Coherence Tomography System Research Articles

Structured illumination microscopy based on Kramers-Kronig relations for quantitative phase reconstruction.

Structured illumination microscopy (SIM) is a widely applied fluorescence super-resolution imaging technique. It can also serve as high-throughput imaging in coherent imaging systems. However, coherent SIM requires additional qualitative/quantitative phase imaging methods to acquire phase information. This paper proposes a structured illumination microscopy technique based on the Kramers-Kronig relations (KK-SIM) that achieves quantitative phase imaging without the need for extra technical assistance and relies solely on the spatial-domain intensity images reconstructed through conventional SIM. KK-SIM utilizes a non-iterative approach to recover intensity into amplitude and phase, maintaining SIM's high acquisition speed and reconstruction efficiency. Our work enables high-throughput quantitative phase imaging using conventional SIM experimental setups and data post-processing, making SIM suitable for label-free, noninvasive dynamic observation.

DGGNets: Deep Gradient-Guidance Networks for Speckle Noise Reduction

Speckle noise is a granular interference that degrades image quality in coherent imaging systems, including underwater sonar, Synthetic Aperture Radar (SAR), and medical ultrasound. This study aims to enhance speckle noise reduction through advanced deep learning techniques. We introduce the Deep Gradient-Guidance Network (DGGNet), which features an architecture comprising one encoder and two decoders—one dedicated to image recovery and the other to gradient preservation. Our approach integrates a gradient map and fractional-order total variation into the loss function to guide training. The gradient map provides structural guidance for edge preservation and directs the denoising branch to focus on sharp regions, thereby preventing over-smoothing. The fractional-order total variation mitigates detail ambiguity and excessive smoothing, ensuring rich textures and detailed information are retained. Extensive experiments yield an average Peak Signal-to-Noise Ratio (PSNR) of 31.52 dB and a Structural Similarity Index (SSIM) of 0.863 across various benchmark datasets, including McMaster, Kodak24, BSD68, Set12, and Urban100. DGGNet outperforms existing methods, such as RIDNet, which achieved a PSNR of 31.42 dB and an SSIM of 0.853, thereby establishing new benchmarks in speckle noise reduction.

Characterization of Defocused Coherent Imaging Systems with Periodic Objects.

Recent advancements in quantum and quantum-inspired imaging techniques have enabled high-resolution 3D imaging through photon correlations. These techniques exhibit reduced degradation of image resolution for out-of-focus samples compared to conventional methods (i.e., intensity-based incoherent imaging). A key advantage of these correlation-based approaches is their independence from the system numerical aperture (NA). Interestingly, both improved resolution of defocused images and NA-independent scaling are linked to the spatial coherence of light. This suggests that while correlation measurements exploit spatial coherence, they are not essential for achieving this imaging advantage. This discovery has led to the development of optical systems that achieve similar performance by using spatially coherent illumination and relying on intensity measurements: direct 3D imaging with NA-independent resolution was recently demonstrated in a correlation-free setup using LED light. Here, we explore the physics behind the enhanced performance of defocused coherent imaging, showing that it arises from the modification of the sample's spatial harmonic content due to diffraction, unlike the blurring seen in conventional imaging. The results we present are crucial for understanding the implications of the physical differences between coherent and incoherent imaging, and are expected to pave the way for the practical application of the discovered phenomena.

Label-free functional imaging of vagus nerve stimulation-evoked potentials at the cortical surface

Vagus nerve stimulation (VNS) is an FDA-approved stimulation therapy to treat patients with refractory epilepsy. In this work, we use a coherent holographic imaging system to characterize vagus nerve-evoked potentials (VEPs) in the cortex in response to VNS stimulation paradigms without electrode placement or any genetic, structural, or functional labels. We analyze stimulation amplitude up to saturation, pulse width up to 800 μs, and frequency from 10 Hz to 30 Hz, finding that stimulation amplitude strongly modulates VEPs response magnitude (effect size 0.401), while pulse width has a moderate modulatory effect (effect size 0.127) and frequency has almost no modulatory effect (effect size 0.009) on the evoked potential magnitude. We find mild interactions between pulse width and frequency. This non-contact label-free functional imaging technique may serve as a non-invasive rapid-feedback tool to characterize VEPs and may increase the efficacy of VNS in patients with refractory epilepsy.

Coherent ghost imaging via phase compensation

The 4-f optical illumination structure is widely used in coherent ghost imaging (CGI) system but its imaging field of view (FOV) is usually restricted by the transmission aperture of the lens. Therefore, this optical illumination structure cannot be applied to remote imaging. Based on conventional imaging illumination structure, a large-FOV CGI via phase compensation (CGI-PC) is proposed and its imaging property is investigated. The experimental results demonstrate that both the FOV and the quality of CGI can be dramatically improved when a lens is introduced into the receiving system to compensate the system’s spatial phase difference. Factors influencing the quality of CGI-PC are also discussed.

Automatic Speckle Noise Reduction in SAR Imagery Using Optimized Kernel Support Vector Machine

Speckle noise (SN) is one of the major types of noise that frequently occurs in different coherent imaging systems such as medical imaging, Synthetic Aperture Radar (SAR) and active Radar images. SAR is a powerful imaging technology that generates fine-resolution images and monitors the earth’s surface in order to identify its physical properties. The satellite images captured by SAR are mainly affected by SN, which reduces the quality of images and complicates the image representation. Therefore, removing SN from SAR images is one of the major challenges and needs significant attention. The proposed study introduces an optimal Machine Learning (ML) classifier named Kernel Support Vector Machine-Improved Aquila Optimization (KSVM-IAO) for reducing SN in SAR images. This study uses a two-step process called filtering and enhanced despeckling to minimize the consequence of speckle suppression. In the first step, different imaging filters, namely Improved Lee Filter (ILF), Improved Frost Filter (IFF), Improved Kuan Filter (IKF) and Improved Boxcar Filter (IBF), are utilized to remove the SN in SAR images. Next, the denoised image is fed to the second stage, which makes use of an optimized KSVM-IAO classifier to obtain an enhanced despeckle image. The hyperparameters of KSVM are tuned using the IAO algorithm, which reduces overfitting issues and increases accuracy. MATLAB is the simulation tool used for the analysis of SAR images. The simulation outcomes reveal that the proposed KSVM-IAO method obtained excellent SN removal while preserving the edges and fine details with low computational complexity.

Blind Image Despeckling Using a Multiscale Attention-Guided Neural Network

Coherent imaging systems have been applied in the detection of target of interest, natural resource exploration, ailment diagnosis, etc. However, it is easy to generate speckle-degraded images due to the coherent interference of reflected echoes, restricting these practical applications. Speckle noise is a granular interference that affects the observed reflectivity. It is often modeled as multiplicative noise with a negative exponential distribution. This non-linear property makes despeckling of imaging data an intractable problem. To enhance despeckling performance, we propose to blindly remove speckle noise using an intelligent computing-enabled multi-scale attention-guided neural network (termed MSANN). In particular, we first introduce the logarithmic transformation to convert the multiplicative speckle noise model to an additive version. Our MSANN, essentially a feature pyramid network (FPN), is then exploited to restore degraded images in the logarithmic domain. To enhance the generalization ability of MSANN, a multi-scale feature enhancement attention module (FEAM) is incorporated into MSANN to extract multi-scale features for improving imaging quality. A multi-scale mixed loss function is further presented to increase network robustness during training. The final despeckled images are naturally equivalent to the exponential versions of the output of MSANN. Experimental results have shown that MSANN has the capacity of effectively removing speckle noise while preserving essential structures. It can achieve superior despeckling results in terms of visual quality and quantitative measures.

An efficient autofocus method for microscope based on the improved first-order derivative Gaussian filtering operator

This paper introduces an improved filtering operator based on the first-order derivative Gaussian filtering operator. As the defocus distance decreases, the edges of an image transition from being smooth to sharp. Using the operator proposed in this study, a parameter can be obtained to quantify the degree of edge sharpness, where a smaller value indicates sharper edges. By using the characteristic, we can judge whether the object is near the focal plane. Combining this information, we proposed two focus method. Both of them can avoid the problem of the hill climbing algorithm locating at local extremum points and one of them can locate the watershed between “coarse search” and “fine search” before the sharpness value crosses the peak value, so as to realize the transition from “coarse search” to “fine search” before crossing the peak value and improve focusing efficiency. This is different from previous passive focusing strategies. The final experiment shows that the proposed solutions in this paper is effective. However, the methods proposed in this study are only applicable to incoherent optical microscopy imaging systems and are not suitable for coherent light source microscopy imaging systems. Besides, the prerequisite for the successful implementation of the two proposed focusing strategies in this study is that the target exhibits sharp edges when it is within the focal plane.

System-level noise performance of coherent imaging systems.

We provide an in-depth analysis of noise considerations in coherent imaging, accounting for speckle and scintillation in addition to "conventional" image noise. Specifically, we formulate closed-form expressions for total effective noise in the presence of speckle only, scintillation only, and speckle combined with scintillation. We find analytically that photon shot noise is uncorrelated with both speckle and weak-to-moderate scintillation, despite their shared dependence on the mean signal. Furthermore, unmitigated speckle and scintillation noise tends to dominate coherent-imaging performance due to a squared mean-signal dependence. Strong coupling occurs between speckle and scintillation when both are present, and we characterize this behavior by fitting a scale factor capable of generating variances in closed form. We verify each of these claims through a series of wave-optics simulations, and we see strong agreement in general between numerical results and theoretical predictions. Our findings allow us to confidently gauge signal-to-noise ratio (SNR) expectations when active illumination produces coherent noise.

On the Arrays Distribution, Scan Sequence and Apodization in Coherent Dual-Array Ultrasound Imaging Systems

Coherent multi-transducer ultrasound (CoMTUS) imaging creates an extended effective aperture through the coherent combination of multiple arrays, which results in images with enhanced resolution, extended field-of-view, and higher sensitivity. However, this also creates a large discontinuous effective aperture that presents additional challenges for current beamforming methods. The discontinuities may increase the level of grating and side lobes and degrade contrast. Also, direct transmissions between multiple arrays, happening at certain transducer relative positions, produce undesirable cross-talk artifacts. Hence, the position of the transducers and the scan sequence play key roles in the beamforming algorithm and imaging performance of CoMTUS. This work investigates the role of the distribution of the individual arrays and the scan sequence in the imaging performance of a coherent dual-array system. First, the imaging performance for different configurations was assessed numerically using the point-spread-function, and then optimized settings were tested on a tissue mimicking phantom. Finally, a subset of the proposed optimum imaging schemes was experimentally validated on two synchronized ULA OP-256 systems equipped with identical linear arrays. Results show that CoMTUS imaging performance can be enhanced by optimizing the relative position of the arrays and the scan sequence together, and that the use of apodization can reduce cross-talk artifacts without degrading spatial resolution. Adding weighted compounding further decreases artifacts and helps to compensate for the differences in the brightness across the image. Setting the maximum steering angle according to the spatial configuration of the arrays reduces the sidelobe energy up to 10 dB plus an extra 4 dB reduction is possible when increasing the number of PWs compounded.

A Phase Recovery Technique Using the Genetic Algorithm for Aberration Correction in a Coherent Imaging System.

For traditional imaging systems, high imaging quality and system miniaturization are often contradictory. In order to meet the requirements of high imaging quality and system miniaturization, this paper proposes a method to correct the aberration of coherent imaging optical systems. The method is based on the idea of phase recovery and the imaging principle of a coherent imaging system to recover the aberrations at the exit pupil of the system. According to the recovered aberrations, conjugate filters are constructed to correct the image quality in the frequency domain. The imaging quality of the system is improved without changing the original optical path, and the simplicity of the system is guaranteed. To solve the pupil frequency domain aberration more accurately, this paper adopts the dual competition and parallel recombination strategy based on the genetic algorithm and introduces the disaster model. The improved genetic algorithm can effectively restrain the appearance of the "precocity" phenomenon. Finally, the paraxial imaging optical path is simulated and verified by experiments. The results show that, after aberration correction, the image sharpness is improved and the edge information is richer, which verifies the feasibility of the coherent imaging system image quality enhancement method proposed in this paper.

A SAR Image-Despeckling Method Based on HOSVD Using Tensor Patches

Coherent imaging systems, such as synthetic aperture radar (SAR), often suffer from granular speckle noise due to inherent defects, which can make interpretation challenging. Although numerous despeckling methods have been proposed in the past three decades, SAR image despeckling remains a challenging task. With the extensive use of non-local self-similarity, despeckling methods under the non-local framework have become increasingly mature. However, effectively utilizing patch similarities remains a key problem in SAR image despeckling. This paper proposes a three-dimensional (3D) SAR image despeckling method based on searching for similar patches and applying the high-order singular value decomposition (HOSVD) theory to better utilize the high-dimensional information of similar patches. Specifically, the proposed method extends two-dimensional (2D) to 3D for SAR image despeckling using tensor patches. A new, non-local similar patch-searching measure criterion is used to classify the patches, and similar patches are stacked into 3D tensors. Lastly, the iterative adaptive weighted tensor cyclic approximation is used for SAR image despeckling based on the HOSVD method. Experimental results demonstrate that the proposed method not only effectively reduces speckle noise but also preserves fine details.

Phase retrieval via nonlocal complex-domain sparsity.

Phase retrieval is indispensable for a number of coherent imaging systems. Owing to limited exposure, it is a challenge for traditional phase retrieval algorithms to reconstruct fine details in the presence of noise. In this Letter, we report an iterative framework for noise-robust phase retrieval with high fidelity. In the framework, we investigate nonlocal structural sparsity in the complex domain by low-rank regularization, which effectively suppresses artifacts caused by measurement noise. The joint optimization of sparsity regularization and data fidelity with forward models enables satisfying detail recovery. To further improve computational efficiency, we develop an adaptive iteration strategy that automatically adjusts matching frequency. The effectiveness of the reported technique has been validated for coherent diffraction imaging and Fourier ptychography, with ≈7 dB higher peak SNR (PSNR) on average, compared with conventional alternating projection reconstruction.

Phase retrieval with a dual recursive scheme.

Since optical sensors cannot detect the phase information of the light wave, recovering the missing phase from the intensity measurements, called phase retrieval (PR), is a natural and important problem in many imaging applications. In this paper, we propose a learning-based recursive dual alternating direction method of multipliers, called RD-ADMM, for phase retrieval with a dual and recursive scheme. This method tackles the PR problem by solving the primal and dual problems separately. We design a dual structure to take advantage of the information embedded in the dual problem that can help with solving the PR problem, and we show that it is feasible to use the same operator for both the primal and dual problems for regularization. To demonstrate the efficiency of this scheme, we propose a learning-based coded holographic coherent diffractive imaging system to generate the reference pattern automatically according to the intensity information of the latent complex-valued wavefront. Experiments on different kinds of images with a high noise level indicate that our method is effective and robust, and can provide higher-quality results than other commonly-used PR methods for this setup.

HoMeNL: A Homogeneity Measure-Based NonLocal Filtering Framework for Detail-Enhanced (Pol)(In)SAR Image Denoising

As an inherent problem in coherent imaging systems, the existence of speckle noise results in SAR images with strong signal-dependent variance and seriously hinders the related properties estimation and the image interpretation. Among many filtering methods, nonlocal means (NLM) have been proven to be effective in reducing noise while preserving details. However, traditional NLM filters still face two core problems: 1) it is difficult for homogeneous pixels selection to construct a patch adaptive to local structure for preventing the omission phenomenon; 2) most central pixel value estimators are still at the stage of suppressing the blurring effect rather than eliminating the wrongly selected heterogeneous pixels. To overcome these two problems in the (Pol)(In)SAR image denoising, this paper proposes a homogeneity measure-based nonlocal (HoMeNL) filtering framework based on the following three innovations: 1) to sufficiently select homogeneous pixels in the patch-wise matching processing, the shape-adaptive (SA) patch can be selected from multiple preset patches with the one-to-many matching strategy; 2) as a general extension of the Lee estimator in (Pol)(In)SAR image denoising, the homogeneity measure (HoMe)-based estimator can achieve an optimal bias-variance tradeoff for the central pixel value; 3) the highlight of the proposed method is that the iterative re-weighted (IRW) estimation combines the residuals statistics and the homogeneity measure to adaptively locate and remove the wrongly selected heterogeneous pixels. Simulated and real experimental results show that the proposed filtering framework owns a superior performance than most state-of-art filters in three aspects of noise reduction, detail enhancement, and coherence magnitude estimation.

Development of the W7-X Alkali Metal Beam Diagnostic Observation System for OP2

On a Wendelstein 7-X (W7-X), an alkali metal beam (AMB) diagnostic system was installed in order to measure the plasma edge electron density profiles and turbulence transport. A sodium beam was injected in the plasma, and the light emission was observed by an optical system. During the last operation phase, OP1.2b campaign trial spectral measurements were performed with a dedicated optical branch. The results showed the emergence of potential CX lines in the light spectra during sodium injection. The lines were identified as Carbon III, which were the dominant lines observed by other diagnostics at the edge plasma. Based on these results, an additional dedicated optical system was developed and installed in 2021 for the upcoming operational phase, OP2. The optics were designed for multiple purposes: spectral measurements for the AMB system and for a He/Ne gas jet. The system was designed to allow implementation of further diagnostics on this port later (e.g., coherence imaging system). The details of the implementation of the design requirements and the main challenges of the manufacturing process and installation are discussed in this paper.

Rapid coherent Raman hyperspectral imaging based on delay-spectral focusing dual-comb method and deep learning algorithm.

Rapid coherent Raman hyperspectral imaging shows great promise for applications in sensing, medical diagnostics, and dynamic metabolism monitoring. However, the spectral acquisition speed of current multiplex coherent anti-Stokes Raman scattering (CARS) microscopy is generally limited by the spectrometer integration time, and as the detection speed increases, the signal-to-noise ratio (SNR) of single spectrum will decrease, leading to a terrible imaging quality. In this Letter, we report a dual-comb coherent Raman hyperspectral microscopy imaging system developed by integrating two approaches, a rapid delay-spectral focusing method and deep learning. The spectral refresh rate is exploited by focusing the relative delay scanning in the effective Raman excitation region, enabling a spectral acquisition speed of 36 kHz, ≈4 frames/s, for a pixel resolution of 95 × 95 pixels and a spectral bandwidth no less than 200 cm-1. To improve the spectral SNR and imaging quality, the deep learning models are designed for spectral preprocessing and automatic unsupervised feature extraction. In addition, by changing the relative delay focusing region of the comb pairs, the detected spectral wavenumber region can be flexibly tuned to the high SNR region of the spectrum.

Fluorescent wavefront shaping using incoherent iterative phase conjugation

Wavefront shaping correction makes it possible to image fluorescent particles deep inside scattering tissue. This requires determining a correction mask to be placed in both the excitation and emission paths. Standard approaches select correction masks by optimizing various image metrics, a process that requires capturing a prohibitively large number of images. To reduce the acquisition cost, iterative phase conjugation techniques use the observation that the desired correction mask is an eigenvector of the tissue transmission operator. They then determine this eigenvector via optical implementations of the power iteration method, which require capturing orders of magnitude fewer images. Existing iterative phase conjugation techniques assume a linear model for the transmission of light through tissue, and thus only apply to fully coherent imaging systems. We extend such techniques to the incoherent case. The fact that light emitted from different sources sums incoherently violates the linear model and makes linear transmission operators inapplicable. We show that, surprisingly, the nonlinearity due to incoherent summation results in an order-of-magnitude acceleration in the convergence of the phase conjugation iteration.

Epidural anesthesia needle guidance by forward-view endoscopic optical coherence tomography and deep learning

Epidural anesthesia requires injection of anesthetic into the epidural space in the spine. Accurate placement of the epidural needle is a major challenge. To address this, we developed a forward-view endoscopic optical coherence tomography (OCT) system for real-time imaging of the tissue in front of the needle tip during the puncture. We tested this OCT system in porcine backbones and developed a set of deep learning models to automatically process the imaging data for needle localization. A series of binary classification models were developed to recognize the five layers of the backbone, including fat, interspinous ligament, ligamentum flavum, epidural space, and spinal cord. The classification models provided an average classification accuracy of 96.65%. During puncture, it is important to maintain a safe distance between the needle tip and the dura mater. Regression models were developed to estimate that distance based on the OCT imaging data. Based on the Inception architecture, our models achieved a mean absolute percentage error of 3.05% ± 0.55%. Overall, our results validated the technical feasibility of using this novel imaging strategy to automatically recognize different tissue structures and measure the distances ahead of the needle tip during the epidural needle placement.

Implementation of free-space Fourier Ptychography with near maximum system numerical aperture.

Over the past decade, the research field of Fourier Ptychographic Microscopy (FPM) has seen numerous innovative developments that significantly expands its utility. Here, we report a high numerical aperture (NA) FPM implementation that incorporates some of these innovations to achieve a synthetic NA of 1.9 - close to the maximum possible synthetic NA of 2 for a free space FPM system. At this high synthetic NA, we experimentally found that it is vital to homogenize the illumination field in order to achieve the best resolution. Our FPM implementation has a full pitch resolution of 266 nm for 465 nm light, and depth of field of 3.6 µm. In comparison, a standard transmission microscope (incoherent) with close to maximum possible NA of 0.95 has a full pitch resolution of 318 nm for 465 nm light, and depth of field of 0.65 µm. While it is generally assumed that a free-space coherent imaging system and a free-space incoherent imaging system operating at their respective maximum NA should give comparable resolution, we experimentally find that an FPM system significantly outperforms its incoherent standard microscopy counterpart in resolution by a factor of 20%. Coupled with FPM's substantially longer effective depth of field (5.5 times longer), our work indicates that, in the near-maximum NA operation regime, the FPM has significant resolution and depth of field advantages over incoherent standard microscopy.