Bifurcation detection in intravascular optical coherence tomography using vision transformer based deep learning.

Objective. Bifurcation detection in intravascular optical coherence tomography (IVOCT) images plays a significant role in guiding optimal revascularization strategies for percutaneous coronary intervention (PCI). We propose a bifurcation detection method using vision transformer (ViT) based deep learning in IVOCT.Approach. Instead of relying on lumen segmentation, the proposed method identifies the bifurcation image using a ViT-based classification model and then estimate bifurcation ostium points by a ViT-based landmark detection model.Main results. By processing 8640 clinical images, the Accuracy and F1-score of bifurcation identification by the proposed ViT-based model are 2.54% and 16.08% higher than that of traditional non-deep learning methods, are similar to the best performance of convolutional neural networks (CNNs) based methods, respectively. The ostium distance error of the ViT-based model is 0.305 mm, which is reduced 68.5% compared with the traditional non-deep learning method and reduced 24.81% compared with the best performance of CNNs based methods. The results also show that the proposed ViT-based method achieves the highest success detection rate are 11.3% and 29.2% higher than the non-deep learning method, and 4.6% and 2.5% higher than the best performance of CNNs based methods when the distance section is 0.1 and 0.2 mm, respectively.Significance. The proposed ViT-based method enhances the performance of bifurcation detection of IVOCT images, which maintains a high correlation and consistency between the automatic detection results and the expert manual results. It is of great significance in guiding the selection of PCI treatment strategies.

Step-adaptive accelerated demodulation algorithm for LFM-pulse-based distributed acoustic sensing.

We propose a novel, to our knowledge, step-adaptive cross-correlation algorithm tailored for distributed acoustic sensing systems based on linear frequency modulation pulses, aiming for rapid demodulation. This algorithm adjusts its step length through an adaptive "successive refinement" search strategy, which greatly improves computational efficiency by reducing the number of cross-correlation computations. Experimental results have shown that the demodulation time can be reduced by approximately 15 times compared to the conventional method, while maintaining the same demodulation result.

Highly sensitive fiber-optic chemical pH sensor based on surface modification of optical fiber with ZnCdSe/ZnS quantum dots.

The pH value plays a vital role in many biological and chemical reactions. In this work, the fiber-optic chemical pH sensors were fabricated based on carboxyl ZnCdSe/ZnS quantum dots (QDs) and tapered optical fiber. The photoluminescence (PL) intensity of QDs is pH-dependence because protonation and deprotonation can affect the process of electron-hole recombination. The evanescent wave of tapered optical fiber was used as excitation source in the process of PL. To obtain higher sensitivity, the end faces of fiber were optimized for cone region. By lengthening the cone region and shrinking the end diameter of optical fiber, evanescent wave was enhanced and the excitation times of QDs were increased, which improved the PL intensity and the sensitivity of the sensor. The sensitivity of sensor can reach as high as 0.139/pH in the range of pH 6.00-9.01. The surface functional modification was adopted to prepare sensing films. The carboxyl groups on the QDs ligands are chemically bonded to the fiber surface, which is good for response time (40 s) and stability (decreased 0.9 % for 5 min). These results demonstrated that ZnCdSe/ZnS QDs-based fiber-optic chemical pH sensors are promising approach in rapid and precise pH detection.

Giant optical absorption of a PtSe2-on-silicon waveguide in mid-infrared wavelengths.

Low-dimensional platinum diselenide (PtSe2) is a promising candidate for high-performance optoelectronics in the short-wavelength mid-infrared band due to its high carrier mobility, excellent stability, and tunable bandgap. However, light usually interacts moderately with low-dimensional PtSe2, limiting the optoelectronic responses of PtSe2-based devices. Here we demonstrated a giant optical absorption of a PtSe2-on-silicon waveguide by integrating a ten-layer PtSe2 film on an ultra-thin silicon waveguide. The weak mode confinement in the ultra-thin waveguide dramatically increases the waveguide mode overlap with the PtSe2 film. Our experimental results show that the absorption coefficient of the PtSe2-on-silicon waveguide is in the range of 0.0648 dB µm-1 to 0.0704 dB µm-1 in a spectral region of 2200 nm to 2300 nm wavelengths. Furthermore, we also studied the optical absorption in an ultra-thin silicon microring resonator. Our study provides a promising approach to developing PtSe2-on-silicon hybrid optoelectronic integrated circuits.

Hollow-microsphere-integrated optofluidic immunochip for myocardial infarction biomarker microanalysis.

This work developed an optofluidic immunochip that uses whispering gallery mode with fiber laser enhancement, for the rapid detection of a key biomarker cardiac troponin I for acute myocardial infarction (AMI). The immunochip adopted an innovative design, using perforated hollow glass microspheres (HGMS) as carriers, with antibodies immobilized on the inner surface of the HGMS, thereby achieving ultra-low sample consumption. The performance of the immunochip was improved by fiber laser, including spectral width compression to 0.019 nm, optical signal-to-noise ratio amplification to 63.17 dB, and an enhancement in the limit of detection to 5 pg/mL. Moreover, this immunochip can provide results within 15 min, making it highly suitable for early AMI risk management. Compared to the standard electrochemiluminescence detection method, although some differences exist in the results of the immunochip due to the principle of detection and differences in antibody affinity, its positive reference value can be calculated as 0.0754 ng/mL, with a successful recognition rate of 88% for positive patients. The immunosensor is integrated on a polydimethylsiloxane substrate, with a compact size suitable for use in point-of-care devices and AMI self-screening, as well as rapid disease screening and microanalysis of various biomarkers, offering new possibilities for applications in these fields.

Simultaneous bond-selective deuterium-based isotopic labeling sensing with disposable ultra-miniature CARS fiber probe.

Deuterium-based isotopic labeling is an important technique for tracking cellular metabolism with the Raman signals analysis of low-wavenumber (LW) C-D bonds and high-wavenumber (HW) C-H bonds. We propose and demonstrate a disposable ultra-miniature fiber probe to detect LW and HW coherent anti-Stokes Raman scattering (CARS) spectra for deuterated compounds simultaneously and bond-selectively sensing. The 10.78 µm diameter disposable fiber probe, comprised of focusing taper as fiber probe head and time-domain walk-off eliminating fiber section with designed length, realizes wide-frequency-interval dual Stokes pulse delivering and focusing. The fiber probe enables quantitative concentration determination with resolution down to 11 mM. The chemical vibration modes of LW region C-D bonds and HW region C-H bonds of the mixture samples of organic compounds and their deuterated counterparts in a simulated cell are simultaneously excited and characterized. The CARS disposable fiber probe introduces a promising handle for in vivo biochemical detection based on isotopic labeling sensing.

Curved fiber compound eye camera inspired by the Strepsiptera vision.

The Strepsiptera vision possesses intriguing features of a large field of view (FOV) and relatively high resolution compared to normal compound eyes. However, it presents a significant challenge of the mismatch between the curved compound eyelet lens array and the planar image sensor to image in a large FOV for artificial compound eyes (ACE). We propose what we believe to be a novel curved fiber compound eye camera (CFCEC) here, which employs coherent fiber bundles as the optical relay system to transmit sub-images curvilinearly. A total of 106 eyelets are arranged based on a scheme similar to the Goldberg polyhedron, with the advantages of uniform interval and minor edge blindness. Then, a prototype of the CFCEC is fabricated and assembled. A series of experiments are conducted to assess the FOV, contrast, resolution, and overlap rate of FOV of the prototype. The results prove that the CFCEC has a total FOV of up to 160°×160° and a total overlap rate of FOV of approximately 65%, demonstrating the promising potential of the CFCEC in various applications, such as panoramic surveillance, 3D detection, and motion tracking.

High sensitivity composite F-P cavity fiber optic sensor based on MEMS for temperature and salinity measurement of seawater.

We proposed an optical fiber salinity sensor with a composite Fabry-Perot (F-P) cavity structure for simultaneous measurement of temperature and salinity based on microelectromechanical system (MEMS) technology. The sensor contains two sensing cavities. The silicon cavity is used for temperature sensing, and the seawater cavity processed by the glass microstructure is sensitive to the refractive index of seawater for salinity sensing. At the same time, the influence of the salinity-temperature cross-sensitivity error of the seawater cavity is effectively compensated by using the temperature single parameter sensitivity characteristics of the silicon cavity. The structural design of the sensor seawater cavity includes a cross-shaped groove and a cylindrical fluid cavity. The surface hydrophilicity treatment was performed on the interior of the cavity to solve the effect of no water injection in the cavity caused by the miniaturization of the sensor. The optical path difference (OPD) demodulation method is used to demodulate the two F-P cavities with large dynamic range and high resolution. In the range of 5∼40°C and 5∼ 40 ‰, the temperature and salinity sensitivity of the sensor can reach 110.25 nm/°C and 178.75 nm/‰, respectively, and the resolution can reach 5.02 × 10-3°C and 0.0138‰. It has the advantages of mass production, high stability, and small size, which give it great potential for marine applications.

Pol2Pol: self-supervised polarimetric image denoising.

In this Letter, we present a self-supervised method, polarization to polarization (Pol2Pol), for polarimetric image denoising with only one-shot noisy images. First, a polarization generator is proposed to generate training image pairs, which are synthesized from one-shot noisy images by exploiting polarization relationships. Second, the Pol2Pol method is extensible and compatible, and any network that performs well in supervised image denoising tasks can be deployed to Pol2Pol after proper modifications. Experimental results show Pol2Pol outperforms other self-supervised methods and achieves comparable performance to supervised methods.

High Sensitivity Temperature Sensing of Long-Period Fiber Grating for the Ocean.

In this study, a new temperature sensor with high sensitivity was achieved by four-layer Ge and B co-doped long-period fiber grating (LPFG) based on the mode coupling principle. By analyzing the mode conversion, the influence of the surrounding refractive index (SRI), the thickness and the refractive index of the film on the sensitivity of the sensor is studied. When 10 nm-thick titanium dioxide (TiO2) film is coated on the surface of the bare LPFG, the refractive index sensitivity of the sensor can be initially improved. Packaging PC452 UV-curable adhesive with a high-thermoluminescence coefficient for temperature sensitization can realize high-sensitivity temperature sensing and meet the requirements of ocean temperature detection. Finally, the effects of salt and protein attachment on the sensitivity are analyzed, which provides a reference for the subsequent application. The sensitivity of 3.8 nm/°C in the range of 5-30 °C was achieved for this new sensor, and the resolution is about 0.00026 °C, which is over 20 times higher than ordinary temperature sensors. This new sensor meets the accuracy and range of general ocean temperature measurements and could be used in various marine monitoring and environmental protection applications.

Highly sensitive and label-free detection of biotin using a liquid crystal-based optofluidic biosensor.

A liquid crystal (LC)-based optofluidic whispering gallery mode (WGM) resonator has been applied as a biosensor to detect biotin. Immobilized streptavidin (SA) act as protein molecules and specifically bind to biotin through strong non-covalent interaction, which can interfere with the orientation of LCs by decreasing the vertical anchoring force of the alignment layer in which the WGM spectral wavelength shift is monitored as a sensing parameter. Due to the double magnification of the LC molecular orientation transition and the resonance of the WGM, the detection limit for SA can reach 1.25 fM (4.7 × 10-13 g/ml). The measurable concentration of biotin and the wavelength shift of the WGM spectrum have an excellent linearity in the range of 0 to 0.1 pg/ml, which can achieve ultra-low detection limit (0.4 fM), i.e., seven orders of magnitude improvement over conventional polarized optical microscope (POM) method. The proposed optofluidic biosensor is highly reproducible and can be used as an ultrasensitive real-time monitoring biosensor, which will open the door for applications to other receptor and ligand models.

2D shape reconstruction of submillimetric irregular rough particles from speckle pattern in interferometric particle imaging measurement.

Particle shape is a significant feature of irregular particles. The interferometric particle imaging (IPI) technique has been introduced to retrieve submillimetric irregular rough particle shapes, while inevitable experimental noises hinder the convergence of two-dimensional (2D) particle shapes from single speckle patterns. In this work, a hybrid input-output algorithm with shrink-wrap support and oversampling smoothness constraints is utilized to suppress the Poisson noise in IPI measurement and recover accurate 2D shapes of particles. Our method is tested in numerical simulations on ice crystal shapes and actual IPI measurements on four different types of irregular, rough particles. The shape similarity of the reconstructed 2D shape has reached an average Jaccard Index score of 0.927, and the relative deviation of the reconstructed size is within 7% for all 60 tested irregular particles at the maximum shot noise level of 7.4%. Furthermore, our method has obviously reduced the uncertainty in the 3D shape reconstruction of irregular, rough particles.

Achievable information rate optimization in C-band optical fiber communication system.

Optical fiber communication networks play an important role in the global telecommunication network. However, nonlinear effects in the optical fiber and transceiver noise greatly limit the performance of fiber communication systems. In this paper, the product of mutual information (MI) and communication bandwidth is used as the metric of the achievable information rate (AIR). The MI loss caused by the transceiver is also considered in this work, and the bit-wise MI, generalized mutual information (GMI), is used to calculate the AIR. This loss is more significant in the use of higher-order modulation formats. The AIR analysis is carried out in the QPSK, 16QAM, 64QAM and 256QAM modulation formats for the communication systems with different communication bandwidths and transmission distances based on the enhanced Gaussian noise (EGN) model. The paper provides suggestions for the selection of the optimal modulation format in different transmission scenarios.

Aqueous Organic Batteries Using the Proton as a Charge Carrier.

Benefiting from the merits of low cost, nonflammability, and high operational safety, aqueous rechargeable batteries have emerged as promising candidates for large-scale energy-storage applications. Among various metal-ion/non-metallic charge carriers, the proton (H+ ) as a charge carrier possesses numerous unique properties such as fast proton diffusion dynamics, a low molar mass, and a small hydrated ion radius, which endow aqueous proton batteries (APBs) with a salient rate capability, a long-term life span, and an excellent low-temperature electrochemical performance. In addition, redox-active organic molecules, with the advantages of structural diversity, rich proton-storage sites, and abundant resources, are considered attractive electrode materials for APBs. However, the charge-storage and transport mechanisms of organic electrodes in APBs are still in their infancy. Therefore, finding suitable electrode materials and uncovering the H+ -storage mechanisms are significant for the application of organic materials in APBs. Herein, the latest research progress on organic materials, such as small molecules and polymers for APBs, is reviewed. Furthermore, a comprehensive summary and evaluation of APBs employing organic electrodes as anode and/or cathode is provided, especially regarding their low-temperature and high-power performances, along with systematic discussions for guiding the rational design and the construction of APBs based on organic electrodes.

Polarized image super-resolution via a deep convolutional neural network.

Reduced resolution of polarized images makes it difficult to distinguish detailed polarization information and limits the ability to identify small targets and weak signals. A possible way to handle this problem is the polarization super-resolution (SR), which aims to obtain a high-resolution polarized image from a low-resolution one. However, compared with the traditional intensity-mode image SR, the polarization SR is more challenging because more channels and their nonlinear cross-links need to be considered as well as the polarization and intensity information need to be reconstructed simultaneously. This paper analyzes the polarized image degradation and proposes a deep convolutional neural network for polarization SR reconstruction based on two degradation models. The network structure and the well-designed loss function have been verified to effectively balance the restoration of intensity and polarization information, and can realize the SR with a maximum scaling factor of four. Experimental results show that the proposed method outperforms other SR methods in terms of both quantitative evaluation and visual effect evaluation for two degradation models with different scaling factors.

Distributed optical fiber biosensor based on optical frequency domain reflectometry.

In situ acquisition of spatial distribution of biochemical substances is important in cell analysis, cancer detection and other fields. Optical fiber biosensors can achieve label-free, fast and accurate measurements. However, current optical fiber biosensors only acquire single-point of biochemical substance content. In this paper, we present a distributed optical fiber biosensor based on tapered fiber in optical frequency domain reflectometry (OFDR) for the first time. To enhance evanescent field at a relative long sensing range, we fabricate a tapered fiber with a taper waist diameter of 6 µm and a total stretching length of 140 mm. Then the human IgG layer is coated on the entire tapered region by polydopamine (PDA) -assisted immobilization as the sensing element to achieve to sense anti-human IgG. We measure shifts of the local Rayleigh backscattering spectra (RBS) caused by the refractive index (RI) change of an external medium surrounding a tapered fiber after immunoaffinity interactions by using OFDR. The measurable concentration of anti-human IgG and RBS shift has an excellent linearity in a range from 0 ng/ml to 14 ng/ml with an effective sensing range of 50 mm. The concentration measurement limit of the proposed distributed biosensor is 2 ng/ml for anti-human IgG. Distributed biosensing based on OFDR can locate a concentration change of anti-human IgG with an ultra-high sensing spatial resolution of 680 µm. The proposed sensor has a potential to realize a micron-level localization of biochemical substances such as cancer cells, which will open a door to transform single-point biosensor to distributed biosensor.

Optimally Configured Optical Fiber Near-Field Enhanced Plasmonic Resonance Immunoprobe for the Detection of Alpha-Fetoprotein.

The detection of trace biomarkers is an important supplementary approach for early screening and diagnoses of tumors. An optical fiber near-field enhanced plasmonic resonance immunoprobe is developed for the detection of the hepatocellular carcinoma biomarker, i.e., the alpha-fetoprotein. Generic principles based on dispersion models and finite element analysis (FEA) models are developed to realize the optimized configuration of spectral characteristics of the immunoprobe. Dispersion models provide theoretical guidance for the design of the multilayer sensing structure from the perspective of the ray optics theory. FEA models provide theoretical guidance for the selection of coating materials from the perspective of the self-defined dielectric constant ratio, i.e., the ratio of the real part to the imaginary part. The optimized configuration of the antibody coupling further improves the biosensing performance of the immunoprobe. The limit of detection (LOD) can reach down to 0.01 ng mL-1 , which is one order of magnitude lower than those relevant reported works. Such a low LOD can more effectively avoid the accuracy degradation of detection results due to measurement errors. Human serum samples have also been detected, with the good precision achieved. This work shows promising prospects in applications of label-free, low-cost, rapid, and convenient early screening of tumors.

Autonomous Microlasers for Profiling Extracellular Vesicles from Cancer Spheroids.

Self-propelled micro/nanomotors are emergent intelligent sensors for analyzing extracellular biomarkers in circulating biological fluids. Conventional luminescent motors are often masked by a highly dynamic and scattered environment, creating challenges to characterize biomarkers or subtle binding dynamics. Here we introduce a strategy to amplify subtle signals by coupling strong light-matter interactions on micromotors. A smart whispering-gallery-mode microlaser that can self-propel and analyze extracellular biomarkers is demonstrated through a liquid crystal microdroplet. Lasing spectral responses induced by cavity energy transfer were employed to reflect the abundance of protein biomarkers, generating exclusive molecular labels for cellular profiling of exosomes derived from 3D multicellular cancer spheroids. Finally, a microfluidic biosystem with different tumor-derived exosomes was employed to elaborate its sensing capability in complex environments. The proposed autonomous microlaser exhibits a promising method for both fundamental biological science and applications in drug screening, phenotyping, and organ-on-chip applications.

Reconstruction error model of distributed shape sensing based on the reentered frame in OFDR.

At present, the reconstruction error of optical fiber shape sensing is commonly represented by Euclidean distance error. However, the Euclidian error of shape reconstruction will be dependent on the shape complexity, which depends on length, curvature and torsion. In this paper, we establish a reconstruction error model of distributed shape sensing in optical frequency domain reflectometry (OFDR) based on the Frenet-Serret frame and the error delivering theory, which illustrates the relationship between the reconstruction error and parameters such as curvature, torsion, fiber length and strain measurement error. We experimentally verify the feasibility and applicability of the proposed reconstruction error model by distributed optical fiber shape sensing system based on OFDR. The proposed reconstruction error model can provide a prediction of the maximal reconstruction error when the estimated range of curvature, torsion, fiber length of a shape needs to be reconstructed and strain measurement errors of OFDR system are known. It is very useful to judge whether the shape reconstruction error meets the requirement according to the shape to be reconstructed.

Defect Detection of MEMS Based on Data Augmentation, WGAN-DIV-DC, and a YOLOv5 Model.

Surface defect detection of micro-electromechanical system (MEMS) acoustic thin film plays a crucial role in MEMS device inspection and quality control. The performances of deep learning object detection models are significantly affected by the number of samples in the training dataset. However, it is difficult to collect enough defect samples during production. In this paper, an improved YOLOv5 model was used to detect MEMS defects in real time. Mosaic and one more prediction head were added into the YOLOv5 baseline model to improve the feature extraction capability. Moreover, Wasserstein divergence for generative adversarial networks with deep convolutional structure (WGAN-DIV-DC) was proposed to expand the number of defect samples and to make the training samples more diverse, which improved the detection accuracy of the YOLOv5 model. The optimal detection model achieved 0.901 mAP, 0.856 F1 score, and a real-time speed of 75.1 FPS. As compared with the baseline model trained using a non-augmented dataset, the mAP and F1 score of the optimal detection model increased by 8.16% and 6.73%, respectively. This defect detection model would provide significant convenience during MEMS production.