projections, deaths associated with antibiotic resistance have become the third leading cause of death worldwide, surpassed only by ischemic heart disease and stroke. Antibiotic resistance poses the greatest threat to the prevention and treatment of increasingly severe infections, and “superbugs” are no longer merely a science fiction scenario.

Currently, the gold-standard clinical testing workflow for pathogenic microorganisms still relies heavily on their culture and proliferation, a process that takes 3–7 days; however, critically ill patients may not be able to wait even a single day.

By leveraging culture-free Raman spectroscopy for pathogen identification and antimicrobial susceptibility testing (AST), we have established a Raman spectral database addressing both identification and AST. By applying machine learning or deep learning recognition algorithms, we achieve rapid pathogen identification and assessment of antibiotic efficacy, with the aim of shortening clinical detection time.

Despite significant advances in the research of spectroscopy technology in the clinical field, products truly applied to clinical practice remain limited. The application of scientific instruments in clinical settings still faces numerous challenges.

First is the repeatability and reliability of the instruments. Scientific research instruments typically have less stringent requirements for repeatability and reliability compared to clinical applications. However, in clinical settings, these metrics are critical. Many companies may not yet be fully prepared to meet clinical demands, thus necessitating further improvements in instrument reliability.

Second is the level of automation. The automation level of scientific research instruments often fails to meet the standards required for clinical use. For example, in the past, I had to operate three different software programs to separately control the light source, detector, and timer when using a certain instrument. While this might be acceptable for researchers, it is not feasible for clinical personnel. Our spectroscopic instrument can automatically identify cell positions, acquire spectra, and perform analysis, which is crucial for clinical applications.

Third is data standardization. Spectroscopic instruments exhibit extremely high sensitivity, with resolutions potentially reaching fractions of a nanometer or even higher; thus, any minor fluctuation may lead to erroneous diagnostic conclusions. Currently, we are dedicated to standardizing data from devices across different laboratories and manufacturers to ensure reliable results in clinical practice.

To overcome these challenges, we need to further enhance instrument reliability, achieve a high degree of automation, and address data standardization issues.

Looking ahead, we anticipate that spectroscopic instruments will achieve greater breakthroughs in clinical practice, providing more accurate tools for medical diagnosis and benefiting a larger number of patients.