Multibranch Wavelet-Based Network for Image Demoiréing.

Moiré patterns caused by aliasing between the camera's sensor and the monitor can severely degrade image quality. Image demoiréing is a multi-task image restoration method that includes texture and color restoration. This paper proposes a new multibranch wavelet-based image demoiréing network (MBWDN) for moiré pattern removal. Moiré images are separated into sub-band images using wavelet decomposition, and demoiréing can be achieved using the different learning strategies of two networks: moiré removal network (MRN) and detail-enhanced moiré removal network (DMRN). MRN removes moiré patterns from low-frequency images while preserving the structure of smooth areas. DMRN simultaneously removes high-frequency moiré patterns and enhances fine details in images. Wavelet decomposition is used to replace traditional upsampling, and max pooling effectively increases the receptive field of the network without losing the spatial information. Through decomposing the moiré image into different levels using wavelet transform, the feature learning results of each branch can be fully preserved and fed into the next branch; therefore, possible distortions in the recovered image are avoided. Thanks to the separation of high- and low-frequency images during feature training, the proposed two networks achieve impressive moiré removal effects. Based on extensive experiments conducted using public datasets, the proposed method shows good demoiréing validity both quantitatively and qualitatively when compared with the state-of-the-art approaches.

Homogeneous Space Construction and Projection for Single-Cell Expression Prediction Based on Deep Learning.

Predicting cellular responses to perturbations is an unsolved problem in biology. Traditional approaches assume that different cell types respond similarly to perturbations. However, this assumption does not take into account the context of genome interactions in different cell types, which leads to compromised prediction quality. More recently, deep learning models used to discover gene-gene relationships can yield more accurate predictions of cellular responses. The huge difference in biological information between different cell types makes it difficult for deep learning models to encode data into a continuous low-dimensional feature space, which means that the features captured by the latent space may not be continuous. Therefore, the mapping relationship between the two conditional spaces learned by the model can only be applied where the real reference data resides, leading to the wrong mapping of the predicted target cells because they are not in the same domain as the reference data. In this paper, we propose an information-navigated variational autoencoder (INVAE), a deep neural network for cell perturbation response prediction. INVAE filters out information that is not conducive to predictive performance. For the remaining information, INVAE constructs a homogeneous space of control conditions, and finds the mapping relationship between the control condition space and the perturbation condition space. By embedding the target unit into the control space and then mapping it to the perturbation space, we can predict the perturbed state of the target unit. Comparing our proposed method with other three state-of-the-art methods on three real datasets, experimental results show that INVAE outperforms existing methods in cell state prediction after perturbation. Furthermore, we demonstrate that filtering out useless information not only improves prediction accuracy but also reveals similarities in how genes in different cell types are regulated following perturbation.

Deep Learning-Based Recognition of Periodontitis and Dental Caries in Dental X-ray Images.

Dental X-ray images are important and useful for dentists to diagnose dental diseases. Utilizing deep learning in dental X-ray images can help dentists quickly and accurately identify common dental diseases such as periodontitis and dental caries. This paper applies image processing and deep learning technologies to dental X-ray images to propose a simultaneous recognition method for periodontitis and dental caries. The single-tooth X-ray image is detected by the YOLOv7 object detection technique and cropped from the periapical X-ray image. Then, it is processed through contrast-limited adaptive histogram equalization to enhance the local contrast, and bilateral filtering to eliminate noise while preserving the edge. The deep learning architecture for classification comprises a pre-trained EfficientNet-B0 and fully connected layers that output two labels by the sigmoid activation function for the classification task. The average precision of tooth detection using YOLOv7 is 97.1%. For the recognition of periodontitis, the area under the curve (AUC) of the receiver operating characteristic (ROC) curve is 98.67%, and the AUC of the precision-recall (PR) curve is 98.38%. For the recognition of dental caries, the AUC of the ROC curve is 98.31%, and the AUC of the PR curve is 97.55%. Different from the conventional deep learning-based methods for a single disease such as periodontitis or dental caries, the proposed approach can provide the recognition of both periodontitis and dental caries simultaneously. This recognition method presents good performance in the identification of periodontitis and dental caries, thus facilitating dental diagnosis.

The Impact of FEV1/FVC Ratio on the Clinical Outcomes in Acute Coronary Syndrome Patients Treated with Dual Anti-Platelet Agents.

Background: Few studies have investigated the clinical efficacy and pulmonary side effects of different P2Y12 inhibitors in acute coronary syndrome (ACS) patients. The aim of this study was to explore the impact of forced expiratory volume in 1 second over forced vital capacity (FEV1/FVC) ratio on the clinical outcomes in ACS patients treated with dual antiplatelet therapy after percutaneous coronary intervention (PCI). Methods: ACS patients who underwent PCI, had documented pre-existing spirometry tests, and received aspirin with either ticagrelor or clopidogrel were enrolled for retrospective analysis. Results: Of the enrolled ACS patients, 275 and 247 received ticagrelor and clopidogrel, respectively. The incidence of wheeze was significantly higher in the ticagrelor group compared to the clopidogrel group within 360 days (14.91% vs. 8.09%, p = 0.016). Multivariable analysis revealed that ticagrelor treatment, as compared to clopidogrel treatment, independently predicted 1-year hospitalization for acute exacerbation (AE) of obstructive airway disease (hazard ratio: 3.44; 95% confidence interval: 1.92 to 6.15; p < 0.01). The receiver operating characteristic curve indicated that an FEV1/FVC ratio of 63.85% had the highest sensitivity and specificity for predicting the incidence of AE of obstructive airway disease within 1 year (p < 0.001). The 1-year hospitalization rate for AE of obstructive airway disease was significantly higher in the ticagrelor group when the FEV1/FVC ratio was < 63%. Conclusions: This study demonstrated higher incidence of wheeze and hospitalization for AE of obstructive airway disease in ACS patients treated with ticagrelor compared to clopidogrel. Furthermore, the FEV1/FVC ratio ≤ 63% in the ACS patients predicted hospitalization for AE of obstructive airway disease in 1 year.

Lightweight Deep Neural Network for Joint Learning of Underwater Object Detection and Color Conversion.

Underwater image processing has been shown to exhibit significant potential for exploring underwater environments. It has been applied to a wide variety of fields, such as underwater terrain scanning and autonomous underwater vehicles (AUVs)-driven applications, such as image-based underwater object detection. However, underwater images often suffer from degeneration due to attenuation, color distortion, and noise from artificial lighting sources as well as the effects of possibly low-end optical imaging devices. Thus, object detection performance would be degraded accordingly. To tackle this problem, in this article, a lightweight deep underwater object detection network is proposed. The key is to present a deep model for jointly learning color conversion and object detection for underwater images. The image color conversion module aims at transforming color images to the corresponding grayscale images to solve the problem of underwater color absorption to enhance the object detection performance with lower computational complexity. The presented experimental results with our implementation on the Raspberry pi platform have justified the effectiveness of the proposed lightweight jointly learning model for underwater object detection compared with the state-of-the-art approaches.

Multi-Scale Deep Residual Learning-Based Single Image Haze Removal via Image Decomposition.

Images/videos captured from outdoor visual devices are usually degraded by turbid media, such as haze, smoke, fog, rain, and snow. Haze is the most common one in outdoor scenes due to the atmosphere conditions. In this paper, a novel deep learning-based architecture (denoted by MSRL-DehazeNet) for single image haze removal relying on multi-scale residual learning (MSRL) and image decomposition is proposed. Instead of learning an end-to-end mapping between each pair of hazy image and its corresponding haze-free one adopted by most existing learningbased approaches, we reformulate the problem as restoration of the image base component. Based on the decomposition of a hazy image into the base and the detail components, haze removal (or dehazing) can be achieved by both of our multi-scale deep residual learning and our simplified U-Net learning only for mapping between hazy and haze-free base components, while the detail component is further enhanced via the other learned convolutional neural network (CNN). Moreover, benefited by the basic building block of our deep residual CNN architecture and our simplified UNet structure, the feature maps (produced by extracting structural and statistical features), and each previous layer can be fully preserved and fed into the next layer. Therefore, possible color distortion in the recovered image would be avoided. As a result, the final haze-removed (or dehazed) image is obtained by integrating the haze-removed base and the enhanced detail image components. Experimental results have demonstrated good effectiveness of the proposed framework, compared with state-ofthe-art approaches.

Using design of experiment for parameter optimization on smart headlamp optics design.

This research focuses on the design of the optical microstructure, and the design of four kinds of light distribution for vehicles' passing beam and driving beam optical structures under the regulation ECE R123. The results show that the passing beam achieves the target light distribution with multiple light patterns superimposed by reflectors, and can meet the four segment light types under the regulations: Class C, Class V, Class E, and Class W. With the structural design method of the reflector, a cutoff line is formed under the structure without a visor to reduce the energy waste caused by the shielding structure, so that the maximum luminosity of the passing beam under the road section can reach 75,980.7 cd and the simulated maximum photometric value can reach 69,705.9 cd under Class W. The driving beam uses the total internal reflection (TIR) lens design to find the optimal 36° angle of the lens to effectively achieve the straightening and brightness enhancement of the light, and then uses the response surface methodology to optimize the optical divergence of the parameters of the microlenticular lens structure on the TIR lens to adjust the width and flatness of the light type. Among them, the radius of curvature, the thickness of the lens, and the length of the single lens are selected as the factors. Using the experimental design method of the reaction surface, the optimal solution of the driving beam design is found. The optimal solution is combined into a radius of curvature of 14.99 mm in the X direction and 25.22 mm in the Y direction, the overall thickness is 1.5 mm, and the length of a single curved surface is 2.43. Each factor is within the limit, and the maximum brightness in the center is 213,866 cd.

Machine Learning for Long Cycle Maintenance Prediction of Wind Turbine.

Within Internet of Things (IoT) sensors, the challenge is how to dig out the potentially valuable information from the collected data to support decision making. This paper proposes a method based on machine learning to predict long cycle maintenance time of wind turbines for efficient management in the power company. Long cycle maintenance time prediction makes the power company operate wind turbines as cost-effectively as possible to maximize the profit. Sensor data including operation data, maintenance time data, and event codes are collected from 31 wind turbines in two wind farms. Data aggregation is performed to filter out some errors and get significant information from the data. Then, the hybrid network is built to train the predictive model based on the convolutional neural network (CNN) and support vector machine (SVM). The experimental results show that the prediction of the proposed method reaches high accuracy, which helps drive up the efficiency of wind turbine maintenance.

Transforming Retinal Photographs to Entropy Images in Deep Learning to Improve Automated Detection for Diabetic Retinopathy.

Entropy images, representing the complexity of original fundus photographs, may strengthen the contrast between diabetic retinopathy (DR) lesions and unaffected areas. The aim of this study is to compare the detection performance for severe DR between original fundus photographs and entropy images by deep learning. A sample of 21,123 interpretable fundus photographs obtained from a publicly available data set was expanded to 33,000 images by rotating and flipping. All photographs were transformed into entropy images using block size 9 and downsized to a standard resolution of 100 × 100 pixels. The stages of DR are classified into 5 grades based on the International Clinical Diabetic Retinopathy Disease Severity Scale: Grade 0 (no DR), Grade 1 (mild nonproliferative DR), Grade 2 (moderate nonproliferative DR), Grade 3 (severe nonproliferative DR), and Grade 4 (proliferative DR). Of these 33,000 photographs, 30,000 images were randomly selected as the training set, and the remaining 3,000 images were used as the testing set. Both the original fundus photographs and the entropy images were used as the inputs of convolutional neural network (CNN), and the results of detecting referable DR (Grades 2-4) as the outputs from the two data sets were compared. The detection accuracy, sensitivity, and specificity of using the original fundus photographs data set were 81.80%, 68.36%, 89.87%, respectively, for the entropy images data set, and the figures significantly increased to 86.10%, 73.24%, and 93.81%, respectively (all p values <0.001). The entropy image quantifies the amount of information in the fundus photograph and efficiently accelerates the generating of feature maps in the CNN. The research results draw the conclusion that transformed entropy imaging of fundus photographs can increase the machinery detection accuracy, sensitivity, and specificity of referable DR for the deep learning-based system.

Machine learning to relate PM2.5 and PM10 concentrations to outpatient visits for upper respiratory tract infections in Taiwan: A nationwide analysis.

AIM: To examine the accuracy of machine learning to relate particulate matter (PM) 2.5 and PM10 concentrations to upper respiratory tract infections (URIs). METHODS: Daily nationwide and regional outdoor PM2.5 and PM10 concentrations collected over 30 consecutive days obtained from the Taiwan Environment Protection Administration were the inputs for machine learning, using multilayer perceptron (MLP), to relate to the subsequent one-week outpatient visits for URIs. The URI data were obtained from the Centers for Disease Control datasets in Taiwan between 2009 and 2016. The testing used the middle month dataset of each season (January, April, July and October), and the training used the other months' datasets. The weekly URI cases were classified by tertile as high, moderate, and low volumes. RESULTS: Both PM concentrations and URI cases peak in winter and spring. In the nationwide data analysis, MLP machine learning can accurately relate the URI volumes of the elderly (89.05% and 88.32%, respectively) and the overall population (81.75% and 83.21%, respectively) with the PM2.5 and PM10 concentrations. In the regional data analyses, greater accuracy is found for PM2.5 than for PM10 for the elderly, particularly in the Central region (78.10% and 74.45%, respectively), whereas greater accuracy is found for PM10 than for PM2.5 for the overall population, particularly in the Northern region (73.19% and 63.04%, respectively). CONCLUSION: Short-term PM2.5 and PM10 concentrations were accurately related to the subsequent occurrence of URIs by using machine learning. Our findings suggested that the effects of PM2.5 and PM10 on URI may differ by age, and the mechanism needs further evaluation.

Tensile force on human macrophage cells promotes osteoclastogenesis through receptor activator of nuclear factor κB ligand induction.

Little is known about the effects of tensile forces on osteoclastogenesis by human monocytes in the absence of mechanosensitive cells, including osteoblasts and fibroblasts. In this study we consider the effects of tensile force on osteoclastogenesis in human monocytes. The cells were treated with receptor activator of nuclear factor κB ligand (RANKL) to promote osteoclastogenesis. Then,expression and secretion of cathepsin K were examined. RANKL and the formation of osteoclasts during the osteoclast differentiation process under continual tensile stress were evaluated by Western blot. It was also found that -100 kPa or lower induces RANKL-enhanced tartrate-resistant acid phosphatase activity in a dose-dependent manner. Furthermore, an increased tensile force raises the expression and secretion of cathepsin K elevated by RANKL, and is concurrent with the increase of TNF-receptor-associated factor 6 induction and nuclear factor κB activation. Overall, the current report demonstrates that tensile force reinforces RANKL-induced osteoclastogenesis by retarding osteoclast differentiation. The tensile force is able to modify every cell through dose-dependent in vitro RANKL-mediated osteoclastogenesis, affecting the fusion of preosteoclasts and function of osteoclasts. However, tensile force increased TNF-receptor-associated factor 6 expression. These results are in vitro findings and were obtained under a condition of tensile force. The current results help us to better understand the cellular roles of human macrophage populations in osteoclastogenesis as well as in alveolar bone remodeling when there is tensile stress.

Poly(dopamine) coating of 3D printed poly(lactic acid) scaffolds for bone tissue engineering.

3D printing is a versatile technique to generate large quantities of a wide variety of shapes and sizes of polymer. The aim of this study is to develop functionalized 3D printed poly(lactic acid) (PLA) scaffolds and use a mussel-inspired surface coating to regulate cell adhesion, proliferation and differentiation of human adipose-derived stem cells (hADSCs). We prepared PLA 3D scaffolds coated with polydopamine (PDA). The chemical composition and surface properties of PDA/PLA were characterized by XPS. PDA/PLA modulated hADSCs' responses in several ways. Firstly, adhesion and proliferation, and cell cycle of hADSCs cultured on PDA/PLA were significantly enhanced relative to those on PLA. In addition, the collagen I secreted from cells was increased and promoted cell attachment and cell cycle progression were depended on the PDA content. In osteogenesis assay, the ALP activity and osteocalcin of hADSCs cultured on PDA/PLA were significantly higher than seen in those cultured on pure PLA scaffolds. Moreover, hADSCs cultured on PDA/PLA showed up-regulation of the ang-1 and vWF proteins associated with angiogenic differentiation. Our results demonstrate that the bio-inspired coating synthetic PLA polymer can be used as a simple technique to render the surfaces of synthetic scaffolds active, thus enabling them to direct the specific responses of hADSCs.

Antibacterial and Odontogenesis Efficacy of Mineral Trioxide Aggregate Combined with CO2 Laser Treatment.

INTRODUCTION: Mineral trioxide aggregate (MTA) has been successfully used in clinical applications in endodontics. Studies show that the antibacterial effects of CO2 laser irradiation are highly efficient when bacteria are embedded in biofilm because of a photothermal mechanism. The aim of this study was to confirm the effects of CO2 laser irradiation on MTA with regard to both material characterization and cell viability. METHODS: MTA was irradiated with a dental CO2 laser using directly mounted fiber optics in the wound healing mode with a spot area of 0.25 cm(2) and then stored in an incubator at 100% relative humidity and 37°C for 1 day to set. The human dental pulp cells cultured on MTA were analyzed along with their proliferation and odontogenic differentiation behaviors. RESULTS: The results indicate that the setting time of MTA after irradiation by the CO2 laser was significantly reduced to 118 minutes rather than the usual 143 minutes. The maximum diametral tensile strength and x-ray diffraction patterns were similar to those obtained without CO2 laser irradiation. However, the CO2 laser irradiation increased the amount of Ca and Si ions released from the MTA and regulated cell behavior. CO2 laser-irradiated MTA promoted odontogenic differentiation of hDPCs, with the increased formation of mineralized nodules on the substrate's surface. It also up-regulated the protein expression of multiple markers of odontogenic and the expression of dentin sialophosphoprotein protein. CONCLUSIONS: The current study provides new and important data about the effects of CO2 laser irradiation on MTA with regard to the decreased setting time and increased ion release. Taking cell functions into account, the Si concentration released from MTA with laser irradiation may be lower than a critical value, and this information could lead to the development of new regenerative therapies for dentin and periodontal tissue.

Poly(Dopamine)-Assisted Immobilization of Xu Duan on 3D Printed Poly(Lactic Acid) Scaffolds to Up-Regulate Osteogenic and Angiogenic Markers of Bone Marrow Stem Cells.

Three-dimensional printing is a versatile technique to generate large quantities of a wide variety of shapes and sizes of polymer. The aim of this study is to develop functionalized 3D printed poly(lactic acid) (PLA) scaffolds and use a mussel-inspired surface coating and Xu Duan (XD) immobilization to regulate cell adhesion, proliferation and differentiation of human bone-marrow mesenchymal stem cells (hBMSCs). We prepared PLA scaffolds and coated with polydopamine (PDA). The chemical composition and surface properties of PLA/PDA/XD were characterized by XPS. PLA/PDA/XD controlled hBMSCs' responses in several ways. Firstly, adhesion and proliferation of hBMSCs cultured on PLA/PDA/XD were significantly enhanced relative to those on PLA. In addition, the focal adhesion kinase (FAK) expression of cells was increased and promoted cell attachment depended on the XD content. In osteogenesis assay, the osteogenesis markers of hBMSCs cultured on PLA/PDA/XD were significantly higher than seen in those cultured on a pure PLA/PDA scaffolds. Moreover, hBMSCs cultured on PLA/PDA/XD showed up-regulation of the ang-1 and vWF proteins associated with angiogenic differentiation. Our results demonstrate that the bio-inspired coating synthetic PLA polymer can be used as a simple technique to render the surfaces of synthetic scaffolds active, thus enabling them to direct the specific responses of hBMSCs.

Stimulatory effects of the fast setting and suitable degrading Ca-Si-Mg cement on both cementogenesis and angiogenesis differentiation of human periodontal ligament cells.

The purpose of this study is to develop a fast setting and suitable degrading Mg-calcium silicate cement (Mg-CS) and a mechanism using Mg ions to stimulate human periodontal ligament cells (hPDLCs). Mechanical strength and stability have been determined by testing the diametral tensile strength; the degradation of cements has been measured by ascertaining the number of ions released in simulated body fluid. Other cell characteristics such as proliferation, differentiation and mineralization, and hPDLCs when cultured on cement surfaces were also examined. The results show that the degradation rate of Mg-CS cements depends on the Mg content in CS. Regarding in vitro bioactivity, the CS cements were covered with clusters of apatite spherulites after immersion for 30 days, while there was less formation of apatite spherulites on the Mg-rich cement surfaces. In addition, researchers also explored the effects of Mg ions on the cementogenesis and angiogenesis differentiation of hPDLCs in comparison with pure CS cement. The proliferation, alkaline phosphatase, cementogenesis-related proteins (CEMP1 and CAP), and angiogenesis-related protein (vWF and ang-1) secretion of hPDLCs were significantly stimulated when the Mg ion concentration of the medium was increased. The research results also suggest that Mg-CS cements with this modified composition stimulate hPDLC behaviour and so may be good biomaterials for bone substitutes and hard tissue regeneration applications as they stimulate cementogenesis/angiogenesis.

Haze effect removal from image via haze density estimation in optical model.

Images/videos captured from optical devices are usually degraded by turbid media such as haze, smoke, fog, rain and snow. Haze is the most common problem in outdoor scenes because of the atmosphere conditions. This paper proposes a novel single image-based dehazing framework to remove haze artifacts from images, where we propose two novel image priors, called the pixel-based dark channel prior and the pixel-based bright channel prior. Based on the two priors with the haze optical model, we propose to estimate atmospheric light via haze density analysis. We can then estimate transmission map, followed by refining it via the bilateral filter. As a result, high-quality haze-free images can be recovered with lower computational complexity compared with the state-of-the-art approach based on patch-based dark channel prior.

Robust laser speckle recognition system for authenticity identification.

This paper proposes a laser speckle recognition system for authenticity verification. Because of the unique imperfection surfaces of objects, laser speckle provides identifiable features for authentication. A Gabor filter, SIFT (Scale-Invariant Feature Transform), and projection were used to extract the features of laser speckle images. To accelerate the matching process, the extracted Gabor features were organized into an indexing structure using the K-means algorithm. Plastic cards were used as the target objects in the proposed system and the hardware of the speckle capturing system was built. The experimental results showed that the retrieval performance of the proposed method is accurate when the database contains 516 laser speckle images. The proposed system is robust and feasible for authenticity verification.