Set Of Images Research Articles

Balancing Innovation and Regulation: A Comprehensive Analysis and Neural Network Approach to AI Copyright Challenges

This article explores emerging issues surrounding artificial intelligence (AI) and copyright through a two-pronged approach. First, it provides an extensive literature review analyzing government and industry strategies for addressing AI copyright concerns and evaluates their rationality. Second, it details experiments conducted using neural networks to examine relevant information and investigate image copyright challenges, assessing mainstream large language models’ efficacy in handling copyright matters.
 The literature review explores AI copyright perspectives of the United Kingdom, China, the European Union, and the United States. It finds that countries emphasize balanced regulation and innovation (UK), ethical content creation (China), regulating high-risk applications (EU), or principles like non-discrimination and privacy (US). However, comprehensive governance frameworks are needed to navigate AI’s ethical, social, and legal intricacies.
 The experimental portion trains a convolutional neural network on a dataset of 41 infringing and non-infringing image sets to identify copyright infringement. While achieving over 80% accuracy, enhancements through expanded training data, segmentation, and multi-domain detection could improve generalization. The paper concludes with an analysis advocating copyright adaptation for AI creations, measured protections for standalone AI works, and constructive policies from interdisciplinary dialogue.

A Spherical Cap Model of Epidural Hematomas.

Background Epidural hematomas (EDHs), which have a characteristic biconvex shape, are a type of post-traumatic intracranial mass. EDHs and other types of intracranial hematomas are often diagnosed with computed tomography (CT). The volumes of EDHs are important in treatment decisions and prognosis. Their volumes are usually estimated on CT using the "ABC" method, which is based on the ellipsoid shape rather than their biconvex shape. Objective To simulate the biconvex shape, we modeled the geometry of EDHs with two spherical caps. We aim to provide simpler estimation of EDH volumes in clinical settings, and eventually recommend a threshold for surgical evacuation. Methods Applying the relationship between the sphere radius, spherical cap height, and base circle radius, we derived formulas for the shape of an EDH, relating its largest diameter and location to the other two diameters. We also estimated EDH volumes using the spherical cap volume and conventional ABC formulas and then constructed a lookup table accordingly. Results Validation of the model was performed using 14 CT image sets from previously reported patients with EDHs. Our geometric model demonstrated accurate predictions. Themodel also allows reducing the number of parameters to be measured in the ABC method from three to one, the hematoma length, showcasing its potential as a reliable tool for clinical decision-making. Based on our model, an EDH longer than 7 cm would occupy more than 30 mL of the intracranial volume. Conclusion The proposed model offers a streamlined approach to estimating EDH volumes, reducing the complexity of parameters required for clinical assessments. We recommend a length of 7 cm as a threshold for surgical evacuation of EDHs. This acceleration in decision-making is crucial for managing critically injured patients with traumatic brain injuries. Further validation across diverse patient populations will enhance the generalizability and utility of this geometric modeling approach in clinical settings.

AI models for automated segmentation of engineered polycystic kidney tubules

Autosomal dominant polycystic kidney disease (ADPKD) is a monogenic, rare disease, characterized by the formation of multiple cysts that grow out of the renal tubules. Despite intensive attempts to develop new drugs or repurpose existing ones, there is currently no definitive cure for ADPKD. This is primarily due to the complex and variable pathogenesis of the disease and the lack of models that can faithfully reproduce the human phenotype. Therefore, the development of models that allow automated detection of cysts’ growth directly on human kidney tissue is a crucial step in the search for efficient therapeutic solutions. Artificial Intelligence methods, and deep learning algorithms in particular, can provide powerful and effective solutions to such tasks, and indeed various architectures have been proposed in the literature in recent years. Here, we comparatively review state-of-the-art deep learning segmentation models, using as a testbed a set of sequential RGB immunofluorescence images from 4 in vitro experiments with 32 engineered polycystic kidney tubules. To gain a deeper understanding of the detection process, we implemented both pixel-wise and cyst-wise performance metrics to evaluate the algorithms. Overall, two models stand out as the best performing, namely UNet++ and UACANet: the latter uses a self-attention mechanism introducing some explainability aspects that can be further exploited in future developments, thus making it the most promising algorithm to build upon towards a more refined cyst-detection platform. UACANet model achieves a cyst-wise Intersection over Union of 0.83, 0.91 for Recall, and 0.92 for Precision when applied to detect large-size cysts. On all-size cysts, UACANet averages at 0.624 pixel-wise Intersection over Union. The code to reproduce all results is freely available in a public GitHub repository.

Automated recognition of the major muscle injury in athletes on X-ray CT images1.

In this research, imaging techniques such as CT and X-ray are used to locate important muscles in the shoulders and legs. Athletes who participate in sports that require running, jumping, or throwing are more likely to get injuries such as sprains, strains, tendinitis, fractures, and dislocations. One proposed automated technique has the overarching goal of enhancing recognition. This study aims to determine how to recognize the major muscles in the shoulder and leg utilizing X-ray CT images as its primary diagnostic tool. Using a shape model, discovering landmarks, and generating a form model are the steps necessary to identify injuries in key shoulder and leg muscles. The method also involves identifying injuries in significant abdominal muscles. The use of adversarial deep learning, and more specifically Deep-Injury Region Identification, can improve the ability to identify damaged muscle in X-ray and CT images. Applying the proposed diagnostic model to 150 sets of CT images, the study results show that Jaccard similarity coefficient (JSC) rate for the procedure is 0.724, the repeatability is 0.678, and the accuracy is 94.9% respectively. The study results demonstrate feasibility of using adversarial deep learning and deep-injury region identification to automatically detect severe muscle injuries in the shoulder and leg, which can enhance the identification and diagnosis of injuries in athletes, especially for those who compete in sports that include running, jumping, and throwing.

Shapelet-based orientation and defect identification method for nanostructured surface imaging

Structure-property relations are of fundamental importance for continued progress in materials research. Determining these relationships for nanomaterials introduces additional challenges, especially when nanostructure is present, either through self-assembly or nano-lithographic processes. Recent advances have been made for quantification of nanostructured surfaces, for which many robust experimental imaging methods exist. One promising approach is based on the use of shapelet functions for image analysis, which may be used as a reduced basis for surface pattern structure resulting from a broad range of phenomena (e.g. self-assembly). These shapelet-based methods enable automated quantification of nanostructured images, guided by the user/researcher, providing pixel-level information of local order without requiring detailed knowledge of order symmetries. In this work, enhancements to the existing shapelet-based response distance method are developed which enable further analysis of local order, including quantification of local orientation and identification of topological defects. The presented shapelet-based methods are applied to a representative set of images of self-assembled surfaces from experimental characterization techniques including scanning electron microscopy, atomic force microscopy, and transmission electron microscopy. These methods are shown to be complementary in implementation and, importantly, provide researchers with a robust and generalized computational approach to comprehensively quantify nanostructure order, including local orientation and boundaries within well-aligned grains.

Criteria for assessing the adequacy of image recognition methods and their verification using examples of artificial series of signals

The article discusses four criteria for assessing the adequacy of the most well-known image recognition methods. Verification of two of these criteria is carried out by empirical analysis using the example of the most well-known signal recognition methods, such as DTW, DDTW, as well as methods based on the Wavelet transform and Fourier transform. Two artificial sets of images are used as recognition objects, formed by uniformly shifting the base image both horizontally and vertically. In general, the goal of this research is to develop a new method for extracting recognition features using the example of the image of the State Emblem of the Republic of Azerbaijan. In the context of this study, a verification of a previously proposed signal recognition algorithm is carried out based on the artificial family of curves, for which the most accessible and acceptable method of displacement is established: horizontally or simultaneously horizontally and vertically.

Elicitation and aggregation of multimodal estimates improve wisdom of crowd effects on ordering tasks

We present a wisdom of crowds study where participants are asked to order a small set of images based on the number of dots they contain and then to guess the respective number of dots in each image. We test two input elicitation interfaces—one elicits the two modalities of estimates jointly and the other independently. We show that the latter interface yields higher quality estimates, even though the multimodal estimates tend to be more self-contradictory. The inputs are aggregated via optimization and voting-rule based methods to estimate the true ordering of a larger universal set of images. We demonstrate that the quality of collective estimates from the simpler yet more computationally-efficient voting methods is comparable to that achieved by the more complex optimization model. Lastly, we find that using multiple modalities of estimates from one group yields better collective estimates compared to mixing numerical estimates from one group with the ordinal estimates from a different group.

What’s in the Shadows? Formal Analysis: Art History Method to Improve Interpretation Skills for Mammography and Chest Radiographs in Resident Education

To assess the feasibility of using an art history tool of formal analysis in resident education for interpretations of mammography and chest radiographs METHODS: In a pre-post study design, residents were shown pre-selected 10 mammograms and 10 chest radiographs for a total of 20 unique anonymized patient cases. After the pretest, residents attended fourformal analysis art history lessons. The formal elements included line, light, dark, shade, proportion and balance. The post-tests were administered utilizing the same set of images given during the pre-test. After the completion of the pre- and post-tests, the participants filled out a written survey. On average, participants improved their image descriptions for a mean of 30% of the total number of mammographic images they evaluated, and similarly they improved their image diagnoses for a mean of 31% of the mammographic images. On average, participants improved their image descriptions for a mean of 37% of the total number of chest radiographs they evaluated, while improving their image diagnoses for a mean of 52% of the chest X-rays. Of the 20 participants, 14 (70%) completed the post study survey. Almost all of the respondents endorsed agreement in understanding the application of formal analysis to radiologic interpretation. Eight out of 14 (57%) participants self-reported improvement in identifying abnormalities and contriving descriptors. This pilot study shows that formal art analysis used by art historians may improve radiologic learners' ability to perceive and describe relevant radiologic abnormalities which in turn would lead to a more accurate radiologic differential diagnosis. The formal analysis process trains the eye to detect and categorize the underlying structure of images. This method provides an alternative arts intervention specifically designed to improve fundamental visual skills in radiology education.

Sensitivity Testing of Stereophotoclinometry for the OSIRIS-REx Mission. I. The Accuracy and Errors of Digital Terrain Modeling

Stereophotoclinometry (SPC) was the prime method of shape modeling for NASA’s OSIRIS-REx mission to asteroid Bennu. Here we describe the extensive testing conducted before launch to certify SPC as NASA Class B flight software, which not only validated SPC for operational use but also quantified the accuracy of this technique. We used a computer-generated digital terrain model (DTM) of a synthetic asteroid as the truth input to render simulated truth images per the planned OSIRIS-REx observing campaign. The truth images were then used as input to SPC to create testing DTMs. Imaging sets, observational parameters, and processing techniques were varied to evaluate their effects on SPC's performance and their relative importance for the quality of the resulting DTMs. We show that the errors in accuracy for SPC models are of the order of the source images’ smallest pixel sizes and that a DTM can be created at any scale, provided there is sufficient imagery at that scale. Uncertainty in the spacecraft’s flight path has minimal impact on the accuracy of SPC models. Subtraction between two DTMs (truth and simulated) is an effective approach for measuring error but has limitations. Comparing the simulated truth images with images rendered from the SPC-derived DTMs provides an excellent metric for DTM quality at smaller scales and can also be applied in flight by using real images of the target. SPC has limitations near steep slopes (e.g., the sides of boulders), leading to height errors of more than 30%. This assessment of the accuracy and sensitivity of SPC provides confidence in this technique and lessons that can be applied to future missions.

Abstract TP170: An Explainable Deep Learning Method to Quantitatively Predict Hematoma Progression After Intracerebral Hemorrhage

In patients with intracerebral hemorrhage (ICH), it is important to quickly and accurately individualize the treatment decision by quantitatively predicting the hematoma progression using medical information collected on admission. Current approaches only qualitatively predict ICH progression trends leveraging anatomical features extracted from non-contrast computed tomography (NCCT) images. However, these approaches do not consider the hemodynamics of the cerebral circulation system, cannot quantitatively predict ICH progression, and lack the needed explainability. To address these technical challenges, in this work, we developed an explainable deep learning method to quantitatively predict hematoma progression after ICH, which quantitatively predicts ICH progression leveraging (i) deep anatomical features (baseline ICH lesion volume) derived from NCCT images, and (ii) deep hemodynamic features focusing veins-of-interest derived from time-resolved CT images collected on admission. A total of 73 patients categorized into the training data set (containing 58 patients) and the test data set (containing 15 patients), with 216 sets of images, were used in the development and validation of the model. For the ICH lesion volume predicted by this model, the mean difference with respect to the reference volume is -0.96 mL, and the agreement limits are [-9.64 mL, +7.71 mL] (Figure 1). In conclusions, by simultaneously leveraging deep anatomical and hemodynamic features, our model can quantitatively predict ICH progression using multimodal CT images collected on admission and hence help individualize the treatment decision during hospitalization.

Auto-CA: Automated Cobb Angle Measurement Based on Vertebrae Detection for Assessment of Spinal Curvature Deformity.

An accurate identification and localization of vertebrae in X-ray images can assist doctors in measuring Cobb angles for treating patients with adolescent idiopathic scoliosis. It is useful for clinical decision support systems for diagnosis, surgery planning, and spinal health analysis. Currently, publicly available annotated datasets on spinal vertebrae are small, making deep-learning-based detection methods that are highly data-dependent less accurate. In this article, we propose an algorithm based on convolutional neural networks that can be trained to detect vertebrae from a small set of images. This method can display critical information on a patient's spine, display vertebrae and their labels on the thoracic and lumbar, calculate the Cobb angle, and evaluate the severity of spinal deformities. The proposed achieved an average accuracy of 0.958 and 0.962 for classifying spinal deformities (i.e., C-shaped, S-shaped type 1, and S-shaped type 2) and severity of Cobb angle (i.e., normal, mild, moderate, and severe), respectively. The Cobb angle measurement had a median difference of less than 5° from the ground-truth with SMAPE of 5.27% and an error on landmark detection of 19.73. In addition, Lenke classification is used to analyze spinal deformities as types A, B, and C, which have an average accuracy of 0.924. Physicians can use the proposed system in clinical practice by providing X-ray images via the user interface.

Deep Metric Learning on the SPD Manifold for Image Set Classification

Deep Metric Learning on the SPD Manifold for Image Set Classification

Sensitivity Testing of Stereophotoclinometry for the OSIRIS-REx Mission. II. Effective Observation Geometry for Digital Terrain Modeling

The OSIRIS-REx mission used stereophotoclinometry (SPC) to generate digital terrain models (DTMs) of its target asteroid, Bennu. Here we present a suite of preflight tests conducted to identify the observing geometry and number of images needed to create DTMs that would enable successful navigation around and to the surface of the asteroid. We demonstrate that high-quality DTMs can be generated by using only five images: four that are focused on topography, in which the spacecraft’s viewing geometry brackets the target (north, south, east, and west), and a fifth that measures the target’s albedo variation, taken from near local noon. We further show that the first 10 iterations of the SPC process can meaningfully improve DTM quality, including in the case of a suboptimal input image set, whereas after 10 iterations the DTM quality approaches an asymptotic maximum. We distill our findings into recommendations for observation planning that can be applied by other missions intending to use SPC to model the shape and terrain of their target.

Comparison between one-jaw and two-jaw designs in virtual surgery planning for patients with class III malocclusion

Orthognathic surgery is highly effective for treating maxillomandibular discrepancies in patients with class III malocclusion. However, whether one- or two-jaw surgery should be selected remains controversial. Our study aimed to evaluate quantitative differences between one-jaw and two-jaw surgical designs.In total, 100 consecutive patients with skeletal class III malocclusion who underwent orthognathic surgery with preoperative three-dimensional simulation between August 2016 and November 2021 were recruited. Based on the same final occlusal setup, a two-jaw surgery design and two types of one-jaw design were created. In total, 400 image sets, including preoperative images and three types of surgical simulation, were measured and compared.The one-jaw mandibular setback design led to improvement in most cephalometric measurements and facial symmetry. Although the one-jaw maxillary advancement design improved the ANB angle and facial convexity, it induced maxillary protrusion and reduced facial symmetry. Compared with the other designs, the two-jaw design provided significantly closer cephalometric measurements to the normative values, better symmetry, and less occlusal cant.Overall, the two-jaw design provided a quantitatively better facial appearance in terms of symmetry, proportion, and profile. Although an optimal surgical design necessitates thorough preoperative evaluation and a shared decision-making process, two-jaw surgery can be considered for improving overall facial esthetics and harmony.

Improving the perception of low-light enhanced images.

Improving images captured under low-light conditions has become an important topic in computational color imaging, as it has a wide range of applications. Most current methods are either based on handcrafted features or on end-to-end training of deep neural networks that mostly focus on minimizing some distortion metric -such as PSNR or SSIM- on a set of training images. However, the minimization of distortion metrics does not mean that the results are optimal in terms of perception (i.e. perceptual quality). As an example, the perception-distortion trade-off states that, close to the optimal results, improving distortion results in worsening perception. This means that current low-light image enhancement methods -that focus on distortion minimization- cannot be optimal in the sense of obtaining a good image in terms of perception errors. In this paper, we propose a post-processing approach in which, given the original low-light image and the result of a specific method, we are able to obtain a result that resembles as much as possible that of the original method, but, at the same time, giving an improvement in the perception of the final image. More in detail, our method follows the hypothesis that in order to minimally modify the perception of an input image, any modification should be a combination of a local change in the shading across a scene and a global change in illumination color. We demonstrate the ability of our method quantitatively using perceptual blind image metrics such as BRISQUE, NIQE, or UNIQUE, and through user preference tests.

Improving the reliability of digital cameras identification by optimizing the noise portraits comparing

The issues of optimization of existing methods for digital camera sensor identification are considered. Ways of improvement of reliability of digital cameras identification is discussed. Homogeneous images were optically recorded to form a noise portrait and test sets of amateur images for 3 cameras of various types. An optimal digital filter was selected to evaluate smoothed images for obtaining noise portraits of identified cameras. Camera identification algorithm was optimized basing on a comparison of light spatial noise portraits. Application of the optimal filter and identity criterion provides an average increase of identification reliability of more than 60 times. The results can be useful in the areas of image registration and processing, security, forensics, big data analysis, etc.

Machine Learning Refinement of In Situ Images Acquired by Low Electron Dose LC-TEM.

We have studied a machine learning (ML) technique for refining images acquired during in situ observation using liquid-cell transmission electron microscopy. Our model is constructed using a U-Net architecture and a ResNet encoder. For training our ML model, we prepared an original image dataset that contained pairs of images of samples acquired with and without a solution present. The former images were used as noisy images, and the latter images were used as corresponding ground truth images. The number of pairs of image sets was 1,204, and the image sets included images acquired at several different magnifications and electron doses. The trained model converted a noisy image into a clear image. The time necessary for the conversion was on the order of 10 ms, and we applied the model to in situ observations using the software Gatan DigitalMicrograph (DM). Even if a nanoparticle was not visible in a view window in the DM software because of the low electron dose, it was visible in a successive refined image generated by our ML model.

Dual contrastive learning based image-to-image translation of unstained skin tissue into virtually stained H&E images

Staining is a crucial step in histopathology that prepares tissue sections for microscopic examination. Hematoxylin and eosin (H&E) staining, also known as basic or routine staining, is used in 80% of histopathology slides worldwide. To enhance the histopathology workflow, recent research has focused on integrating generative artificial intelligence and deep learning models. These models have the potential to improve staining accuracy, reduce staining time, and minimize the use of hazardous chemicals, making histopathology a safer and more efficient field. In this study, we introduce a novel three-stage, dual contrastive learning-based, image-to-image generative (DCLGAN) model for virtually applying an "H&E stain" to unstained skin tissue images. The proposed model utilizes a unique learning setting comprising two pairs of generators and discriminators. By employing contrastive learning, our model maximizes the mutual information between traditional H&E-stained and virtually stained H&E patches. Our dataset consists of pairs of unstained and H&E-stained images, scanned with a brightfield microscope at 20 × magnification, providing a comprehensive set of training and testing images for evaluating the efficacy of our proposed model. Two metrics, Fréchet Inception Distance (FID) and Kernel Inception Distance (KID), were used to quantitatively evaluate virtual stained slides. Our analysis revealed that the average FID score between virtually stained and H&E-stained images (80.47) was considerably lower than that between unstained and virtually stained slides (342.01), and unstained and H&E stained (320.4) indicating a similarity virtual and H&E stains. Similarly, the mean KID score between H&E stained and virtually stained images (0.022) was significantly lower than the mean KID score between unstained and H&E stained (0.28) or unstained and virtually stained (0.31) images. In addition, a group of experienced dermatopathologists evaluated traditional and virtually stained images and demonstrated an average agreement of 78.8% and 90.2% for paired and single virtual stained image evaluations, respectively. Our study demonstrates that the proposed three-stage dual contrastive learning-based image-to-image generative model is effective in generating virtual stained images, as indicated by quantified parameters and grader evaluations. In addition, our findings suggest that GAN models have the potential to replace traditional H&E staining, which can reduce both time and environmental impact. This study highlights the promise of virtual staining as a viable alternative to traditional staining techniques in histopathology.

Parallel Processing of Image Databases for Accelerated Morphological Analysis

Nuclear forensics relies on different signatures to identify the source of nuclear material. Such signatures include crystalline structure, chemical composition, and particle morphology. One way to quantify morphology in electron microscope imagery is through image segmentation, where pixels are assigned to several partitions (or groups) that correspond to particles, grains, and other objects of interest within the image. Once pixels are assigned to segments, it is relatively straightforward to quantify other quantities of interest, such as grain size, circularity, etc. However, the range and diversity of microscope images make it difficult to obtain an accurate segmentation automatically. The accuracy of segmentation can be improved through supervised learning, but this requires many images to be manually segmented. Another way to improve the accuracy is to use interactive segmentation. Interactive segmentation requires a human to provide image-specific user input to improve performance. However, the amount of user input (effort) is generally far less than is required for supervised learning. In this paper, we investigate several parallelization strategies to automatically explore the user input parameter space of interactive segmentation algorithms across a large number of images. Specifically, we investigate four different parallelization mechanisms in a high-performance computing (HPC) environment and use the Amdahl fraction to evaluate efficiency on multiple processor cores across multiple nodes. Ultimately, the parallelization strategy that was most efficient utilized the message passing interface integrated with the segmentation and quantification code. This strategy had an Amdahl fraction of 0.985 and a performance of about 0.251 s/image. These results indicate that the parameter space of interactive segmentation algorithms can be efficiently explored using HPC. This opens the door to future work where user input is reduced and where interactive image segmentation algorithms are automatically applied to large image sets.

Deep Learning Models for the Screening of Cognitive Impairment Using Multimodal Fundus Images

ObjectiveWe aimed to develop a deep learning system capable of identifying subjects with cognitive impairment quickly and easily based on multimodal ocular images. DesignCross-sectional study SubjectsParticipants of Beijing Eye Study 2011 and patients attending Beijing Tongren Eye Center and Beijing Tongren Hospital Physical Examination Center. MethodsWe trained and validated a deep learning algorithm to assess cognitive impairment using retrospectively collected data from the Beijing Eye Study 2011. Cognitive impairment was defined as a Mini–Mental State Examination (MMSE) score <24. Based on fundus photographs and optical coherence tomography (OCT) images, we developed five models based on the following sets of images: macula-centered fundus photographs, optic disc-centered fundus photographs, fundus photographs of both fields, optical coherence tomography (OCT) images, and fundus photographs of both fields with OCT (multi-modal). The performance of the models was evaluated and compared in an external validation dataset, which was collected from patients attending Beijing Tongren Eye Center and Beijing Tongren Hospital Physical Examination Center. Main Outcome MeasuresArea under the curve (AUC). ResultsA total of 9,424 retinal photographs and 4,712 OCT images were used to develop the model. The external validation sets from each center included 1,180 fundus photographs and 590 OCT images. Model comparison revealed that the multi-modal performed best, achieving an AUC of 0.820 in the internal validation set, 0.786 in external validation set 1 and 0.784 in external validation set 2. We evaluated the performance of the multi-model in different sexes and different age groups; there were no significant differences. The heatmap analysis showed that signals around the optic disc in fundus photographs and the retina and choroid around the macular and optic disc regions in OCT images were used by the multi-modal to identify participants with cognitive impairment. ConclusionsFundus photographs and OCT can provide valuable information on cognitive function. Multi-modal models provide richer information compared to single-mode models. Deep learning algorithms based on multimodal retinal images may be capable of screening cognitive impairment. This technique has potential value for broader implementation in community-based screening or clinic settings.