This PDF file contains the front matter associated with SPIE Proceedings Volume 12753, including the Title Page, Copyright information, Table of Contents and Conference Committee lists.

The Raman spectral data feature is generally the Raman wavelength of the sample, and there is a correlation between the feature attributes. Too many features will lead to weak generalization ability of the model, so a Recursive Feature Elimination (RFE) dimensionality reduction method combined with BP neural network is proposed to classify the Raman spectrum of the COVID-19. Firstly, the collected serum Raman spectral data of the population were processed, the maximum and minimum standard scaling method (Min-Max), the Savitzky-Golay smoothing filter method, and then the recursive feature elimination (RFE-RF) based on the random forest base model and two different dimensionality reduction methods of PCA reduce the dimensionality of Raman spectral data and classify them through the BP neural network algorithm model. The experimental results show that the RFE-RF dimensionality reduction method can improve the accuracy of the classification algorithm, providing a new idea for the detection of the COVID-19, with high accuracy, and the classification accuracy of the model is 92.47%

Evaluating tissue mechanical properties is an important issue in the biomedical field. While traditional in vitro tissue deformation experiments have been used to measure mechanical properties, optical methods are becoming increasingly popular due to their non-invasive and non-contact advantages. In this study, we utilized Mueller matrix polarimetry to quantify the mechanical properties of bovine tendon tissue. We acquired 3×3 Mueller matrix images of the tendon tissue samples under various stretching states using a backscattering measurement setup based on a polarization camera, enabling us to examine changes in both structural information and optical properties. Subsequently, we extracted frequency distribution histograms of Mueller matrix elements to elucidate the structural changes in the tendon tissue during the stretching process. We then calculated the Mueller matrix transformation parameters, namely the total anisotropy t1 and anisotropy direction α1 of the tendon tissue samples under different stretching processes, to characterize their structural changes quantitatively. For better discrimination of tendon tissues under different stretching states, we trained an image classification neural network using the derived MMT parameters as input. Ultimately, we obtained a highly accurate model with 90% precision. The results demonstrate the potential of Mueller matrix polarimetry as a tool for evaluating tissue mechanical properties.

Digital holographic microscopy (DHM) is an efficient optical measurement and imaging technology with the advantages of non- invasion, non-damage, high sensitivity and high resolution. However, its resolution is still limited by the diffraction limit of the system. The structured light illumination microscopy (SIM) is a good super resolution imaging technology, which obtains high frequency information of objects by changing the lighting mode to achieve improved imaging resolution. In order to discuss the factors that affect structured light phase imaging, we first simulated the entire imaging process of a structured light digital holography system using Matlab software, and then systematically analyzed the effects of four factors, structured light spatial frequency, loading direction, modulation level, and noise level, on the imaging situation, and reached corresponding conclusions. The above research results can provide a reference for the study of structured light phase imaging methods.

Characterization and evaluation of skin tissue structures are crucial for diagnosis of skin diseases. Recently, the Mueller matrix polarimetry and second harmonic generation microscopy have been widely used in skin tissue imaging due to their unique advantages. However, it is difficult to evaluate comprehensive structural characteristics of complex layered skin tissue by a single imaging modality. Therefore, we propose a dual-modality imaging method combining Mueller matrix polarimetry and second harmonic generation microscopy to achieve the automatic differentiation and quantitative evaluation of different layers of skin tissue. It is demonstrated that the dual-modality method can well divide the mouse tail skin tissue specimens’ images into three layers of stratum corneum, epidermis, and dermis. Then, the texture feature vector extracted by gray level co-occurrence matrix is used to quantitatively characterize the structure features of each layer. Finally, we propose an index named S-Health based on cosine similarity and the gray level co-occurrence matrix texture parameters to quantitatively measure the structural differences between damaged and normal skin areas. The experiments validate the efficacy of the dual-modality imaging method in discriminating and assessing skin tissue structures. It confirms the potential of the proposed method for dermatological practices

The whispering gallery mode ( WGM ) microcavity is a widely used microresonator. With its ultra-high quality factor Q ( above 10⁹ ) and small mode volume ( µm³ ) , it can amplify the interaction between substances. The giant magnetostrictive material Terfenol-D is a rare earth material that can respond to changes in the magnetic field. With its ultra-high magnetostrictive coefficient ( 1500∼2000 ppm ), ultra-fast response speed( less than 1 µs ), and efficient energy conversion efficiency ( about 50% ), it is widely used in magnetic field sensors. Therefore, a high-sensitivity cavity optomechanical magnetometer can be designed and manufactured by combining the WGM and Terfenol-D. By analyzing the inherent characteristic frequency and magnetostrictive material characteristics of the cavity optomechanical magnetometer, the sensitivity of the magnetometer in magnetic field detection is greatly improved by mixing the bias magnetic field with the excitation magnetic field. In our simulation, compared with the environment without bias magnetic field, the deformation of the cavity optomechanical magnetometer is about 10⁶ times higher of the original after adding the bias magnetic field, which will greatly enhance the detection sensitivity of the excitation magnetic field and provide an effective method for detecting biomagnetic signals.

As one of the important means of quantitative evaluation of liver injury, biochemical analyzer is a conventional testing method. However, it has not been widely used in low - middle - income countries (LMICs) due to the limited personnel, lack of funding and complicated environmental problems. How to acquire the information of liver injury quickly with simple and efficient method providing powerful support for clinical decision-making is still a challenge for LMICs. To solve this problem, we designed a novel and easy-to-use point of care system for liver injury. This system consists of a detection device which is based on a two-photon macro photochemical sensor and a paper-based test card with a built-in system for blood cell filtration. The two-photon structure is used to reduce the overall volume and cost of the system, it makes up for the inter-station errors introduced during the instrument assembly process. The blood cell filtration system reduces the operational complexity, and the blood can be filtered, react and change color directly on the test card. The detection device obtains the continuous light reflection signal of the paper-based test card and quantifies the signal to complete the quantitative test of liver damage indexes. We simulated the national environment of LMICs to evaluate the performance of the system: under the environment of 35℃ and 90% relative humidity, 40 Heparin whole blood, the correlation R2 between our system and MindrayBS350s was greater than 0.95. Both of the Randox Quality Control level 2 and level 3 repeatability CV were less than 7.5%. The results show that the system is not only small in size, low in cost and simple in operation, its measurement performance and stability meet the clinical requirements of high temperature and high humidity environment, and so it can be used in LMICs for primary liver function screening and liver disease progression assessment.

Deep learning-based super-resolution models have the potential to revolutionize biomedical imaging and diagnoses by effectively tackling various challenges associated with early detection, personalized medicine, and clinical automation. However, the requirement of an extensive collection of high-resolution images presents limitations for widespread adoption in clinical practice. In our experiment, we proposed an approach to effectively train the deep learning-based super-resolution models using only one real image by leveraging self-generated high resolution images. We employed a mixed metric of image screening to automatically select images with a distribution similar to ground truth, creating an incrementally curated training data set that encourages the model to generate improved images over time. After five training iterations, the proposed deep learning-based super-resolution model experienced a 7.5% and 5.49% improvement in structural similarity and peak-signal-to-noise ratio, respectively. Significantly, the model consistently produces visually enhanced results for training, improving its performance while preserving the characteristics of original biomedical images. These findings indicate a potential way to train a deep neural network in a self-revolution manner independent of real-world human data.

SERF (Spin-exchange relaxation free) atomic magnetometers have been demonstrated as the ultra-high sensitive, noncryogenic sensor. Single-beam configured sensors often choose VCSEL (Vertical-Cavity Surface-Emitting Laser) as the optical source. The existing commercial VCSELs are often with magnetic package and heater, which will affect the sensitivity of atomic magnetometers. In our work, we presented a novel nonmagnetic VCSEL package and its driver circuit. VCSEL chip, heater, and PT1000 are assembled on FPC (Flexible Printed Circuit), and its driver are optimized in terms of low noise, safe protection and AC temperature control. The result showed that the magnetic field generated by heating current can be ignored based on proposed layout method, and the RMS noise of current and temperature are lower than 0.1 µA and 20 mK, respectively. Our proposed method will expand the possibility for commercialization process of miniature SERF atomic magnetometers.

In photoacoustic tomography (PAT), image reconstruction has a fundamental impact on image quality and imaging speed. Among various reconstruction algorithms, the analytical filtered back-projection (FBP) and the numerical time reversal (TR) algorithms are two commonly used image reconstruction techniques in PAT. However, so far, no comprehensive studies are reported on the comparisons of FBP and TR algorithms. In this work, we compare these two algorithms from the perspectives of computational efficiency, robustness to non-ideal detection surfaces, and applicability to heterogeneous media. The results show that: 1) In terms of computational efficiency, FBP is typically faster than TR due to its flexibility in the selection of the reconstruction region. 2) For non-ideal detection surfaces-based reconstruction, FBP can provide more accurate amplitude information for the limited-view reconstruction and can produce fewer image artifacts for the sparse-view reconstruction. 3) For acoustically heterogeneous media, the TR algorithm can incorporate acoustic properties and thus can yield high-quality images, while FBP fails in this case. This study can help researchers gain a deeper understanding of the FBP and TR algorithms and is expected to provide a guide for the reasonable selection of PAT image reconstruction algorithms.

Diffuse optical tomography (DOT) has shown promise in biomedical research, such as breast cancer diagnostics and brain imaging, by reconstructing hidden objects within scattering media. However, the conventional reconstruction framework faces challenges due to the highly ill-posed inverse problem of reconstructing optical properties. This work introduces a novel approach, neural field-based diffuse optical tomography (NeuDOT), which leverages a multi-layer perceptron (MLP) to learn an implicit function that maps spatial coordinates to their corresponding optical absorption coefficients. The performance of the NeuDOT method has been evaluated through several phantom studies, demonstrating its potential for high spatial resolution DOT reconstruction

In recent years, two-photon endomicroscopy has developed as a promising label-free optical biopsy technique for diagnosing gastrointestinal tumors. In this study, we optimize the imaging resolution of the lensed fiber-optic scanning two-photon endomicroscopic imaging scheme. By fabricating a lensed fiber for fiber-optic scanning two-photon endomicroscopy, a lateral resolution of 2.1 μm and a field of view of 600 μm in two-photon endomicroscopic imaging is achieved. Furthermore, the objective-lens-free imaging capability is also validated using gastric

In intravascular laser ablation procedures, rapidly expanding bubbles can inflict mechanical damage on the vessel walls, thereby rendering the transient characteristics of laser-induced bubble formation and evolution within the vessels a significant research topic. The high-density energy of the laser produces tension in the liquid; when this tension amplitude exceeds the stability limit of the medium, the liquid undergoes a phase transition, resulting in bubble formation. In this study, we employed numerical simulation techniques to construct a simplified model of laser-induced cavitation, capturing the velocity, pressure, and phase distribution within the fluid region during bubble evolution. Additionally, we analyzed trends in bubble variation throughout its formation and evolution within the lumen environment. This research could facilitate preliminary prediction of vessel wall damage in laser ablation procedures and provide direction for improvement

As milk consumption increases, rapid detection of milk concentration is becoming more and more crucial. In this research, we propose an LED-based milk concentration detection system, which is a new detection method that detects milk concentration by loading the LED with RF signals. The system includes an arbitrary waveform generator (AWG), LED, lampshade, silicon photo dioxide (PD), oscilloscope, and other components. After passing through different concentrations of milk solution, the light signal carrying a specific frequency is received by the PD, and after photoelectric conversion, the signal carrying a specific frequency in the light signal is converted out, and the concentration of the milk solution to be measured is obtained by reading the peak value measured by the oscilloscope. The experiments validated different brands of milk solutions with varying concentrations, demonstrating the feasibility of LED-based milk concentration detection, and the test analysis was performed for the same brand of milk solutions with varying concentrations, yielding a quadratic fit R2=0.9169, confirming that the system has good sensing performance for milk concentration and validating the method's feasibility.

During the last decade, Miniaturized microscopy, or Miniscope, has gained popularity in neuroscience, particularly for behavioral studies in awake rodents. However, image quality control and standardization remain challenging for both users and developers. To address these challenges, we present MiniMounter, a cost-effective and multi-functional accessory toolkit that includes a hardware platform with customized grippers and four-degree-of-freedom adjustment for Miniscope, as well as software for displacement control and image quality evaluation. Our toolkit enables auto-focusing and accurate measurement of spatial resolution and field of view (FOV). We have demonstrated the effectiveness of such a toolkit through comprehensive phantom and animal experiments.

Macroscopic-level diffuse optical imaging has been widely used in small animal imaging for preclinical research. Due to severe light scattering, 3D reconstruction in diffuse optics is highly ill-posed and sensitive to small noise in measurement. Bringing prior information such as the inner structural or surface information of the imaging object may largely reduce the ill-posed nature of the inverse problem and improve the reconstruction accuracy. Most existing solutions use additional equipment or multimodal techniques (e.g., CT, MRI, etc.). However, these methods pose new challenges such as increased cost and image alignment between different modalities. Herein, we present a novel compact optical tomography system that enables surface extraction using a single programmable scanning module and pinhole modeling. Experiments on phantom and mice show that the system is capable of achieving high-fidelity surface extraction with a minimal error of less than 0.1 mm, which in turn improves the accuracy of 3D fluorescence reconstruction