Hyperspectral Interferometry and AI Unlock Precise Single-Cell Imaging and Dispersion Analysis

Interferometry is a cornerstone of precise optical measurement, yet its sensitivity to external disturbances often limits performance. Kamyar Behrouzi, Tanveer Ahmed Siddique, and Megan Teng, alongside colleagues from the University of California Berkeley and Lawrence Berkeley National Laboratory, have addressed this challenge with a novel approach to broadband interferometry. Their research introduces a technique called general polarization common-path interferometry (GPCPI), enhanced by artificial intelligence, to decouple polarization and improve phase stability. This innovation achieves an order of magnitude improvement in phase stability, paving the way for more accurate measurements in fields like molecular diagnostics and drug discovery. By combining advanced optics with a deep learning model, the team demonstrates the ability to distinguish between normal and cancerous skin cells at a single-cell level, offering a powerful new tool for disease diagnosis.

AI Boosts Robust Common-Path Interferometry Measurements

Interferometry techniques are essential for extracting phase information from optical systems, enabling precise measurements of dispersion and highly sensitive detection of perturbations. While phase sensing offers enhanced sensitivity compared to conventional spectroscopy methods, this sensitivity often makes systems more vulnerable to external factors such as vibrations, introducing instability and noise. In this work, researchers demonstrate a broadband and AI-enhanced interferometry method, denoted general polarization common-path interferometry (GPCPI), that relaxes the polarization constraints commonly found in traditional common-path interferometry. The approach utilises a polarisation diversity scheme combined with machine learning algorithms to mitigate the effects of environmental disturbances and improve measurement stability.

GPCPI achieves high sensitivity while remaining robust against external vibrations and noise by employing a common-path configuration, minimising the impact of air currents and mechanical vibrations on the interference signal. Furthermore, the method incorporates a polarisation diversity scheme, capturing multiple polarisation states of the interfering beams to enhance signal quality and reduce sensitivity to polarisation drift. A key component of the approach is the implementation of machine learning algorithms trained to identify and suppress residual noise, further improving the signal-to-noise ratio. The research contributes a novel interferometric setup that combines the benefits of common-path interferometry with advanced signal processing techniques. The GPCPI method demonstrates a broadband operation range, extending from 650nm to 1650nm, and achieves a phase noise of 2.34 picoradians/√Hz at 1kHz. This work establishes a foundation for developing more robust and sensitive interferometric sensors for a range of applications, including environmental monitoring, biomedical diagnostics, and precision metrology.

Polarization-Independent Common-Path Interferometer Design and Setup

This research paper details a novel approach to quantitative phase and dispersion measurement using a polarization-independent common-path interferometer. The system overcomes limitations of traditional methods, such as sensitivity to polarization, complex setups, or difficulty in achieving high accuracy and stability. Existing interferometric techniques can be susceptible to environmental noise and require precise alignment. The researchers developed a common-path interferometer featuring polarization independence, simplifying the setup and improving robustness. The design also incorporates a broadband light source to enable simultaneous measurement of phase and dispersion.

A ConvNeXt V2 deep learning model is employed for robust fringe analysis, even with noisy or imperfect fringe patterns, allowing for accurate phase extraction and unwrapping. Vector fitting is used to accurately extract the complex refractive index and dispersion characteristics from the measured phase data. The system’s performance was validated by measuring the refractive index and dispersion of known materials (ethylene glycol-water solutions) and comparing the results to established values. Researchers also demonstrated the system’s ability to measure the dispersion of metamaterials. The technique is applicable to a wide range of materials, including transparent media, metamaterials, and biological samples, with potential applications in material characterization, biosensing, metamaterial research, optical component testing, and real-time monitoring of physical properties.

GPCPI Demonstrates Tenfold Phase Stability Improvement A new

Scientists have achieved a ten-fold improvement in phase stability using a new broadband interferometry method, designated general polarization common-path interferometry (GPCPI). This breakthrough relaxes the traditional polarization constraints of common-path interferometry, enabling simultaneous amplitude and phase measurements. The team measured phase patterns subjected to external shock, revealing that GPCPI exhibits significantly reduced variations compared to state-of-the-art interferometry techniques. Experiments demonstrated an 87% contrast reduction in the conventional method immediately after a vertical shock, while GPCPI experienced only a 50% reduction, with a shock damping time of 1.6 seconds.

Further analysis involved recording normalized contrast of the phase pattern over 14 seconds, showing standard deviations of approximately 31% for the conventional method and 14% for GPCPI during shock-free periods, confirming the enhanced stability of the new approach. Vertical phase pattern variations at consecutive time stamps revealed minimal changes in the GPCPI method, maintaining consistent patterns where the conventional method showed significant shifts. The research extends to plasmonic metasurface-based refractive index sensing. Scientists fabricated a plasmonic metasurface consisting of nanorods and encapsulated it within a custom flow cell.

By flowing mixtures of water and ethylene glycol, they induced bulk refractive index variations, and measurements using GPCPI successfully detected a minimum concentration of 20%, corresponding to a refractive index change of 0.02, relying solely on transmittance measurements. The designed plasmonic sensor achieved a sensitivity of approximately 1400nm per refractive index unit. Integration of GPCPI with a ConvNeXt V2 deep learning model, comprising 28 million parameters pre-trained on the ImageNet dataset, enables single-shot, real-time tracking of phase variation with minimized noise. This combination allows for robust cell classification and disease diagnosis at a single-cell level, differentiating between normal and cancerous skin cells through analysis of interference fringes.

GPCPI Enables Stable Single-Cell Dispersion Imaging

Researchers have developed a new interferometry technique, general polarization common-path interferometry (GPCPI), which significantly improves the stability and accuracy of broadband phase measurements. By relaxing the polarization constraints of traditional methods and incorporating a deep learning algorithm, specifically a customized ConvNeXt V2 model, GPCPI achieves an order of magnitude improvement in phase stability while simultaneously measuring both amplitude and phase. This advancement allows for real-time, noise-suppressed phase sensing, even with samples exhibiting arbitrary polarization. The technique’s capabilities were demonstrated through the characterization of plasmonic metasurfaces and, crucially, through hyperspectral single-cell dispersion imaging.

Analysis of interference fringes enabled robust classification of skin cells, distinguishing between normal and cancerous types, and offers a platform for biological studies and medical diagnostics. The authors acknowledge a limitation in that the performance of the deep learning model is dependent on the quality and diversity of the training data. Future work will likely focus on expanding the application of GPCPI to areas such as molecular diagnostics, drug discovery, and quantum sensing, building upon its demonstrated versatility and broad applicability to optical metrology.