Search

Microscopy is essential for detecting, identifying, analyzing, and measuring small objects. Access to modern microscopy equipment is crucial for scientific research, especially in the biomedical and analytical sciences. However, the high cost of equipment, limited availability of parts, and challenges associated with transporting equipment often limit the accessibility and operational capabilities of these tools, particularly in field sites and other remote or resource-limited settings. Thus, there is a need for affordable and accessible alternatives to traditional microscopy systems. We address this challenge by investigating the feasibility of using a simple microcontroller board not only as a portable and field-ready digital microscope, but furthermore as a versatile platform which can easily be adapted to a variety of imaging applications. By adding a few external components, we demonstrate that a low-cost ESP32 camera board can be used to build an autonomous in situ platform for digital time-lapse imaging of cells. Our prototype of this approach, which we call ESPressoscope, can be adapted to applications ranging from monitoring incubator cell cultures in the lab to observing ecological phenomena in the sea, and it can be adapted for other techniques such as microfluidics or spectrophotometry. Our prototype of the ESPressoscope concept achieves a low power consumption and small size, which makes it ideal for field research in environments and applications where microscopy was previously infeasible. Its Wi-Fi connectivity enables integration with external image processing and storage systems, including on cloud platforms when internet access is available. Finally, we present several web browser-based tools to help users operate and manage our prototype's software. Our findings demonstrate the potential for low-cost, portable microscopy solutions to enable new and more accessible experiments for biological and analytical applications. [ABSTRACT FROM AUTHOR]

Background: The pathology in Duchenne muscular dystrophy (DMD) is characterized by degenerating muscle fibers, inflammation, fibro-fatty infiltrate, and edema, and these pathological processes replace normal healthy muscle tissue. The mdx mouse model is one of the most commonly used preclinical models to study DMD. Mounting evidence has emerged illustrating that muscle disease progression varies considerably in mdx mice, with inter-animal differences as well as intra-muscular differences in pathology in individual mdx mice. This variation is important to consider when conducting assessments of drug efficacy and in longitudinal studies. We developed a magnetic resonance imaging (MRI) segmentation and analysis pipeline to rapidly and non-invasively measure the severity of muscle disease in mdx mice. Methods: Wildtype and mdx mice were imaged with MRI and T2 maps were obtained axially across the hindlimbs. A neural network was trained to rapidly and semi-automatically segment the muscle tissue, and the distribution of resulting T2 values was analyzed. Interdecile range and Pearson Skew were identified as biomarkers to quickly and accurately estimate muscle disease severity in mice. Results: The semiautomated segmentation tool reduced image processing time approximately tenfold. Measures of Pearson skew and interdecile range based on that segmentation were repeatable and reflected muscle disease severity in healthy wildtype and diseased mdx mice based on both qualitative observation of images and correlation with Evans blue dye uptake. Conclusion: Use of this rapid, non-invasive, semi-automated MR image segmentation and analysis pipeline has the potential to transform preclinical studies, allowing for pre-screening of dystrophic mice prior to study enrollment to ensure more uniform muscle disease pathology across treatment groups, improving study outcomes. [ABSTRACT FROM AUTHOR]

Advances in fluorescence imaging technology have been crucial to the progress of neuroscience. Whether it was specific expression of indicator proteins, detection of neurotransmitters, or miniaturization of fluorescence microscopes, fluorescence imaging has improved upon electrophysiology, the gold standard for monitoring brain activity, and enabled novel methods to sense activity in the brain. Hence, we developed a lightweight and compact implantable CMOS-based lensless Ca2+ imaging device for freely moving transgenic G-CaMP mouse experiments. However, without a lens system, determination of regions of interest (ROI) has proven challenging. Localization of fluorescence activity and separation of signal from noise are difficult. In this study, we report an ROI selection method using a series of adaptive binarizations with a gaussian method and morphological image processing. The parameters for each operation such as the kernel size, sigma and footprint size were optimized. We then validated the utility of the algorithm with simulated data and freely moving nociception experiments using the lensless devices. The device was implanted in the dorsal raphe nucleus to observe pain-related brain activity following a formalin test to stimulate pain. We observed significant increases in fluorescence activity after formalin injection compared to the control group when using the ROI determination algorithm. [ABSTRACT FROM AUTHOR]

In the field of image processing, common noise types include Gaussian noise, salt and pepper noise, speckle noise, uniform noise and pulse noise. Different types of noise require different denoising algorithms and techniques to maintain image quality and fidelity. Traditional image denoising methods not only remove image noise, but also result in the detail loss in the image. It cannot guarantee the clean removal of noise information while preserving the true signal of the image. To address the aforementioned issues, an image denoising method combining an improved threshold function and wavelet transform is proposed in the experiment. Unlike traditional threshold functions, the improved threshold function is a continuous function that can avoid the pseudo Gibbs effect after image denoising and improve image quality. During the process, the output image of the finite ridge wave transform is first combined with the wavelet transform to improve the denoising performance. Then, an improved threshold function is introduced to enhance the quality of the reconstructed image. In addition, to evaluate the performance of different algorithms, different densities of Gaussian noise are added to Lena images of black, white, and color in the experiment. The results showed that when adding 0.010.01 variance Gaussian noise to black and white images, the peak signal-to-noise ratio of the research method increased by 2.58dB in a positive direction. The mean square error decreased by 0.10dB. When using the algorithm for denoising, the research method had a minimum denoising time of only 13ms, which saved 9ms and 3ms compared to the hard threshold algorithm (Hard TA) and soft threshold algorithm (Soft TA), respectively. The research method exhibited higher stability, with an average similarity error fluctuating within 0.89%. The above results indicate that the research method has smaller errors and better system stability in image denoising. It can be applied in the field of digital image denoising, which can effectively promote the positive development of image denoising technology to a certain extent. [ABSTRACT FROM AUTHOR]

The global impact of the SARS-CoV-2 pandemic has underscored the need for a deeper understanding of viral evolution to anticipate new viruses or variants. Genetic recombination is a fundamental mechanism in viral evolution, yet it remains poorly understood. In this study, we conducted a comprehensive research on the genetic regions associated with genetic recombination features in SARS-CoV-2. With this aim, we implemented a two-phase transfer learning approach using genomic spectrograms of complete SARS-CoV-2 sequences. In the first phase, we utilized a pre-trained VGG-16 model with genomic spectrograms of HIV-1, and in the second phase, we applied HIV-1 VGG-16 model to SARS-CoV-2 spectrograms. The identification of key recombination hot zones was achieved using the Grad-CAM interpretability tool, and the results were analyzed by mathematical and image processing techniques. Our findings unequivocally identify the SARS-CoV-2 Spike protein (S protein) as the pivotal region in the genetic recombination feature. For non-recombinant sequences, the relevant frequencies clustered around 1/6 and 1/12. In recombinant sequences, the sharp prominence of the main hot zone in the Spike protein prominently indicated a frequency of 1/6. These findings suggest that in the arithmetic series, every 6 nucleotides (two triplets) in S may encode crucial information, potentially concealing essential details about viral characteristics, in this case, recombinant feature of a SARS-CoV-2 genetic sequence. This insight further underscores the potential presence of multifaceted information within the genome, including mathematical signatures that define an organism's unique attributes. [ABSTRACT FROM AUTHOR]

In view of the special engineering properties of red clay and the waste of phosphogypsum resources, the expansion and contraction deformation and fissure evolution of phosphogypsum stabilized red clay under different conditions were investigated by laboratory tests and image processing system. The research results show that: (1) the absolute expansion and absolute shrinkage of phosphogypsum stabilized red clay are positively correlated with the compaction degree, the number of dry and wet cycles and the cement dosage, and negatively correlated with the initial water content and the phosphogypsum dosage; (2) the fissure rate increases with the increase of the number of dry and wet cycles, and decreases with the increase of the initial water content, the compaction degree, the cement, and the phosphogypsum dosage; (3) The relationship among absolute expansion rate (absolute shrinkage), degree of compaction and fracture rate can be fitted by the equation f(x,y) = ax+by+cx2+dy2+e; (4) Phosphogypsum has an obvious inhibiting effect on the expansion, shrinkage and cracking of the mix. It is recommended that the cement mixing amount of 6% and phosphogypsum: red clay = 1:1~1:2 as roadbed filler. [ABSTRACT FROM AUTHOR]

With the development of society and changes in the human living environment, people are increasingly attaching importance to their own health. Regarding medical imaging examinations of certain parts of the body, the process of medical image segmentation has become extremely important. This paper presents a novel hybrid algorithm: SAOBL-IA, a fusion of the Simulated Annealing(SA), Opposition-based Learning(OBL)and Island Algorithm(IA). The Island Algorithm itself suffers from slow convergence speed and the tendency to get stuck in local optimum. To address these limitations, we introduce opposition-based learning to enhance the search range and avoid local optimum. Furthermore, we leverage the simulated annealing approach to accelerate the convergence of SAOBL-IA. Comparing the experimental results, it can be seen that SAOBL-IA has better comprehensive performance. Subsequently, the SAOBL-IA algorithm is utilized in conjunction with an optimized two-dimensional OTSU fusion segmentation technique for the purpose of medical image processing. This study proposes an application of image segmentation based on the SAOBL-IA. The segmentation of pixels around the background and target regions using the two-dimensional OTSU method faces challenges in terms of accuracy. To address this issue, an adaptive thresholding technique known as Adaptive Forking is employed for optimization. By determining the slope of the fork based on the misclassified pixel ratio, enhanced segmentation accuracy can be achieved. This improved approach is then integrated with the SAOBL-IA algorithm and applied to the segmentation of lung medical images. The experimental findings show that the amalgamation of SAOBL-IA with the adaptive two-dimensional OTSU segmentation approach, as proposed in this study, manifests superior segmentation speed and enhanced precision in the context of medical image segmentation. [ABSTRACT FROM AUTHOR]

The classification of vehicles presents notable challenges within the domain of image processing. Traditional models suffer from inefficiency, prolonged training times for datasets, intricate feature extraction, and variable assignment complexities for classification. Conventional methods applied to categorize vehicles from extensive datasets often lead to errors, misclassifications, and unproductive outcomes. Consequently, leveraging machine learning techniques emerges as a promising solution to tackle these challenges. This study adopts a machine learning approach to alleviate image misclassifications and manage large quantities of vehicle images effectively. Specifically, a contrast enhancement technique is employed in the pre-processing stage to highlight pixel values in vehicle images. In the feature segmentation stage, Mask-R-CNN is utilized to categorize pixels into predefined classes. VGG16 is then employed to extract features from vehicle images, while an autoencoder aids in selecting features by learning non-linear input features and compressing representation features. Finally, the CatBoost (CB) algorithm is implemented for vehicle classification (VC) in diverse critical environments, such as inclement weather, twilight, and instances of vehicle blockage. Extensive experiments are conducted using different large-scale datasets with various machine learning platforms. The findings indicate that CB (presumably a specific method or algorithm) attains the highest level of performance on the large-scale dataset named UFPR-ALPR, with an accuracy rate of 98.89%. [ABSTRACT FROM AUTHOR]

In this study, we employed the principle of Relative Mode Transfer Method (RMTM) to establish a model for a single pendulum subjected to sudden changes in its length. An experimental platform for image processing was constructed to accurately track the position of a moving ball, enabling experimental verification of the pendulum's motion under specific operating conditions. The experimental data demonstrated excellent agreement with simulated numerical results, validating the effectiveness of the proposed methodology. Furthermore, we performed simulations of a double obstacle pendulum system, investigating the effects of different parameters, including obstacle pin positions, quantities, and initial release angles, on the pendulum's motion through numerical simulations. This research provides novel insights into addressing the challenges associated with abrupt and continuous changes in pendulum length. [ABSTRACT FROM AUTHOR]

In the realm of digital image applications, image processing technology occupies a pivotal position, with image segmentation serving as a foundational component. As the digital image application domain expands across industries, the conventional segmentation techniques increasingly challenge to cater to modern demands. To address this gap, this paper introduces an MCMC-based image segmentation algorithm based on the Markov Random Field (MRF) model, marking a significant stride in the field. The novelty of this research lies in its method that capitalizes on domain information in pixel space, amplifying the local segmentation precision of image segmentation algorithms. Further innovation is manifested in the development of an adaptive segmentation image denoising algorithm based on MCMC sampling. This algorithm not only elevates image segmentation outcomes, but also proficiently denoises the image. In the experimental results, MRF-MCMC achieves better segmentation performance, with an average segmentation accuracy of 94.26% in Lena images, significantly superior to other common image segmentation algorithms. In addition, the study proposes that the denoising model outperforms other algorithms in peak signal-to-noise ratio and structural similarity in environments with noise standard deviations of 15, 25, and 50. In essence, these experimental findings affirm the efficacy of this study, opening avenues for refining digital image segmentation methodologies. [ABSTRACT FROM AUTHOR]

To facilitate the characterization of unlabeled induced pluripotent stem cells (iPSCs) during culture and expansion, we developed an AI pipeline for nuclear segmentation and mitosis detection from phase contrast images of individual cells within iPSC colonies. The analysis uses a 2D convolutional neural network (U-Net) plus a 3D U-Net applied on time lapse images to detect and segment nuclei, mitotic events, and daughter nuclei to enable tracking of large numbers of individual cells over long times in culture. The analysis uses fluorescence data to train models for segmenting nuclei in phase contrast images. The use of classical image processing routines to segment fluorescent nuclei precludes the need for manual annotation. We optimize and evaluate the accuracy of automated annotation to assure the reliability of the training. The model is generalizable in that it performs well on different datasets with an average F1 score of 0.94, on cells at different densities, and on cells from different pluripotent cell lines. The method allows us to assess, in a non-invasive manner, rates of mitosis and cell division which serve as indicators of cell state and cell health. We assess these parameters in up to hundreds of thousands of cells in culture for more than 36 hours, at different locations in the colonies, and as a function of excitation light exposure. [ABSTRACT FROM AUTHOR]

The integration of artificial intelligence (AI) in diagnosing diabetic retinopathy, a major contributor to global vision impairment, is becoming increasingly pronounced. Notably, the detection of vision-threatening diabetic retinopathy (VTDR) has been significantly fortified through automated techniques. Traditionally, the reliance on manual analysis of retinal images, albeit slow and error-prone, constituted the conventional approach. Addressing this, our study introduces a novel methodology that amplifies the robustness and precision of the detection process. This is complemented by the groundbreaking Hierarchical Block Attention (HBA) and HBA-U-Net architecture, which notably propel attention mechanisms in image segmentation. This innovative model refines image processing without imposing excessive computational demands by honing in on individual pixel intricacies, spatial relationships, and channel-specific attention. Building upon this innovation, our proposed method employs a multi-stage strategy encompassing data pre-processing, feature extraction via a hybrid CNN-SVD model, and classification employing an amalgamation of Improved Support Vector Machine-Radial Basis Function (ISVM-RBF), DT, and KNN techniques. Rigorously tested on the IDRiD dataset classified into five severity tiers, the hybrid model yields remarkable performance, achieving a 99.18% accuracy, 98.15% sensitivity, and 100% specificity in VTDR detection, thus surpassing existing methods. These results underscore a more potent avenue for diagnosing and addressing this crucial ocular condition while underscoring AI's transformative potential in medical care, particularly in ophthalmology. [ABSTRACT FROM AUTHOR]

In the context of intelligent driving, pedestrian detection faces challenges related to low accuracy in target recognition and positioning. To address this issue, a pedestrian detection algorithm is proposed that integrates a large kernel attention mechanism with the YOLOV5 lightweight model. The algorithm aims to enhance long-term attention and dependence during image processing by fusing the large kernel attention module with the C3 module. Furthermore, it addresses the lack of long-distance relationship information in channel and spatial feature extraction and representation by introducing the Coordinate Attention mechanism. This mechanism effectively extracts local information and focused location details, thereby improving detection accuracy. To improve the positioning accuracy of obscured targets, the alpha CIOU bounding box regression loss function is employed. It helps mitigate the impact of occlusions and enhances the algorithm's ability to precisely localize pedestrians. To evaluate the effectiveness of trained model, experiments are conducted on the BDD100K pedestrian dataset as well as the Pascal VOC dataset. Experimental results demonstrate that the improved attention fusion YOLOV5 lightweight model achieves an average accuracy of 60.3%. Specifically, the detection accuracy improves by 1.1% compared to the original YOLOV5 algorithm, and the accuracy performance index reaches 73.0%. These findings strongly indicate the proposed algorithm in significantly enhancing the accuracy of pedestrian detection in road scenes. [ABSTRACT FROM AUTHOR]

In the context of Industry 4.0, manufacturing metrology is crucial for inspecting and measuring machines. The Internet of Things (IoT) technology enables seamless communication between advanced industrial devices through local and cloud computing servers. This study investigates the use of the MQTT protocol to enhance the performance of circularity measurement data transmission between cloud servers and round-hole data sources through Open CV. Accurate inspection of circular characteristics, particularly roundness errors, is vital for lubricant distribution, assemblies, and rotational force innovation. Circularity measurement techniques employ algorithms like the minimal zone circle tolerance algorithm. Vision inspection systems, utilizing image processing techniques, can promptly and accurately detect quality concerns by analyzing the model's surface through circular dimension analysis. This involves sending the model's image to a computer, which employs techniques such as Hough Transform, Edge Detection, and Contour Analysis to identify circular features and extract relevant parameters. This method is utilized in the camera industry and component assembly. To assess the performance, a comparative experiment was conducted between the non-contact-based 3SMVI system and the contact-based CMM system widely used in various industries for roundness evaluation. The CMM technique is known for its high precision but is time-consuming. Experimental results indicated a variation of 5 to 9.6 micrometers between the two methods. It is suggested that using a high-resolution camera and appropriate lighting conditions can further enhance result precision. [ABSTRACT FROM AUTHOR]

Low-dose computed tomography (LDCT) has attracted significant attention in the domain of medical imaging due to the inherent risks of normal-dose computed tomography (NDCT) based X-ray radiations to patients. However, reducing radiation dose in CT imaging produces noise and artifacts that degrade image quality and subsequently hinders medical disease diagnostic performance. In order to address these problems, this research article presents a competent low-dose computed tomography image denoising algorithm based on a constructive non-local means algorithm with morphological residual processing to achieve the task of removing noise from the LDCT images. We propose an innovative constructive non-local image filtering algorithm by means of applications in low-dose computed tomography technology. The nonlocal mean filter that was recently proposed was modified to construct our denoising algorithm. It constructs the discrete property of neighboring filtering to enable rapid vectorized and parallel implantation in contemporary shared memory computer platforms while simultaneously decreases computing complexity. Subsequently, the proposed method performs faster computation compared to a non-vectorized and serial implementation in terms of speed and scales linearly with image dimension. In addition, the morphological residual processing is employed for the purpose of edge-preserving image processing. It combines linear lowpass filtering with a nonlinear technique that enables the extraction of meaningful regions where edges could be preserved while removing residual artifacts from the images. Experimental results demonstrate that the proposed algorithm preserves more textural and structural features while reducing noise, enhances edges and significantly improves image quality more effectively. The proposed research article obtains better results both qualitatively and quantitively when compared to other comparative algorithms on publicly accessible datasets. [ABSTRACT FROM AUTHOR]

Parts produced with metal additive manufacturing often suffer from a poor surface finish. Surface finishing techniques are effective to improve the quality of 3D printed surfaces, however they have as downsides that they also slightly change the geometry of the part, in an unpredictable way. This effect on the geometrical features of complex parts has received little attention. In this research, we illustrate a method to visualize the impact of surface finishing techniques on geometrical features, as well as their effectiveness on parts with high shape-complexity, by using centrifugal disk finishing as a case study. We designed and 3D printed test parts with different features using selective laser melting, which were coated with a blue metal lacquer prior to polishing. After polishing, the blue lacquer was eroded away from the spots that were easily reached by the polishing process, yet had remained on the surfaces that could not be reached by the process. We used measurements of material removal and image processing of the remaining blue lacquer on the surfaces to analyze these effects. Using this method, we were able to derive a number of specific design guidelines that can be incorporated while designing metal AM parts for centrifugal disk finishing. We suggest that this visualization method can be applied to different polishing methods to gain insight into their influence, as well as being used as an aid in the design process. [ABSTRACT FROM AUTHOR]

In the latest research progress, deep neural networks have been revolutionized by frameworks to extract image features more accurately. In this study, we focus on an attention model that can be useful in deep neural networks and propose a simple but strong feature extraction deep network architecture, W-Net. The architecture of our W-Net network has two mutually independent path structures, and it is designed with the following advantages. (1) There are two independent effective paths in our proposed network structure, and the two paths capture more contextual information from different scales in different ways. (2) The two paths acquire different feature images, and in the upsampling approach, we use bilinear interpolation thus reducing the feature map distortion phenomenon and integrating the different images processed. (3) The feature image processing is at a bottleneck, and a hierarchical attention module is constructed at the bottleneck by reclassifying after the channel attention module and the spatial attention module, resulting in more efficient and accurate processing of feature images. During the experiment, we also tested iSAID, a massively high spatial resolution remote sensing image dataset, with further experimental data comparison to demonstrate the generality of our method for remote sensor image segmentation. [ABSTRACT FROM AUTHOR]

Eye diseases such as diabetic retinopathy are progressive with various changes in the retinal vessels, and it is difficult to analyze the disease for future treatment. There are many computerized algorithms implemented for retinal vessel segmentation, but the tiny vessels drop off, impacting the performance of the overall algorithms. This research work contains the new image processing techniques such as enhancement filters, coherence filters and binary thresholding techniques to handle the different color retinal fundus image problems to achieve a vessel image that is well-segmented, and the proposed algorithm has improved performance over existing work. Our developed technique incorporates morphological techniques to address the center light reflex issue. Additionally, to effectively resolve the problem of insufficient and varying contrast, our developed technique employs homomorphic methods and Wiener filtering. Coherent filters are used to address the coherence issue of the retina vessels, and then a double thresholding technique is applied with image reconstruction to achieve a correctly segmented vessel image. The results of our developed technique were evaluated using the STARE and DRIVE datasets and it achieves an accuracy of about 0.96 and a sensitivity of 0.81. The performance obtained from our proposed method proved the capability of the method which can be used by ophthalmology experts to diagnose ocular abnormalities and recommended for further treatment. [ABSTRACT FROM AUTHOR]

This work addresses the declining physical fitness levels observed in both football players and the general population. The objective is to investigate the impact of functional strength training on the physical capabilities of football players and to develop a machine learning-based approach for posture recognition. A total of 116 adolescents aged 8 to 13 participating in football training are randomly assigned to either an experimental group (n = 60) or a control group (n = 56). Both groups underwent 24 training sessions, with the experimental group engaging in 15–20 minutes of functional strength training after each session. Machine learning techniques, specifically the backpropagation neural network (BPNN) in deep learning, are utilized to analyze the kicking actions of football players. Movement speed, sensitivity, and strength are employed as input vectors for the BPNN to compare the images of players' movements, while the similarity between the kicking actions and standard movements served as the output result to enhance training efficiency. The experimental group's kicking scores are compared to their pre-experiment scores, demonstrating a statistically significant improvement. Moreover, statistically significant differences are observed in the 5*25m shuttle running, throwing, and set kicking between the control and experimental groups. These findings highlight the significant enhancement in strength and sensitivity achieved through functional strength training in football players. The results contribute to the development of training programs for football players and the overall improvement of training efficiency. [ABSTRACT FROM AUTHOR]

Aerial photography is a long-range, non-contact method of target detection technology that enables qualitative or quantitative analysis of the target. However, aerial photography images generally have certain chromatic aberration and color distortion. Therefore, effective segmentation of aerial images can further enhance the feature information and reduce the computational difficulty for subsequent image processing. In this paper, we propose an improved version of Golden Jackal Optimization, which is dubbed Helper Mechanism Based Golden Jackal Optimization (HGJO), to apply multilevel threshold segmentation to aerial images. The proposed method uses opposition-based learning to boost population diversity. And a new approach to calculate the prey escape energy is proposed to improve the convergence speed of the algorithm. In addition, the Cauchy distribution is introduced to adjust the original update scheme to enhance the exploration capability of the algorithm. Finally, a novel "helper mechanism" is designed to improve the performance for escape the local optima. To demonstrate the effectiveness of the proposed algorithm, we use the CEC2022 benchmark function test suite to perform comparison experiments. the HGJO is compared with the original GJO and five classical meta-heuristics. The experimental results show that HGJO is able to achieve competitive results in the benchmark test set. Finally, all of the algorithms are applied to the experiments of variable threshold segmentation of aerial images, and the results show that the aerial photography images segmented by HGJO beat the others. Noteworthy, the source code of HGJO is publicly available at https://github.com/Vang-z/HGJO. [ABSTRACT FROM AUTHOR]

Myopic Choroidal neovascularization (mCNV) is one of the most common vision-threatening com- plications of pathological myopia among many retinal diseases. Optical Coherence Tomography Angiography (OCTA) is an emerging newer non-invasive imaging technique and is recently being included in the investigation and treatment of mCNV. However, there exists no standard tool for time-efficient and dependable analysis of OCTA images of mCNV. In this study, we propose a customizable ImageJ macro that automates the OCTA image processing and lets users measure nine mCNV biomarkers. We developed a three-stage image processing pipeline to process the OCTA images using the macro. The images were first manually delineated, and then denoised using a Gaussian Filter. This was followed by the application of the Frangi filter and Local Adaptive thresholding. Finally, skeletonized images were obtained using the Mexican Hat filter. Nine vascular biomarkers including Junction Density, Vessel Diameter, and Fractal Dimension were then computed from the skeletonized images. The macro was tested on a 26 OCTA image dataset for all biomarkers. Two trends emerged in the computed biomarker values. First, the lesion-size dependent parameters (mCNV Area (mm2) Mean = 0.65, SD = 0.46) showed high variation, whereas normalized parameters (Junction Density(n/mm): Mean = 10.24, SD = 0.63) were uniform throughout the dataset. The computed values were consistent with manual measurements within existing literature. The results illustrate our ImageJ macro to be a convenient alternative for manual OCTA image processing, including provisions for batch processing and parameter customization, providing a systematic, reliable analysis of mCNV. [ABSTRACT FROM AUTHOR]

More than 50% of the images captured by optical satellites are covered by clouds, which reduces the available information in the images and seriously affects the subsequent applications of satellite images. Therefore, the identification and segmentation of cloud regions come to be one of the most important problems in current satellite image processing. Due to the complexity and variability of satellite images, especially when the ground is covered with snow, the boundary information of cloud regions is difficult to be accurately identified. The fast and accurate segmentation of cloud regions is a difficult point in the current research. We propose a lightweight convolutional neural network. Firstly, channel attention is used to optimize the effective information in the feature maps as a way to improve the network's ability to extract semantic information at each scale. Then, we fuse high and low-dimensional feature maps to enhance the network's ability to obtain small-scale semantic information. In addition, the feature aggregation module automatically adjusts the input multi-level feature weights to highlight the details of different features. Finally, we design the fully connected conditional random field to solve the problem that some noise in the input image and local minima during training is passed to the output layer resulting in the loss of edge features. Experimental results show that the proposed method achieves 0.9695 and 0.8218 for overall accuracy and recall, respectively, which has higher segmentation accuracy with the shortest time consumption compared with other state-of-the-art methods. [ABSTRACT FROM AUTHOR]

Quantitative bovine gait analysis using technology has evolved significantly over the last two decades. However, subjective methods of gait assessment using visual locomotion scoring remain the primary on-farm and experimental approach. The objective of this review is to map research trends in quantitative bovine gait analysis and to explore the technologies that have been utilized to measure biomechanical parameters of gait. A scoping literature review was conducted according to PRISMA guidelines. A search algorithm based on PICO framework generated three components–bovine, gait, and technology–to address our objectives. Three online databases were searched for original work published from January 2000 to June 2020. A two-step screening process was then conducted, starting with the review of article titles and abstracts based on inclusion criteria. A remaining 125 articles then underwent a full-text assessment, resulting in 82 final articles. Thematic analysis of research aims resulted in four major themes among the studies: gait/claw biomechanics, lameness detection, intervention/comparison, and system development. Of the 4 themes, lameness detection (55% of studies) was the most common reason for technology use. Within the literature identified three main technologies were used: force and pressure platforms (FPP), vision-based systems (VB), and accelerometers. FPP were the first and most popular technologies to evaluate bovine gait and were used in 58.5% of studies. They include force platforms, pressure mapping systems, and weight distribution platforms. The second most applied technology was VB (34.1% of studies), which predominately consists of video analysis and image processing systems. Accelerometers, another technological method to measure gait characteristics, were used in 14.6% of studies. In sum, the strong demand for automatic lameness detection influenced the path of development for quantitative gait analysis technologies. Among emergent technologies, deep learning and wearable sensors (e.g., accelerometers) appear to be the most promising options. However, although progress has been made, more research is needed to develop more accurate, practical, and user-friendly technologies. [ABSTRACT FROM AUTHOR]

Research on underwater image processing has increased significantly in the past decade due to the precious resources that exist underwater. However, it is still a challenging problem to restore degraded underwater images. Existing prior-based methods show limited performance in many cases due to their reliance on hand-crafted features. Therefore, in this paper, we propose an effective unsupervised generative adversarial network(GAN) for underwater image restoration. Specifically, we embed the idea of contrastive learning into the model. The method encourages two elements (corresponding patches) to map the similar points in the learned feature space relative to other elements (other patches) in the data set, and maximizes the mutual information between input and output through PatchNCE loss. We design a query attention (Que-Attn) module, which compares feature distances in the source domain, and gives an attention matrix and probability distribution for each row. We then select queries based on their importance measure calculated from the distribution. We also verify its generalization performance on several benchmark datasets. Experiments and comparison with the state-of-the-art methods show that our model outperforms others. [ABSTRACT FROM AUTHOR]

Despite current advancement in the field of biomedical image processing, propelled by the deep learning revolution, multimodal image registration, due to its several challenges, is still often performed manually by specialists. The recent success of image-to-image (I2I) translation in computer vision applications and its growing use in biomedical areas provide a tempting possibility of transforming the multimodal registration problem into a, potentially easier, monomodal one. We conduct an empirical study of the applicability of modern I2I translation methods for the task of rigid registration of multimodal biomedical and medical 2D and 3D images. We compare the performance of four Generative Adversarial Network (GAN)-based I2I translation methods and one contrastive representation learning method, subsequently combined with two representative monomodal registration methods, to judge the effectiveness of modality translation for multimodal image registration. We evaluate these method combinations on four publicly available multimodal (2D and 3D) datasets and compare with the performance of registration achieved by several well-known approaches acting directly on multimodal image data. Our results suggest that, although I2I translation may be helpful when the modalities to register are clearly correlated, registration of modalities which express distinctly different properties of the sample are not well handled by the I2I translation approach. The evaluated representation learning method, which aims to find abstract image-like representations of the information shared between the modalities, manages better, and so does the Mutual Information maximisation approach, acting directly on the original multimodal images. We share our complete experimental setup as open-source (https://github.com/MIDA-group/MultiRegEval), including method implementations, evaluation code, and all datasets, for further reproducing and benchmarking. [ABSTRACT FROM AUTHOR]

Otoliths (ear-stones) in the inner ears of vertebrates containing visible year zones are used extensively to determine fish age. Analysis of otoliths is a time-consuming and difficult task that requires the education of human experts. Human age estimates are inconsistent, as several readings by the same human expert might result in different ages assigned to the same otolith, in addition to an inherent bias between readers. To improve efficiency and resolve inconsistent results in the age reading from otolith images by human experts, an automated procedure based on convolutional neural networks (CNNs), a class of deep learning models suitable for image processing, is investigated. We applied CNNs that perform image regression to estimate the age of Greenland halibut (Reinhardtius hippoglossoides) with good results for individual ages as well as the overall age distribution, with an average CV of about 10% relative to the read ages by experts. In addition, the density distribution of predicted ages resembles the density distribution of the ground truth. By using k*l-fold cross-validation, we test all available samples, and we show that the results are rather sensitive to the choice of test set. Finally, we apply explanation techniques to analyze the decision process of deep learning models. In particular, we produce heatmaps indicating which input features that are the most important in the computation of predicted age. [ABSTRACT FROM AUTHOR]

Deep learning, in recent times, has made remarkable strides when it comes to impressive performance for many tasks, including medical image processing. One of the contributing factors to these advancements is the emergence of large medical image datasets. However, it is exceedingly expensive and time-consuming to construct a large and trustworthy medical dataset; hence, there has been multiple research leveraging medical reports to automatically extract labels for data. The majority of this labor, however, is performed in English. In this work, we propose a data collecting and annotation pipeline that extracts information from Vietnamese radiology reports to provide accurate labels for chest X-ray (CXR) images. This can benefit Vietnamese radiologists and clinicians by annotating data that closely match their endemic diagnosis categories which may vary from country to country. To assess the efficacy of the proposed labeling technique, we built a CXR dataset containing 9,752 studies and evaluated our pipeline using a subset of this dataset. With an F1-score of at least 0.9923, the evaluation demonstrates that our labeling tool performs precisely and consistently across all classes. After building the dataset, we train deep learning models that leverage knowledge transferred from large public CXR datasets. We employ a variety of loss functions to overcome the curse of imbalanced multi-label datasets and conduct experiments with various model architectures to select the one that delivers the best performance. Our best model (CheXpert-pretrained EfficientNet-B2) yields an F1-score of 0.6989 (95% CI 0.6740, 0.7240), AUC of 0.7912, sensitivity of 0.7064 and specificity of 0.8760 for the abnormal diagnosis in general. Finally, we demonstrate that our coarse classification (based on five specific locations of abnormalities) yields comparable results to fine classification (twelve pathologies) on the benchmark CheXpert dataset for general anomaly detection while delivering better performance in terms of the average performance of all classes. [ABSTRACT FROM AUTHOR]

This paper details the development and testing of the GIFT macro, which is a freely available program for ImageJ for the automated measurement of fiber diameters in SEM images of electrospun materials. The GIFT macro applies a validated method which distinguishes fiber diameters based on distance frequencies within an image. In this work, we introduce an applied version of the GIFT method which has been designed to be user-friendly while still allowing complete control over the various parameters involved in the image processing steps. The macro quickly processes large data sets and creates results that are reproducible and accurate. The program outputs both raw data and fiber diameter averages, so that the user can quickly assess the results and has the opportunity for further analysis if desired. The GIFT macro was compared directly to other software designed for fiber diameter measurements and was found to have comparable or lower average error, especially when measuring very small fibers, and reduced processing times per image. The macro, detailed instructions for use, and sample images are freely available online (https://github.com/IBMTRostock/GIFT). We believe that the GIFT macro is a valuable new tool for researchers looking to quickly, easily and reliably assess fiber diameters in electrospun materials. [ABSTRACT FROM AUTHOR]

Introduction: Magnetic resonance imaging (MRI) of the brain could be a key diagnostic and research tool for understanding the neuropsychiatric complications of COVID-19. For maximum impact, multi-modal MRI protocols will be needed to measure the effects of SARS-CoV-2 infection on the brain by diverse potentially pathogenic mechanisms, and with high reliability across multiple sites and scanner manufacturers. Here we describe the development of such a protocol, based upon the UK Biobank, and its validation with a travelling heads study. A multi-modal brain MRI protocol comprising sequences for T1-weighted MRI, T2-FLAIR, diffusion MRI (dMRI), resting-state functional MRI (fMRI), susceptibility-weighted imaging (swMRI), and arterial spin labelling (ASL), was defined in close approximation to prior UK Biobank (UKB) and C-MORE protocols for Siemens 3T systems. We iteratively defined a comparable set of sequences for General Electric (GE) 3T systems. To assess multi-site feasibility and between-site variability of this protocol, N = 8 healthy participants were each scanned at 4 UK sites: 3 using Siemens PRISMA scanners (Cambridge, Liverpool, Oxford) and 1 using a GE scanner (King's College London). Over 2,000 Imaging Derived Phenotypes (IDPs), measuring both data quality and regional image properties of interest, were automatically estimated by customised UKB image processing pipelines (S2 File). Components of variance and intra-class correlations (ICCs) were estimated for each IDP by linear mixed effects models and benchmarked by comparison to repeated measurements of the same IDPs from UKB participants. Intra-class correlations for many IDPs indicated good-to-excellent between-site reliability. Considering only data from the Siemens sites, between-site reliability generally matched the high levels of test-retest reliability of the same IDPs estimated in repeated, within-site, within-subject scans from UK Biobank. Inclusion of the GE site resulted in good-to-excellent reliability for many IDPs, although there were significant between-site differences in mean and scaling, and reduced ICCs, for some classes of IDP, especially T1 contrast and some dMRI-derived measures. We also identified high reliability of quantitative susceptibility mapping (QSM) IDPs derived from swMRI images, multi-network ICA-based IDPs from resting-state fMRI, and olfactory bulb structure IDPs from T1, T2-FLAIR and dMRI data. Conclusion: These results give confidence that large, multi-site MRI datasets can be collected reliably at different sites across the diverse range of MRI modalities and IDPs that could be mechanistically informative in COVID brain research. We discuss limitations of the study and strategies for further harmonisation of data collected from sites using scanners supplied by different manufacturers. These acquisition and analysis protocols are now in use for MRI assessments of post-COVID patients (N = 700) as part of the ongoing COVID-CNS study. [ABSTRACT FROM AUTHOR]

Cerebrovascular Reactivity (CVR) provides an assessment of the brain's vascular reserve and has been postulated to be a sensitive marker in cerebrovascular diseases. MRI-based CVR measurement typically employs alterations in arterial carbon dioxide (CO2) level while continuously acquiring Blood-Oxygenation-Level-Dependent (BOLD) images. CO2-inhalation and resting-state methods are two commonly used approaches for CVR MRI. However, processing of CVR MRI data often requires special expertise and may become an obstacle in broad utilization of this promising technique. The aim of this work was to develop CVR-MRICloud, a cloud-based CVR processing pipeline, to enable automated processing of CVR MRI data. The CVR-MRICloud consists of several major steps including extraction of end-tidal CO2 (EtCO2) curve from raw CO2 recording, alignment of EtCO2 curve with BOLD time course, computation of CVR value on a whole-brain, regional, and voxel-wise basis. The pipeline also includes standard BOLD image processing steps such as motion correction, registration between functional and anatomic images, and transformation of the CVR images to canonical space. This paper describes these algorithms and demonstrates the performance of the CVR-MRICloud in lifespan healthy subjects and patients with clinical conditions such as stroke, brain tumor, and Moyamoya disease. CVR-MRICloud has potential to be used as a data processing tool for a variety of basic science and clinical applications. [ABSTRACT FROM AUTHOR]

The analysis of pathological images, such as cell counting and nuclear morphological measurement, is an essential part in clinical histopathology researches. Due to the diversity of uncertain cell boundaries after staining, automated nuclei segmentation of Hematoxylin-Eosin (HE) stained pathological images remains challenging. Although better performances could be achieved than most of classic image processing methods do, manual labeling is still necessary in a majority of current machine learning based segmentation strategies, which restricts further improvements of efficiency and accuracy. Aiming at the requirements of stable and efficient high-throughput pathological image analysis, an automated Feature Global Delivery Connection Network (FGDC-net) is proposed for nuclei segmentation of HE stained images. Firstly, training sample patches and their corresponding asymmetric labels are automatically generated based on a Full Mixup strategy from RGB to HSV color space. Secondly, in order to add connections between adjacent layers and achieve the purpose of feature selection, FGDC module is designed by removing the jumping connections between codecs commonly used in UNet-based image segmentation networks, which learns the relationships between channels in each layer and pass information selectively. Finally, a dynamic training strategy based on mixed loss is used to increase the generalization capability of the model by flexible epochs. The proposed improvements were verified by the ablation experiments on multiple open databases and own clinical meningioma dataset. Experimental results on multiple datasets showed that FGDC-net could effectively improve the segmentation performances of HE stained pathological images without manual interventions, and provide valuable references for clinical pathological analysis. [ABSTRACT FROM AUTHOR]

With the exploration and development of marine resources, deep learning is more and more widely used in underwater image processing. However, the quality of the original underwater images is so low that traditional semantic segmentation methods obtain poor segmentation results, such as blurred target edges, insufficient segmentation accuracy, and poor regional boundary segmentation effects. To solve these problems, this paper proposes a semantic segmentation method for underwater images. Firstly, the image enhancement based on multi-spatial transformation is performed to improve the quality of the original images, which is not common in other advanced semantic segmentation methods. Then, the densely connected hybrid atrous convolution effectively expands the receptive field and slows down the speed of resolution reduction. Next, the cascaded atrous convolutional spatial pyramid pooling module integrates boundary features of different scales to enrich target details. Finally, the context information aggregation decoder fuses the features of the shallow network and the deep network to extract rich contextual information, which greatly reduces information loss. The proposed method was evaluated on RUIE, HabCam UID, and UIEBD. Compared with the state-of-the-art semantic segmentation algorithms, the proposed method has advantages in segmentation integrity, location accuracy, boundary clarity, and detail in subjective perception. On the objective data, the proposed method achieves the highest MIOU of 68.3 and OA of 79.4, and it has a low resource consumption. Besides, the ablation experiment also verifies the effectiveness of our method. [ABSTRACT FROM AUTHOR]

Purpose: Considering the scarcity of normal and strabismic images, this study proposed a method that combines a meta-learning approach with image processing methods to improve the classification accuracy when meta-learning alone is used for screening strabismus. Methods: The meta-learning approach was first pre-trained on a public dataset to obtain a well-generalized embedding network to extract distinctive features of images. On the other hand, the image processing methods were used to extract the position features of eye regions (e.g., iris position, corneal light reflex) as supplementary features to the distinctive features. Afterward, principal component analysis was applied to reduce the dimensionality of distinctive features for integration with low-dimensional supplementary features. The integrated features were then used to train a support vector machine classifier for performing strabismus screening. Sixty images (30 normal and 30 strabismus) were used to verify the effectiveness of the proposed method, and its classification performance was assessed by computing the accuracy, specificity, and sensitivity through 5,000 experiments. Results: The proposed method achieved a classification accuracy of 0.805 with a sensitivity (correct classification of strabismus) of 0.768 and a specificity (correct classification of normal) of 0.842, whereas the classification accuracy of using meta-learning alone was 0.709 with a sensitivity of 0.740 and a specificity of 0.678. Conclusion: The proposed strabismus screening method achieved promising classification accuracy and gained significant accuracy improvement over using meta-learning alone under data scarcity. [ABSTRACT FROM AUTHOR]

A quality detection system for the "Red Fuji" apple in Luochuan was designed for automatic grading. According to the Chinese national standard, the grading principles of apple appearance quality and Brix detection were determined. Based on machine vision and image processing, the classifier models of apple defect, contour, and size were constructed. And then, the grading thresholds were set to detect the defective pixel ratio t, aspect ratio λ, and the cross-sectional diameter Wp in the image of the apple. Spectral information of apples in the wavelength range of 400 nm~1000 nm was collected and the multiple scattering correction (MSC) and standard normal variable (SNV) transformation methods were used to preprocess spectral reflectance data. The competitive adaptive reweighted sampling (CARS) algorithm and the successive projections algorithm (SPA) were used to extract characteristic wavelength points containing Brix information, and the CARS-PLS (partial least squares) algorithm was used to establish a Brix prediction model. Apple defect, contour, size, and Brix were combined as grading indicators. The apple quality online grading detection platform was built, and apple's comprehensive grading detection algorithm and upper computer software were designed. The experiments showed that the average accuracy of apple defect, contour, and size grading detection was 96.67%, 95.00%, and 94.67% respectively, and the correlation coefficient Rp of the Brix prediction set was 0.9469. The total accuracy of apple defect, contour, size, and Brix grading was 96.67%, indicating that the detection system designed in this paper is feasible to classify "Red Fuji" apple in Luochuan. [ABSTRACT FROM AUTHOR]

A fault detection method using skeleton extraction based on orientation field consistency is proposed to improve the efficiency of fault detection, reduce the influence of transverse nonstructural factors on fault detection, and realize automatic fault extraction. In fingerprint image processing, the consistency of the orientation field reaches a maximum value when all orientations are parallel and takes a smaller value when not all orientations are parallel. The orientation field ceases to be parallel in the presence of a stratigraphic discontinuity, and the consistency of the orientation field in the corresponding region is lower than that in parallel regions. This characteristic can be exploited to extract discontinuous regions from seismic data. Then, binarization and closing operations are used to extract fault areas and increase fault continuity. Finally, a skeleton extraction method based on extracting the longitudinal center point is used to identify the fault lines. Compared with the classical ant tracking method, the proposed method requires the adjustment of fewer parameters, thus simplifying fault identification process to a certain extent. Moreover, the proposed method effectively suppresses transverse discontinuities, highlights the longitudinal fault characteristics, and strengthens fault continuity. [ABSTRACT FROM AUTHOR]

Plant leaf area (LA) is a key metric in plant monitoring programs. Machine learning methods were used in this study to estimate the LA of four plum genotypes, including three greengage genotypes (Prunus domestica [subsp. italica var. claudiana.]) and a single myrobalan plum (prunus ceracifera), using leaf length (L) and width (W) values. To develop reliable models, 5548 leaves were subjected to experiments in two different years, 2019 and 2021. Image processing technique was used to extract dimensional leaf features, which were then fed into Linear Multivariate Regression (LMR), Support Vector Regression (SVR), Artificial Neural Networks (ANN), and the Adaptive Neuro-Fuzzy Inference System (ANFIS). Model evaluation on 2019 data revealed that the LMR structure LA = 0.007+0.687 L×W was the most accurate among the various LMR structures, with R2 = 0.9955 and Root Mean Squared Error (RMSE) = 0.404. In this case, the linear kernel-based SVR yielded an R2 of 0.9955 and an RMSE of 0.4871. The ANN (R2 = 0.9969; RMSE = 0.3420) and ANFIS (R2 = 0.9971; RMSE = 0.3240) models demonstrated greater accuracy than the LMR and SVR models. Evaluating the models mentioned above on data from various genotypes in 2021 proved their applicability for estimating LA with high accuracy in subsequent years. In another research segment, LA prediction models were developed using data from 2021, and evaluations demonstrated the superior performance of ANN and ANFIS compared to LMR and SVR models. ANFIS, ANN, LMR, and SVR exhibited R2 values of 0.9971, 0.9969, 0.9950, and 0.9948, respectively. It was concluded that by combining image analysis and modeling through ANFIS, a highly accurate smart non-destructive LA measurement system could be developed. [ABSTRACT FROM AUTHOR]

The ratio of Na+ and K+ is an important determinant of the magnitude of Na+ toxicity and osmotic stress in plant cells. Traditional analytical approaches involve destructive tissue sampling and chemical analysis, where real-time observation of spatio-temporal experiments across genetic or breeding populations is unrealistic. Such an approach can also be very inaccurate and prone to erroneous biological interpretation. Analysis by Hyperspectral Imaging (HSI) is an emerging non-destructive alternative for tracking plant nutrient status in a time-course with higher accuracy and reduced cost for chemical analysis. In this study, the feasibility and predictive power of HSI-based approach for spatio-temporal tracking of Na+ and K+ levels in tissue samples was explored using a panel recombinant inbred line (RIL) of rice (Oryza sativa L.; salt-sensitive IR29 x salt-tolerant Pokkali) with differential activities of the Na+ exclusion mechanism conferred by the SalTol QTL. In this panel of RILs the spectrum of salinity tolerance was represented by FL499 (super-sensitive), FL454 (sensitive), FL478 (tolerant), and FL510 (super-tolerant). Whole-plant image processing pipeline was optimized to generate HSI spectra during salinity stress at EC = 9 dS m-1. Spectral data was used to create models for Na+ and K+ prediction by partial least squares regression (PLSR). Three datasets, i.e., mean image pixel spectra, smoothened version of mean image pixel spectra, and wavelength bands, with wide differences in intensity between control and salinity facilitated the prediction models with high R2. The smoothened and filtered datasets showed significant improvements over the mean image pixel dataset. However, model prediction was not fully consistent with the empirical data. While the outcome of modeling-based prediction showed a great potential for improving the throughput capacity for salinity stress phenotyping, additional technical refinements including tissue-specific measurements is necessary to maximize the accuracy of prediction models. [ABSTRACT FROM AUTHOR]

The bin packing problem has a wide range of applications in industry. With the upgrade of the task difficulty, the traditional 2d rectangular layout algorithm can no longer meet the needs of modern industry, such as express packing task and exoplanet ore collection task. The express or ore samples come in heterogeneous shapes so they cannot all be treated as rectangular pieces. In this paper, we propose a three-stage method called covering, corner-searching and occupying (C,S&O) to solve the two-dimensional multishape part packing problem. The objective of the packing problem variant is to ensure maximum use of the raw material and minimize the trim loss. The algorithm cannot make use of information about the sequence of future objects that are going to arrive, only knowing the shape and size of the coming one, and the coming part must be packed into the bin immediately after its arrival without buffering or readjusting. The method of C,S&O hybridizes the idea of "gold corners, silver edges and grass belly" in the Chinese game Go and the method of finding picture corners in machine vision. In the first stage, the rectangular bin and the coming part are transformed into matrix representation, and generating the position matrix that indicates possible ways of packing the part into the bin. In the second stage, the suitable layout position of the coming part is obtained using machine vision image processing technology for reference. The third stage is calculating the environment matching degree to determine the current optimal placement orientation. In order to facilitate the display of the simulation results, only three shapes of parts are considered in the simulation, rectangle, circle and triangle. The experimental results show the effectiveness of this method. Consulting the literature, it is found that this paper is the first to propose a layout method for multishape manufacturing parts. [ABSTRACT FROM AUTHOR]

Image processing algorithms are used to improve digital image representations in either their appearance or storage efficiency. The merit of these algorithms depends, in part, on visual perception by human observers. However, in practice, most are assessed numerically, and the perceptual metrics that do exist are criterion sensitive with several shortcomings. Here we propose an objective performance-based perceptual measure of image quality and demonstrate this by comparing the efficacy of a denoising algorithm for a variety of filters. For baseline, we measured detection thresholds for a white noise signal added to one of a pair of natural images in a two-alternative forced-choice (2AFC) paradigm where each image was selected randomly from a set of n = 308 on each trial. In a series of experimental conditions, the stimulus image pairs were passed through various configurations of a denoising algorithm. The differences in noise detection thresholds with and without denoising are objective perceptual measures of the ability of the algorithm to render noise invisible. This was a factor of two (6dB) in our experiment and consistent across a range of filter bandwidths and types. We also found that thresholds in all conditions converged on a common value of PSNR, offering support for this metric. We discuss how the 2AFC approach might be used for other algorithms including compression, deblurring and edge-detection. Finally, we provide a derivation for our Cartesian-separable log-Gabor filters, with polar parameters. For the biological vision community this has some advantages over the more typical (i) polar-separable variety and (ii) Cartesian-separable variety with Cartesian parameters. [ABSTRACT FROM AUTHOR]

Robust and accurate behavioral tracking is essential for ethological studies. Common methods for tracking and extracting behavior rely on user adjusted heuristics that can significantly vary across different individuals, environments, and experimental conditions. As a result, they are difficult to implement in large-scale behavioral studies with complex, heterogenous environmental conditions. Recently developed deep-learning methods for object recognition such as Faster R-CNN have advantages in their speed, accuracy, and robustness. Here, we show that Faster R-CNN can be employed for identification and detection of Caenorhabditis elegans in a variety of life stages in complex environments. We applied the algorithm to track animal speeds during development, fecundity rates and spatial distribution in reproductive adults, and behavioral decline in aging populations. By doing so, we demonstrate the flexibility, speed, and scalability of Faster R-CNN across a variety of experimental conditions, illustrating its generalized use for future large-scale behavioral studies. Author summary: Behavioral and ethological studies often rely on the ability to accurately track and process large amounts of behavioral recording in complex, heterogeneous environments and experimental conditions. Under these conditions, traditional image processing techniques may fail and prevent extraction of basic behavioral information, such as location and speed. We present an easy-to-use deep-learning-based tool to allow researchers to easily track and detect a commonly used model organism (Caenorhabditis elegans) in a variety of different environments. We examine behavior and movement throughout their lifespan, along with reproductive fecundity. We also provide a generalized, free, web-based tool for developing deep learning object detection models for any object of interest. [ABSTRACT FROM AUTHOR]

The distortion measurement plays an important role in panoramic image processing. Most measurement algorithms judge the panoramic image quality by means of weighting the quality of the local areas. However, such a calculation fails to globally reflect the quality of the panoramic image. Therefore, the multi-region adjacent pixels correlation (MRAPC) is proposed as the efficient feature for no-reference panoramic images quality assessment in this paper. Specifically, from the perspective of the statistical characteristics, the differences of the adjacent pixels in panoramic image are proved to be highly related to the degree of distortion and independent of image content. Besides, the difference map has limited pixel value range, which can improve the efficiency of quality assessment. Based on these advantages, the proposed MRAPC feature collaborates with the support vector regression to globally predict the quality of panoramic images. Extensive experimental results show that the proposed no-reference panoramic image quality assessment algorithm achieves higher evaluation performance than the existing algorithms. [ABSTRACT FROM AUTHOR]

The Euler's elastica energy regularizer has been widely used in image processing and computer vision tasks. However, finding a fast and simple solver for the term remains challenging. In this paper, we propose a new dual method to simplify the solution. Classical fast solutions transform the complex optimization problem into simpler subproblems, but introduce many parameters and split operators in the process. Hence, we propose a new dual algorithm to maintain the constraint exactly, while using only one dual parameter to transform the problem into its alternate optimization form. The proposed dual method can be easily applied to level-set-based segmentation models that contain the Euler's elastic term. Lastly, we demonstrate the performance of the proposed method on both synthetic and real images in tasks image processing tasks, i.e. denoising, inpainting, and segmentation, as well as compare to the Augmented Lagrangian method (ALM) on the aforementioned tasks. [ABSTRACT FROM AUTHOR]

This report describes how image processing harnessed to multivariate analysis techniques can be used as a bio-analytical tool for mastitis screening in cows using milk samples collected from 48 animals (32 from Jersey, 7 from Gir, and 9 from Guzerat cow breeds), totalizing a dataset of 144 sequential images was collected and analyzed. In this context, this methodology was developed based on the lactoperoxidase activity to assess mastitis using recorded images of a cuvette during a simple experiment and subsequent image treatments with an R statistics platform. The color of the sample changed from white to brown upon its exposure to reagents, which is a consequence of lactoperoxidase enzymatic reaction. Data analysis was performed to extract the channels from the RGB (Red-Green-Blue) color system, where the resulting dataset was evaluated with Principal Component Analysis (PCA), Multiple Linear Regression (MLR), and Second-Order Regression (SO). Interesting results in terms of enzymatic activity correlation (R2 = 0.96 and R2 = 0.98 by MLR and SO, respectively) and of somatic cell count (R2 = 0.97 and R2 = 0.99 by MLR and SO, respectively), important mastitis indicators, were obtained using this simple method. Additionally, potential advantages can be accessed such as quality control of the dairy chain, easier bovine mastitis prognosis, lower cost, analytical frequency, and could serve as an evaluative parameter to verify the health of the mammary gland. [ABSTRACT FROM AUTHOR]

Purpose: We compared the visibility and surgeon posture between image-processing-assisted trabeculotomy (IP-LOT) using the NGENUITY® 3D visual system and conventional microsurgery (microscope-assisted trabeculotomy; MS-LOT). Methods: IP-LOT was performed for five pig eyes. The visibility of the trabecular mesh work was evaluated on images of the trabecular mesh work and the posterior surface of the cornea (Cor) obtained under three different conditions. Images were then analyzed using ImageJ® to measure differences in luminance between the trabecular mesh work and Cor. IP-LOT was also performed for eleven human eyes, and the data were analyzed using the same approach as that used for the pig eyes. The length from the surgeon's abdomen to the operative eye (working distance) during MS-LOT and IP-LOT was measured for 12 different surgeons and compared to evaluate surgeon posture. Results: Image processing significantly increased the difference in luminance between the trabecular mesh work and Cor in both pig and human eyes (p < 0.05). Moreover, the working distance in IP-LOT was significantly shorter than that in MS-LOT (p < 0.05). Conclusion: Our findings suggest that the NGENUITY® 3D visual system provides better trabecular mesh work visibility than a normal microscope in conventional surgical methods, and it allows surgeons to operate without moving far from the operative eye. [ABSTRACT FROM AUTHOR]

Focus of this study is to design an automated image processing pipeline for handling uncontrolled acquisition conditions of images acquired in the field. The pipeline has been tested on the automated identification and count of uncapped brood cells in honeybee (Apis Mellifera) comb images to reduce the workload of beekeepers during the study of the hygienic behavior of honeybee colonies. The images used to develop and test the model were acquired by beekeepers on different days and hours in summer 2020 and under uncontrolled conditions. This resulted in images differing for background noise, illumination, color, comb tilts, scaling, and comb sizes. All the available 127 images were manually cropped to approximately include the comb area. To obtain an unbiased evaluation, the cropped images were randomly split into a training image set (50 images), which was used to develop and tune the proposed model, and a test image set (77 images), which was solely used to test the model. To reduce the effects of varied illuminations or exposures, three image enhancement algorithms were tested and compared followed by the Hough Transform, which allowed identifying individual cells to be automatically counted. All the algorithm parameters were automatically chosen on the training set by grid search. When applied to the 77 test images the model obtained a correlation of 0.819 between the automated counts and the experts' counts. To provide an assessment of our model with publicly available images acquired by a different equipment and under different acquisition conditions, we randomly extracted 100 images from a comb image dataset made available by a recent literature work. Though it has been acquired under controlled exposure, the images in this new set have varied illuminations; anyhow, our pipeline obtains a correlation between automatic and manual counts equal to 0.997. In conclusion, our tests on the automatic count of uncapped honey bee comb cells acquired in the field and on images extracted from a publicly available dataset suggest that the hereby generated pipeline successfully handles varied noise artifacts, illumination, and exposure conditions, therefore allowing to generalize our method to different acquisition settings. Results further improve when the acquisition conditions are controlled. [ABSTRACT FROM AUTHOR]

Developments in high-throughput microscopy have made it possible to collect huge amounts of cell image data that are difficult to analyse manually. Machine learning (e.g., deep learning) is often employed to automate the extraction of information from these data, such as cell counting, cell type classification and image segmentation. However, the effects of different imaging methods on the accuracy of image processing have not been examined systematically. We studied the effects of different imaging methods on the performance of machine learning-based cell type classifiers. We observed lymphoid-primed multipotential progenitor (LMPP) and pro-B cells using three imaging methods: differential interference contrast (DIC), phase contrast (Ph) and bright-field (BF). We examined the classification performance of convolutional neural networks (CNNs) with each of them and their combinations. CNNs achieved an area under the receiver operating characteristic (ROC) curve (AUC) of ~0.9, which was significantly better than when the classifier used only cell size or cell contour shape as input. However, no significant differences were found between imaging methods and focal positions. [ABSTRACT FROM AUTHOR]

Automated image processing approaches are indispensable for many biomedical experiments and help to cope with the increasing amount of microscopy image data in a fast and reproducible way. Especially state-of-the-art deep learning-based approaches most often require large amounts of annotated training data to produce accurate and generalist outputs, but they are often compromised by the general lack of those annotated data sets. In this work, we propose how conditional generative adversarial networks can be utilized to generate realistic image data for 3D fluorescence microscopy from annotation masks of 3D cellular structures. In combination with mask simulation approaches, we demonstrate the generation of fully-annotated 3D microscopy data sets that we make publicly available for training or benchmarking. An additional positional conditioning of the cellular structures enables the reconstruction of position-dependent intensity characteristics and allows to generate image data of different quality levels. A patch-wise working principle and a subsequent full-size reassemble strategy is used to generate image data of arbitrary size and different organisms. We present this as a proof-of-concept for the automated generation of fully-annotated training data sets requiring only a minimum of manual interaction to alleviate the need of manual annotations. [ABSTRACT FROM AUTHOR]

Rupture of the scapholunate interosseous ligament can cause the dissociation of scaphoid and lunate bones, resulting in impaired wrist function. Current treatments (e.g., tendon-based surgical reconstruction, screw-based fixation, fusion, or carpectomy) may restore wrist stability, but do not address regeneration of the ruptured ligament, and may result in wrist functional limitations and osteoarthritis. Recently a novel multiphasic bone-ligament-bone scaffold was proposed, which aims to reconstruct the ruptured ligament, and which can be 3D-printed using medical-grade polycaprolactone. This scaffold is composed of a central ligament-scaffold section and features a bone attachment terminal at either end. Since the ligament-scaffold is the primary load bearing structure during physiological wrist motion, its geometry, mechanical properties, and the surgical placement of the scaffold are critical for performance optimisation. This study presents a patient-specific computational biomechanical evaluation of the effect of scaffold length, and positioning of the bone attachment sites. Through segmentation and image processing of medical image data for natural wrist motion, detailed 3D geometries as well as patient-specific physiological wrist motion could be derived. This data formed the input for detailed finite element analysis, enabling computational of scaffold stress and strain distributions, which are key predictors of scaffold structural integrity. The computational analysis demonstrated that longer scaffolds present reduced peak scaffold stresses and a more homogeneous stress state compared to shorter scaffolds. Furthermore, it was found that scaffolds attached at proximal sites experience lower stresses than those attached at distal sites. However, scaffold length, rather than bone terminal location, most strongly influences peak stress. For each scaffold terminal placement configuration, a basic metric was computed indicative of bone fracture risk. This metric was the minimum distance from the bone surface to the internal scaffold bone terminal. Analysis of this minimum bone thickness data confirmed further optimisation of terminal locations is warranted. [ABSTRACT FROM AUTHOR]

The ability to automatically detect and classify populations of cells in tissue sections is paramount in a wide variety of applications ranging from developmental biology to pathology. Although deep learning algorithms are widely applied to microscopy data, they typically focus on segmentation which requires extensive training and labor-intensive annotation. Here, we utilized object detection networks (neural networks) to detect and classify targets in complex microscopy images, while simplifying data annotation. To this end, we used a RetinaNet model to classify genetically labeled neurons and glia in the brains of Mosaic Analysis with Double Markers (MADM) mice. Our initial RetinaNet-based model achieved an average precision of 0.90 across six classes of cells differentiated by MADM reporter expression and their phenotype (neuron or glia). However, we found that a single RetinaNet model often failed when encountering dense and saturated glial clusters, which show high variability in their shape and fluorophore densities compared to neurons. To overcome this, we introduced a second RetinaNet model dedicated to the detection of glia clusters. Merging the predictions of the two computational models significantly improved the automated cell counting of glial clusters. The proposed cell detection workflow will be instrumental in quantitative analysis of the spatial organization of cellular populations, which is applicable not only to preparations in neuroscience studies, but also to any tissue preparation containing labeled populations of cells. [ABSTRACT FROM AUTHOR]

In this study, the feasibility of classifying soybean frogeye leaf spot (FLS) is investigated. Leaf images and hyperspectral reflectance data of healthy and FLS diseased soybean leaves were acquired. First, image processing was used to classify FLS to create a reference for subsequent analysis of hyperspectral data. Then, dimensionality reduction methods of hyperspectral data were used to obtain the relevant information pertaining to FLS. Three single methods, namely spectral index (SI), principal component analysis (PCA), and competitive adaptive reweighted sampling (CARS), along with a PCA and SI combined method, were included. PCA was used to select the effective principal components (PCs), and evaluate SIs. Characteristic wavelengths (CWs) were selected using CARS. Finally, the full wavelengths, CWs, effective PCs, SIs, and significant SIs were divided into 14 datasets (DS1–DS14) and used as inputs to build the classification models. Models' performances were evaluated based on the classification accuracy for both the overall and individual classes. Our results suggest that the FLS comprised of five classes based on the proportion of total leaf surface covered with FLS. In the PCA and SI combination model, 5 PCs and 20 SIs with higher weight coefficient of each PC were extracted. For hyperspectral data, 20 CWs and 26 effective PCs were also selected. Out of the 14 datasets, the model input variables provided by five datasets (DS2, DS3, DS4, DS10, and DS11) were more superior than those of full wavelengths (DS1) both in support vector machine (SVM) and least squares support vector machine (LS-SVM) classifiers. The models developed using these five datasets achieved overall accuracies ranging from 91.8% to 94.5% in SVM, and 94.5% to 97.3% in LS-SVM. In addition, they improved the classification accuracies by 0.9% to 3.6% (SVM) and 0.9% to 3.7% (LS-SVM). [ABSTRACT FROM AUTHOR]