Development of a Novel Label-Free Functional, Molecular, and Structural Imaging System Combining Optical Coherence Tomography and Raman Spectroscopy for In Vivo Measurement of Rat Retina

Optics Medical Imaging Life Sciences Physics Bioengineering Ophthalmology Medicine
optical coherence tomography Raman spectroscopy retina in vivo measurement multimodal imaging label-free technology neurodegenerative diseases

A novel label-free functional molecular and structural imaging system combining Optical Coherence Tomography and Raman Spectroscopy

A Revolution in Optical Imaging: Development of a Multimodal Retinal Imaging System Combining OCT and Raman Spectroscopy

Research Background and Significance

Accessing molecular information of retinal tissue is pivotal for the early diagnosis of ophthalmic and neurodegenerative diseases. However, the current gold standard in retinal imaging—Optical Coherence Tomography (OCT) and its functional extension OCT Angiography (OCTA), only provides structural and blood perfusion information. Despite their significant value in diagnosing diabetic retinopathy and retinal and vascular changes associated with central nervous system diseases (e.g., Alzheimer’s disease and multiple sclerosis), their specificity in discerning disease origins is insufficient due to strong overlap among structural and vascular biomarkers.

To address this limitation, Raman Spectroscopy (RS), a light-based molecular sensing technology, was proposed. RS detects inelastic scattering of light to capture the vibration modes of chemicals bonds, thereby providing unique molecular fingerprints. While RS has successfully differentiated tumor tissues from healthy ones, its application in retinal imaging has been limited due to the challenges posed by optical power restrictions and high levels of retinal autofluorescence.

Previous studies have attempted to combine OCT and RS in dual-modality imaging for tissue analysis and pathological research but have yet to fully demonstrate its feasibility for in vivo animal or human retinal imaging. Thus, the primary objective of this study is to develop a multimodal ophthalmic imaging system that combines OCT and RS in compliance with laser safety standards. The study provides early evidence of the system’s capability for structural, functional, and molecular measurements of the retina in live rats.

Article Source and Authors

This study was published in the February 1, 2025 issue of Biomedical Optics Express (Vol. 16, No. 2) by an international team led by Ryan Sentosa. Collaborating institutions include the Center for Medical Physics and Biomedical Engineering at the Medical University of Vienna (Austria), Carl Zeiss AG (Germany), TNO Optics Department (Netherlands), Horiba France SAS (France), among others. The study was supported by technical contributions from Carl Zeiss Meditec Inc. and Innolume GmbH. Funding came from the Optica Open Access Publishing Agreement.

Research Methods and Workflow

1. System Design and Development

The study introduced a multimodal imaging system incorporating Infrared Fundus Imaging (IR Fundus Imaging), High-Speed Swept-Source OCT (SS-OCT), and Non-Resonant Raman Spectroscopy (NR-RS). The system was calibrated to ensure compliance with laser safety standards, supporting in vivo imaging of small rodents.

System Architecture:

  • IR Fundus Imaging System: Used for initial sample alignment. Based on 730 nm broadband LEDs, with optical power ~1 mW and a refresh rate of 5 Hz.
  • Swept-Source OCT: Utilized a swept-source laser with a central wavelength of 1060 nm to provide 3D structural data and blood perfusion information. The system supports a wide field of view (FoV) of 59°.
  • Raman Spectroscopy Module: Employed a 785 nm continuous wave (CW) diode laser as the excitation source, with optical power of 1 mW and an optimized optical collection design.

Experimental Models:

  • Multimodal Eye Model Validation: A custom-designed artificial eye model, replicating a three-layered retina with vascular structures.
  • Experimental Rats: One Sprague Dawley albino rat (8 weeks old) was tested under anesthesia. Albino rats, with low retinal pigment autofluorescence, were selected to avoid signal interference during RS measurements.

2. Data Acquisition and Processing

Experimental Procedure:

  1. Align and focus the sample with IR Fundus Imaging.
  2. Use OCT to acquire volumetric and perfusion data.
  3. Select specific locations for Raman spectroscopy based on OCT guidance.
  4. Raman spectra were acquired over 30 seconds, with 15 accumulations averaged for in vivo experiments.

Data Processing:

  • Image Processing: Fundus and OCT data were processed using Fourier filtering, dispersion compensation, and phase variance-based blood flow extraction algorithms.
  • Spectral Analysis: Raman data underwent Savitzky–Golay filtering for smoothing, baseline correction, and normalization, focusing on wavenumber ranges between 400–2400 cm⁻¹.

Research Results

1. Multimodal Eye Model Validation

Using the artificial eye model, the following results were observed: - IR Fundus Imaging: Clearly identified the absorption features of three quasi-vessels embedded in the retinal phantom. - OCT Images: Successfully captured the distinct three-layer retinal structure and vascular features, although regions with minimal refractive index differences showed low contrast. - Raman Spectra: Differentiated between background and vessel regions, with characteristic wavenumbers such as 679, 747, 953, and 1450 cm⁻¹ that correspond to blue pigment molecules (Phthalo Blue) incorporated in the phantom.

2. In Vivo Albino Rat Imaging

  • OCT Results: Delivered detailed 3D structural and perfusion maps of the retina.
  • Raman Spectra: Provided precise molecular vibrational patterns related to biomolecules, highlighting:
    • Cholesterol features (702 cm⁻¹)
    • C-S stretching mode of cystine (661 cm⁻¹)
    • Ring breathing of lysine and tyrosine (852 cm⁻¹)

Significance and Innovation

1. Scientific Impact

This study achieved the first in vivo molecular imaging of rodent retinas using non-resonant Raman spectroscopy. It offers groundbreaking insights into retinal molecular composition and advances the study of molecular mechanisms underlying ophthalmic and neurodegenerative diseases.

2. Clinical and Applied Value

The research presents a non-invasive, label-free diagnostic tool with significant potential for detecting eye diseases (e.g., diabetic retinopathy) and central nervous system disorders. Its multimodal nature paves the way for early diagnosis and disease monitoring applications.

3. Highlights and Innovations

  • The system seamlessly integrates OCT and RS technologies, combining structural and molecular imaging capabilities.
  • Maintains laser safety compliance, making it suitable for longitudinal studies.
  • Adopts albino rat models to minimize retinal pigment interference, setting a new standard for small animal studies.

Outlook

Future directions include optimizing Raman signal intensity, extending the molecular detection range to higher wavenumbers, and exploring its application in the human eye. Additionally, combining the system with adaptive optics to enhance Raman spatial resolution represents a critical avenue for research. Overall, this study opens exciting possibilities for multimodal optical imaging in biomedical applications.