Set Of Images Research Articles

That’s BAD: blind anomaly detection by implicit local feature clustering

Recent studies on visual anomaly detection (AD) of industrial objects/textures have achieved quite good performance. They consider an unsupervised setting, specifically the one-class setting, in which we assume the availability of a set of normal (i.e., anomaly-free) images for training. In this paper, we consider a more challenging scenario of unsupervised AD, in which we detect anomalies in a given set of images that might contain both normal and anomalous samples. The setting does not assume the availability of known normal data and thus is completely free from human annotation, which differs from the standard AD considered in recent studies. For clarity, we call the setting blind anomaly detection (BAD). We show that BAD can be converted into a local outlier detection problem and propose a novel method named PatchCluster that can accurately detect image- and pixel-level anomalies. Experimental results show that PatchCluster shows a promising performance without the knowledge of normal data, even comparable to the SOTA methods applied in the one-class setting needing it.

A Deep Learning Approach to Classification Pneumonia in Thorax Images

An algorithm of automatic learning was developed. That is abke to identify radiographic images with a pneumonia and not pneumonia diagnose based on a data set published in the “Kaggle” platform by the Radiological Society of North America, from which we obtained a set of specific images with their labels that were divided by 70% to train a convolutional neuronal network model consisting of two convolutional layers for the extraction of characteristics in each image, and ends with a classification stage in the training of the model, to conclude with 74% in metrics of accuracy given by test tests with 30% of the data set.

The universal law of generalization holds for naturalistic stimuli.

Shepard's universal law of generalization is a remarkable hypothesis about how intelligent organisms should perceive similarity. In its broadest form, the universal law states that the level of perceived similarity between a pair of stimuli should decay as a concave function of their distance when embedded in an appropriate psychological space. While extensively studied, evidence in support of the universal law has relied on low-dimensional stimuli and small stimulus sets that are very different from their real-world counterparts. This is largely because pairwise comparisons-as required for similarity judgments-scale quadratically in the number of stimuli. We provide strong evidence for the universal law in a naturalistic high-dimensional regime by analyzing an existing data set of 214,200 human similarity judgments and a newly collected data set of 390,819 human generalization judgments (N = 2,406 U.S. participants) across three sets of natural images. (PsycInfo Database Record (c) 2024 APA, all rights reserved).

A robust pairing method for two-pulse particle tracking velocimetry based on coherent point drift

Particle tracking velocity has reached a high level of maturity in time-resolved measurements since the introduction and development of the Shake-The-Box algorithm. The effectiveness of this approach lies, in part, in its ability to exploit the temporal coherence of particle trajectories to reject the ghost particles while increasing the density of true particles. However, certain situations may prevent time-resolved measurements. In those cases, a Two-Pulse configuration is often the only option. This raises a challenge with regard to the capacity in separating the ghost from the true particles due to the lack of long-term trajectories. This article proposes a new approach to solve this problem using the coherent point drift (CPD) method. This method identifies a spatially coherent deformation field that models the transformation between two correlated sets of points. In the context of particle tracking velocimetry, the imposed spatial coherence of this calculation is believed to act in the same way as the temporal coherence that made Shake-The-Box successful. The CPD is governed by three parameters whose optimal values have been evaluated in the present contribution. These values were found to be weakly sensitive to the characteristics of the flow under study, ensuring that this method is robust without further tuning of the parameters. The method is then compared with the Two-Pulse implementation of Shake-The-Box (2P-STB) available in Davis 10.2. For this purpose, sets of realistic images were generated at two successive times for different configurations based of synthetically generated turbulent flows. The Iterative-Particle-Reconstruction in Davis 10.2 was then used to extract the list of particles to be processed by CPD. The comparison shows a better recall with 2P-STB than CPD, especially for large time intervals between frames, but an overall better rejection of ghost particles by CPD than 2P-STB, which was the expected benefit of this method.

First deep search of tidal tails in the Milky Way globular cluster NGC 6362

I present results of the analysis of a set of images obtained in the field of the Milky Way globular cluster NGC 6362 using the Dark Energy Camera, which is mounted in the 4.0 m Victor Blanco telescope of the Cerro-Tololo Interamerican Observatory. The cluster was selected as a science case for deep high-quality photometry because of the controversial observational findings and theoretical predictions on the existence of cluster tidal tails. The collected data allowed me to build an unprecedented deep cluster field color-magnitude diagram, from which I filtered stars to produce a stellar density map, to trace the stellar density variation as a function of the position angle for different concentric annuli centered on the cluster, and to construct a cluster stellar density radial profile. I also built a stellar density map from a synthetic color-magnitude diagram generated from a model of the stellar population distribution in the Milky Way. The entire analysis approach converged toward a relatively smooth stellar density between 1 and ∼3.8 cluster Jacobi radii, with a slight difference smaller than two times the background stellar density fluctuation between the mean stellar density of the southeastern hemisphere and that of the northwestern one, with the latter being higher. Moreover, the spatial distribution of the recently claimed tidal tail stars agrees well not only with the observed composite star field distribution, but also with the region least affected by interstellar absorption. Nevertheless, I detected a low stellar density excess around the cluster’s Jacobi radius, from which I conclude that NGC 6362 presents a thin extra tidal halo.

Trypophobia, skin disease, and the visual discomfort of natural textures

In the last decade, the behavioral sciences have described the phenomenon of trypophobia, which is the discomfort felt by some individuals when viewing images containing clusters of bumps or holes. One evolutionary hypothesis for this phenomenon is that this visual discomfort represents an adaptation which helps organisms avoid skin disease and/or ectoparasites. Even though trypophobic imagery and disease imagery are both examples of visual textures, to date there has been no comparison of the visual discomfort elicited by these two specific kinds of textures within the larger context of the visual comfort elicited by natural texture imagery more generally. In the present study, we administered the Trypophobia Questionnaire (TQ) and recorded the visual comfort ratings elicited by a large set of standard natural texture images, including several trypophobic and skin disease images. In two independent samples, we found that while all observers find skin diseases uncomfortable to view, only those scoring high on the TQ rated trypophobic imagery equally uncomfortable. Comparable effects were observed using both standard ANOVA procedures as well as linear mixed effects modeling. Comparing the ratings of both high-TQ and low-TQ participants to the standard textures, we find remarkably consistent rank-order preferences, with the most unpleasant textures (as rated by both groups) exhibiting qualitative similarities to trypophobic imagery. However, we also find that low-level image statistics which have been previously shown to affect visual comfort are poor predictors of the visual comfort elicited by natural textures, including trypophobic and disease imagery. Our results suggest that a full understanding of the visual comfort elicited by natural textures, including those arising from skin disease, will ultimately depend upon a better understanding of cortical areas specialized for the perception of surface and material properties, and how these visual regions interact with emotional brain areas to evoke appropriate behavioral responses, like disgust.

“Searching for order at all levels”. Antonio Lima-de-Faria (July 4, 1921 – December 27, 2023)

“Searching for order at all levels”. Antonio Lima-de-Faria (July 4, 1921 – December 27, 2023)

Predicting Time-to-Healing from a Digital Wound Image: A Hybrid Neural Network and Decision Tree Approach Improves Performance

Despite the societal burden of chronic wounds and despite advances in image processing, automated image-based prediction of wound prognosis is not yet in routine clinical practice. While specific tissue types are known to be positive or negative prognostic indicators, image-based wound healing prediction systems that have been demonstrated to date do not (1) use information about the proportions of tissue types within the wound and (2) predict time-to-healing (most predict categorical clinical labels). In this work, we analyzed a unique dataset of time-series images of healing wounds from a controlled study in dogs, as well as human wound images that are annotated for the tissue type composition. In the context of a hybrid-learning approach (neural network segmentation and decision tree regression) for the image-based prediction of time-to-healing, we tested whether explicitly incorporating tissue type-derived features into the model would improve the accuracy for time-to-healing prediction versus not including such features. We tested four deep convolutional encoder–decoder neural network models for wound image segmentation and identified, in the context of both original wound images and an augmented wound image-set, that a SegNet-type network trained on an augmented image set has best segmentation performance. Furthermore, using three different regression algorithms, we evaluated models for predicting wound time-to-healing using features extracted from the four best-performing segmentation models. We found that XGBoost regression using features that are (i) extracted from a SegNet-type network and (ii) reduced using principal components analysis performed the best for time-to-healing prediction. We demonstrated that a neural network model can classify the regions of a wound image as one of four tissue types, and demonstrated that adding features derived from the superpixel classifier improves the performance for healing-time prediction.

Impact of cone-beam computed tomography artifacts on dose calculation accuracy for lung cancer.

To implement image-guided adaptive radiotherapy (IGART), many studies investigated dose calculations on cone-beam computed tomography (CBCT). A high HU accuracy is crucial for a high dose calculation accuracy and many imaging sites showed satisfactory results. It has been shown that the dose calculation accuracy for lung cancer lags behind. To examine why the dose calculation accuracy for lung is insufficient, the relative effects of the field-of-view (FOV), breathing motion, and scatter on dose calculation accuracy were studied. A framework was built to simulate CBCT scans for lung cancer patients by forward projecting repeat CT (rCT) scans for two scan geometries: small (SFOV) and medium FOV (MFOV). Breathing motion was modeled by applying a 4D deformation vector field to the mid-position rCT. Scatter was modeled by Monte-Carlo simulations with/without an anti-scatter grid (ASG). Simulated projections were reconstructed using filtered back-projection with/without scatter correction. In case of the SFOV, the CBCT images were patched with the planning CT scan in axial direction. The treatment plan was recalculated on the rCT and simulated CBCT. The mean Hounsfield unit (HU) difference (ΔHUmean), the structural similarity index measure (SSIM), and γ metrics were calculated for the CBCT datasets of various imaging settings. The differences in HU, SSIM and dose calculation accuracy for CBCTs with and without breathing motion were negligible (mean ΔHUmean=6.4vs. 13.7, mean SSIM=0.941vs. 0.957, mean γ (ref=MFOV)=0.75). The SFOV resulted in a lower HU (mean ΔHUmean=-9.2vs. 13.7) and SSIM (mean SSIM=0.912vs. 0.957), and therefore in dose differences compared to the MFOV (mean γ=1.22). Scatter led to considerable discrepancies in all metrics. Adding only the ASG improved the results more than only applying a scatter correction algorithm. Combining ASG and scatter correction algorithm resulted in an even higher dose calculation accuracy. Scatter and FOV are the main contributors to dose inaccuracies and motion has only a minor effect on dose calculation accuracy. Therefore, utilizing an appropriate scatter correction and FOV is important to achieve sufficient dose calculation accuracy to facilitate IGART for lung.

Segmented Acoustic Displays

For more than a century, segmented displays, such as the seven‐segment display, have been a popular and cost‐effective option. They can display various commonly used characters by controlling a few custom‐shaped binary light emitters. Herein, the acoustic equivalent of segmented displays, which uses heterogeneous sound modulators to generate a limited set of acoustic holographic images, is introduced. Designing segmented acoustic displays is more challenging than optical ones due to the complex relationship between emitted sound and generated holographic images. To address this challenge, a design methodology based on unsupervised learning techniques is proposed. The approach balances the cost of acoustic displays and the quality of the images they generate, resulting in segmented displays that outperform existing general‐purpose ones when generating a finite set of acoustic images. Using simulations and physical fabrication of metamaterial‐based acoustic displays, it is proven that the approach can create segmented acoustic displays that produce high‐quality images at a lower cost. Additionally, the methodology is applied to multifrequency acoustic displays and its scalability is assessed as the number of images increases. The cost reduction through segmentation presented herein is expected to democratize sound manipulators for acoustic displays and other applications like acoustic levitation and noise cancellation.

Adaptive convolutional sparsity with sub-band correlation in the NSCT domain for MRI image fusion

Objective. Multimodal medical image fusion (MMIF) technologies merges diverse medical images with rich information, boosting diagnostic efficiency and accuracy. Due to global optimization and single-valued nature, convolutional sparse representation (CSR) outshines the standard sparse representation (SR) in significance. By addressing the challenges of sensitivity to highly redundant dictionaries and robustness to misregistration, an adaptive convolutional sparsity scheme with measurement of the sub-band correlation in the non-subsampled contourlet transform (NSCT) domain is proposed for MMIF. Approach. The fusion scheme incorporates four main components: image decomposition into two scales, fusion of detail layers, fusion of base layers, and reconstruction of the two scales. We solved a Tikhonov regularization optimization problem with source images to obtain the base and detail layers. Then, after CSR processing, detail layers were sparsely decomposed using pre-trained dictionary filters for initial coefficient maps. NSCT domain’s sub-band correlation was used to refine fusion coefficient maps, and sparse reconstruction produced the fused detail layer. Meanwhile, base layers were fused using averaging. The final fused image was obtained via two-scale reconstruction. Main results. Experimental validation of clinical image sets revealed that the proposed fusion scheme can not only effectively eliminate the interference of partial misregistration, but also outperform the representative state-of-the-art fusion schemes in the preservation of structural and textural details according to subjective visual evaluations and objective quality evaluations. Significance. The proposed fusion scheme is competitive due to its low-redundancy dictionary, robustness to misregistration, and better fusion performance. This is achieved by training the dictionary with minimal samples through CSR to adaptively preserve overcompleteness for detail layers, and constructing fusion activity level with sub-band correlation in the NSCT domain to maintain CSR attributes. Additionally, ordering the NSCT for reverse sparse representation further enhances sub-band correlation to promote the preservation of structural and textural details.

Single-stage approach for estimating optical parameters in spectral quantitative photoacoustictomography.

In quantitative photoacoustic tomography, the optical parameters of a target, most importantly the concentrations of chromophores such as deoxygenated and oxygenated hemoglobin, are estimated from photoacoustic data measured on the boundary of the target. In this work, a numerical approximation of a forward model for spectral quantitative photoacoustic tomography is constructed by utilizing the diffusion approximation for light propagation, the acoustic wave equation for ultrasound propagation, and spectral models of optical absorption and scattering to describe the wavelength dependence of the optical parameters. The related inverse problem is approached in the framework of Bayesian inverse problems. Concentrations of four chromophores (deoxygenated and oxygenated hemoglobin, water, and lipid), two scattering parameters (reference scattering and scattering power), and the Grüneisen parameter are estimated in a single-stage from photoacoustic data. The methodology is evaluated using numerical simulations in different full-view and limited-view imaging settings. The results show that, utilizing spectral data and models, the spectral optical parameters and the Grüneisen parameter can be simultaneously estimated. Furthermore, the approach can also be utilized in limited-view imaging situations.

New Vessel Extraction Method by Using Skew Normal Distribution for MRA Images

Vascular-related diseases pose significant public health challenges and are a leading cause of mortality and disability. Understanding the complex structure of the vascular system and its processes is crucial for addressing these issues. Recent advancements in medical imaging technology have enabled the generation of high-resolution 3D images of vascular structures, leading to a diverse array of methods for vascular extraction. While previous research has often assumed a normal distribution of image data, this paper introduces a novel vessel extraction method that utilizes the skew normal distribution for more accurate probability distribution modeling. The proposed method begins with a preprocessing step to enhance vessel structures and reduce noise in Magnetic Resonance Angiography (MRA) images. The skew normal distribution, known for its ability to model skewed data, is then employed to characterize the intensity distribution of vessels. By estimating the parameters of the skew normal distribution using the Expectation-Maximization (EM) algorithm, the method effectively separates vessel pixels from the background and non-vessel regions. To extract vessels, a thresholding technique is applied based on the estimated skew normal distribution parameters. This segmentation process enables accurate vessel extraction, particularly in detecting thin vessels and enhancing the delineation of vascular edges with low contrast. Experimental evaluations on a diverse set of MRA images demonstrate the superior performance of the proposed method compared to previous approaches in terms of accuracy and computational efficiency. The presented vessel extraction method holds promise for improving the diagnosis and treatment of vascular-related diseases. By leveraging the skew normal distribution, it provides accurate and efficient vessel segmentation, contributing to the advancement of vascular imaging in the field of medical image analysis.

Semi-supervised auto-segmentation method for pelvic organ-at-risk in magnetic resonance images based on deep-learning.

In radiotherapy, magnetic resonance (MR) imaging has higher contrast for soft tissues compared to computed tomography (CT) scanning and does not emit radiation. However, manual annotation of the deep learning-based automatic organ-at-risk (OAR) delineation algorithms is expensive, making the collection of large-high-quality annotated datasets a challenge. Therefore, we proposed the low-cost semi-supervised OAR segmentation method using small pelvic MR image annotations. We trained a deep learning-based segmentation model using 116 sets of MR images from 116 patients. The bladder, femoral heads, rectum, and small intestine were selected as OAR regions. To generate the training set, we utilized a semi-supervised method and ensemble learning techniques. Additionally, we employed a post-processing algorithm to correct the self-annotation data. Both 2D and 3D auto-segmentation networks were evaluated for their performance. Furthermore, we evaluated the performance of semi-supervised method for 50 labeled data and only 10 labeled data. The Dice similarity coefficient (DSC) of the bladder, femoral heads, rectum and small intestine between segmentation results and reference masks is 0.954, 0.984, 0.908, 0.852 only using self-annotation and post-processing methods of 2D segmentation model. The DSC of corresponding OARs is 0.871, 0.975, 0.975, 0.783, 0.724 using 3D segmentation network, 0.896, 0.984, 0.890, 0.828 using 2D segmentation network and common supervised method. The outcomes of our study demonstrate that it is possible to train a multi-OAR segmentation model using small annotation samples and additional unlabeled data. To effectively annotate the dataset, ensemble learning and post-processing methods were employed. Additionally, when dealing with anisotropy and limited sample sizes, the 2D model outperformed the 3D model in terms of performance.

Evaluation of multi-camera images in different SfM-MVS based photogrammetric software and comparison of digital products in generating 3D city models

High-resolution 3D city modeling is generated using the oblique and vertical (nadir) imaging features of multi-camera systems. There are several SfM-MVS based photogrammetric software available. They enable high-resolution image processing and matching and acquire successful outcomes. The orientation of large photogrammetric blocks consisting of multiple sets of images and the accuracy of the 3D city models to be obtained vary according to the ability of the software using these approaches to solve large image matrix problems. In this study, a 3D city model was produced with vertical and oblique images obtained with UltraCam Osprey Mark 3 Premium camera in a selected application area in Kütahya city center. The main purpose of the study is to evaluate the point cloud, textured mesh model, digital elevation model, true orthophoto products and vectorial roof detail drawings via utilizing the vertical + oblique images with different SfM-MVS based photogrammetric software and to compare the results.

Technical and clinical considerations of a physical liver phantom for CT radiomics analysis.

This study identifies key characteristics to help build a physical liver computed tomography (CT) phantom for radiomics harmonization; particularly, the higher-order texture metrics. CT scans of a radiomics phantom comprising of 18 novel 3D printed inserts with varying size, shape, and material combinations were acquired on a 64-slice CT scanner (Brilliance 64, Philips Healthcare). The images were acquired at 120kV, 250mAs, CTDIvol of 16.36mGy, 2mm slice thickness, and iterative noise-reduction reconstruction (iDose, Philips Healthcare, Andover, MA). Radiomics analysis was performed using the Cancer Imaging Phenomics Toolkit (CaPTk), following automated segmentation of 3D regions of interest (ROI) of the 18 inserts. The findings were compared to three additional ROI obtained of an anthropomorphic liver phantom, a patient liver CT scan, and a water phantom, at comparable imaging settings. Percentage difference in radiomic metrics values between phantom and tissue was used to assess the biological equivalency and <10% was used to claim equivalent. The HU for all 18 ROI from the phantom ranged from -30 to 120 which is within clinically observed HU range of the liver, showing that our phantom material (T3-6B) is representative of biological CT tissue densities (liver) with >50% radiomic features having <10% difference from liver tissue. Based on the assessment of the Neighborhood Gray Tone Difference Matrix (NGTDM) metrics it is evident that the water phantom ROI show extreme values compared to the ROIs from the phantom. This result may further reinforce the difference between a structureless quantity such as water HU values and tissue HU values found in liver. The 3-D printed patterns of the constructed radiomics phantom cover a wide span of liver tissue textures seen in CT images. Using our results, texture metrics can be selectively harmonized to establish clinically relevant and reliable radiomics panels.

Identifying optical microscope images of CVD-grown two-dimensional MoS2 by convolutional neural networks and transfer learning.

In Complementary Metal-Oxide Semiconductor (CMOS) technology, scaling down has been a key strategy to improve chip performance and reduce power losses. However, challenges such as sub-threshold leakage and gate leakage, resulting from short-channel effects, contribute to an increase in distributed static power. Two-dimensional transition metal dichalcogenides (2D TMDs) emerge as potential solutions, serving as channel materials with steep sub-threshold swings and lower power consumption. However, the production and development of these 2-dimensional materials require some time-consuming tasks. In order to employ them in different fields, including chip technology, it is crucial to ensure that their production meets the required standards of quality and uniformity; in this context, deep learning techniques show significant potential. This research introduces a transfer learning-based deep convolutional neural network (CNN) to classify chemical vapor deposition (CVD) grown molybdenum disulfide (MoS2) flakes based on their uniformity or the occurrence of defects affecting electronic properties. Acquiring and labeling a sufficient number of microscope images for CNN training may not be realistic. To address this challenge, artificial images were generated using Fresnel equations to pre-train the CNN. Subsequently, accuracy was improved through fine-tuning with a limited set of real images. The proposed transfer learning-based CNN method significantly improved all measurement metrics with respect to the ordinary CNNs. The initial CNN, trained with limited data and without transfer learning, achieved 68% average accuracy for binary classification. Through transfer learning and artificial images, the same CNN achieved 85% average accuracy, demonstrating an average increase of approximately 17%. While this study specifically focuses on MoS2 structures, the same methodology can be extended to other 2-dimensional materials by simply incorporating their specific parameters when generating artificial images.

Segmentation of liver and liver lesions using deep learning.

Segmentation of organs and lesions could be employed for the express purpose of dosimetry in nuclear medicine, assisted image interpretations, and mass image processing studies. Deep leaning created liver and liver lesion segmentation on clinical 3D MRI data has not been fully addressed in previous experiments. To this end, the required data were collected from 128 patients, including their T1w and T2w MRI images, and ground truth labels of the liver and liver lesions were generated. The collection of 110 T1w-T2w MRI image sets was divided, with 94 designated for training and 16 for validation. Furthermore, 18 more datasets were separately allocated for use as hold-out test datasets. The T1w and T2w MRI images were preprocessed into a two-channel format so that they were used as inputs to the deep learning model based on the Isensee 2017 network. To calculate the final Dice coefficient of the network performance on test datasets, the binary average of T1w and T2w predicted images was used. The deep learning model could segment all 18 test cases, with an average Dice coefficient of 88% for the liver and 53% for the liver tumor. Liver segmentation was carried out with rather a high accuracy; this could be achieved for liver dosimetry during systemic or selective radiation therapies as well as for attenuation correction in PET/MRI scanners. Nevertheless, the delineation of liver lesions was not optimal; therefore, tumor detection was not practical by the proposed method on clinical data.

An ultra-lightweight method for individual identification of cow-back pattern images in an open image set

Cow recognition forms the foundation of smart livestock management. Current closed-set cow recognition models can exclusively identify cows within their training set and are ill-equipped to recognize other subjects in the open world. Additionally, there is a dearth of comprehensive research on open-set cow recognition. Existing open-set recognition models not only exhibit suboptimal accuracy but also impose a substantial burden on hardware resources due to their extensive parameterization. These challenges hinder the practical deployment of cow recognition models. This study introduces an innovative open-set metric learning-based recognition framework for cow-back patterns, aiming to rectify a significant limitation of individual cow recognition models, which is their inability to identify cows not included in the training set. The proposed framework combines various loss functions, metric methods, and backbone networks. This integration enables the open-set recognition of cow back pattern images, thereby facilitating the identification of cows previously unrecognized by the model. To mitigate hardware resource consumption, we have devised a novel ultra-lightweight backbone network known as LightCowsNet. This network draws inspiration from existing lightweight backbone networks such as MobileFaceNet, MobileViT, and EfficientNetV2. Its purpose is to extract features from cow back pattern images efficiently. LightCowsNet employs attention mechanisms, inverted residual structures, and depthwise separable convolutions. LightCowsNet first uses attention mechanisms, inverted residual structures, and depthwise separable convolutions to design multiple new feature extraction modules. These modules, which have different structures, are subsequently placed to extract different levels of image information, which enables complete extraction of the textural information of the cows’ back patterns. The present study reorganized the Cows2021 dataset to render it suitable for open-set recognition of cow-back patterns, with the test set comprising image pairs. The results revealed that using A-SoftMax as the loss function and Euclidean distance as the metric method enabled LightCowsNet to achieve the highest accuracy of 94.26 %, with a model weight space occupation of only 4.06 MB. Compared to MobileFaceNet, MobileViT, and EfficientNetV2, LightCowsNet achieved 6.24 %, 8.82 %, and 2.3 % greater accuracy on the test set, respectively. Its weight also decreased by 0.71 MB, 0.24 MB, and 4.3 MB, respectively. These results demonstrate that the proposed model achieved both accuracy improvement and model lightweightness, and these findings could serve as reference solutions for intelligent farming on cow farms.

Frequency and associated factors of accommodation and non-strabismic binocular vision dysfunction among medical university students.

To investigate the frequency and associated factors of accommodation and non-strabismic binocular vision dysfunction among medical university students. Totally 158 student volunteers underwent routine vision examination in the optometry clinic of Guangxi Medical University. Their data were used to identify the different types of accommodation and non-strabismic binocular vision dysfunction and to determine their frequency. Correlation analysis and logistic regression were used to examine the factors associated with these abnormalities. The results showed that 36.71% of the subjects had accommodation and non-strabismic binocular vision issues, with 8.86% being attributed to accommodation dysfunction and 27.85% to binocular abnormalities. Convergence insufficiency (CI) was the most common abnormality, accounting for 13.29%. Those with these abnormalities experienced higher levels of eyestrain (χ2=69.518, P<0.001). The linear correlations were observed between the difference of binocular spherical equivalent (SE) and the index of horizontal esotropia at a distance (r=0.231, P=0.004) and the asthenopia survey scale (ASS) score (r=0.346, P<0.001). Furthermore, the right eye's SE was inversely correlated with the convergence of positive and negative fusion images at close range (r=-0.321, P<0.001), the convergence of negative fusion images at close range (r=-0.294, P<0.001), the vergence facility (VF; r=-0.234, P=0.003), and the set of negative fusion images at far range (r=-0.237, P=0.003). Logistic regression analysis indicated that gender, age, and the difference in right and binocular SE did not influence the emergence of these abnormalities. Binocular vision abnormalities are more prevalent than accommodation dysfunction, with CI being the most frequent type. Greater binocular refractive disparity leads to more severe eyestrain symptoms.