Set Of Standard Images Research Articles

Using positional tracking to improve abdominal ultrasound machine learning classification

Diagnostic abdominal ultrasound screening and monitoring protocols are based around gathering a set of standard cross sectional images that ensure the coverage of relevant anatomical structures during the collection procedure. This allows clinicians to make diagnostic decisions with the best picture available from that modality. Currently, there is very little assistance provided to sonographers to ensure adherence to collection protocols, with previous studies suggesting that traditional image only machine learning classification can provide only limited assistance in supporting this task, for example it can be difficult to differentiate between multiple liver cross sections or those of the left and right kidney from image post collection. In this proof of concept, positional tracking information was added to the image input of a neural network to provide the additional context required to recognize six otherwise difficult to identify edge cases. In this paper optical and sensor based infrared tracking (IR) was used to track the position of an ultrasound probe during the collection of clinical cross sections on an abdominal phantom. Convolutional neural networks were then trained using both image-only and image with positional data, the classification accuracy results were then compared. The addition of positional information significantly improved average classification results from ∼90% for image-only to 95% for optical IR position tracking and 93% for Sensor-based IR in common abdominal cross sections. While there is further work to be done, the addition of low-cost positional tracking to machine learning ultrasound classification will allow for significantly increased accuracy for identifying important diagnostic cross sections, with the potential to not only provide validation of adherence to protocol but also could provide navigation prompts to assist in user training and in ensuring adherence in capturing cross sections in future.

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.

Optimizing skin disease diagnosis: harnessing online community data with contrastive learning and clustering techniques

Skin diseases pose significant challenges in China. Internet health forums offer a platform for millions of users to discuss skin diseases and share images for early intervention, leaving large amount of valuable dermatology images. However, data quality and annotation challenges limit the potential of these resources for developing diagnostic models. In this study, we proposed a deep-learning model that utilized unannotated dermatology images from diverse online sources. We adopted a contrastive learning approach to learn general representations from unlabeled images and fine-tuned the model on coarsely annotated images from Internet forums. Our model classified 22 common skin diseases. To improve annotation quality, we used a clustering method with a small set of standardized validation images. We tested the model on images collected by 33 experienced dermatologists from 15 tertiary hospitals and achieved a 45.05% top-1 accuracy, outperforming the published baseline model by 3%. Accuracy increased with additional validation images, reaching 49.64% with 50 images per category. Our model also demonstrated transferability to new tasks, such as detecting monkeypox, with a 61.76% top-1 accuracy using only 50 additional images in the training process. We also tested our model on benchmark datasets to show the generalization ability. Our findings highlight the potential of unannotated images from online forums for future dermatology applications and demonstrate the effectiveness of our model for early diagnosis and potential outbreak mitigation.

Impact of a short online course on the accuracy of non-ophthalmic diabetic retinopathy graders in recognising glaucomatous optic nerves in Vietnam

PurposeTo test an online training course for non-ophthalmic diabetic retinopathy (DR) graders for recognition of glaucomatous optic nerves in Vietnam.MethodsThis was an uncontrolled, experimental, before-and-after study in which 43 non-ophthalmic...

Evaluation of metric and representation learning approaches: Effects of representations driven by relative distance on the performance.

Several deep neural network architectures have emerged recently for metric learning. We asked which architecture is the most effective in measuring the similarity or dissimilarity among images. To this end, we evaluated six networks on a standard image set. We evaluated variational autoencoders, Siamese networks, triplet networks, and variational auto-encoders combined with Siamese or triplet networks. These networks were compared to a baseline network consisting of multiple separable convolutional layers. Our study revealed the following: (i) the triplet architecture proved the most effective one due to learning a relative distance - not an absolute distance; (ii) combining auto-encoders with networks that learn metrics (e.g., Siamese or triplet networks) is unwarranted; and (iii) an architecture based on separable convolutional layers is a reasonable simple alternative to triplet networks. These results can potentially impact our field by encouraging architects to develop advanced networks that take advantage of separable convolution and relative distance.

LOW-COST PORTABLE WIRELESS ELECTROENCEPHALOGRAPHY TO DETECT EMOTIONAL RESPONSES TO VISUAL CUES - VALIDATION AND POTENTIAL APPLICATIONS

This paper validates the using a low-cost EEG headset – Emotiv Insight 2.0 – for detecting emotional responses to visual stimuli. The researchers detected, based on brainwave activity, the viewer’s emotional states in reference to a series of visuals and mapped them on valance and arousal axes. Valence in this research is defined as the viewer’s positive or negative state, and arousal is defined as the intensity of the emotion or how calm or excited the viewer is. A set of thirty images – divided into two categories: Objects and Scenes – was collected from the Open Affective Standard Image Set (OASIS) and used as a reference for validation. We collected a total of 720 data points for six different emotional states: Engagement, Excitement, Focus, Interest, Relaxation, and Stress. To validate the emotional state score generated by the EEG headset, we created a regression model using those six parameters to estimate the valence and arousal level, and compare them to values reported by OASIS. The results show the significance of the Engagement parameter in predicting the valence level in the Objects category and the significance of the Excitement parameter in the Scenes category. With the emergence of personal EEG headsets, understanding the emotional reaction in different contexts will help in various fields such as urban design, digital art, and neuromarketing. In architecture, the findings can enable designers to generate more dynamic and responsive design solutions informed by users’ emotions.

A Pixel Based Method for Image Compression

The basic solution to overcome difficult issues related to huge size of digital images is to recruited image compression techniques to reduce images size for efficient storage and fast transmission. In this paper, a new scheme of pixel base technique is proposed for grayscale image compression that implicitly utilize hybrid techniques of spatial modelling base technique of minimum residual along with transformed technique of Discrete Wavelet Transform (DWT) that also impels mixed between lossless and lossy techniques to ensure highly performance in terms of compression ratio and quality. The proposed technique has been applied on a set of standard test images and the results obtained are significantly encourage compared with Joint Photographic Experts Group (JPEG).

External Validation of an Ensemble Model for Automated Mammography Interpretation by Artificial Intelligence

With a shortfall in fellowship-trained breast radiologists, mammography screening programs are looking toward artificial intelligence (AI) to increase efficiency and diagnostic accuracy. External validation studies provide an initial assessment of how promising AI algorithms perform in different practice settings. To externally validate an ensemble deep-learning model using data from a high-volume, distributed screening program of an academic health system with a diverse patient population. In this diagnostic study, an ensemble learning method, which reweights outputs of the 11 highest-performing individual AI models from the Digital Mammography Dialogue on Reverse Engineering Assessment and Methods (DREAM) Mammography Challenge, was used to predict the cancer status of an individual using a standard set of screening mammography images. This study was conducted using retrospective patient data collected between 2010 and 2020 from women aged 40 years and older who underwent a routine breast screening examination and participated in the Athena Breast Health Network at the University of California, Los Angeles (UCLA). Performance of the challenge ensemble method (CEM) and the CEM combined with radiologist assessment (CEM+R) were compared with diagnosed ductal carcinoma in situ and invasive cancers within a year of the screening examination using performance metrics, such as sensitivity, specificity, and area under the receiver operating characteristic curve (AUROC). Evaluated on 37 317 examinations from 26 817 women (mean [SD] age, 58.4 [11.5] years), individual model AUROC estimates ranged from 0.77 (95% CI, 0.75-0.79) to 0.83 (95% CI, 0.81-0.85). The CEM model achieved an AUROC of 0.85 (95% CI, 0.84-0.87) in the UCLA cohort, lower than the performance achieved in the Kaiser Permanente Washington (AUROC, 0.90) and Karolinska Institute (AUROC, 0.92) cohorts. The CEM+R model achieved a sensitivity (0.813 [95% CI, 0.781-0.843] vs 0.826 [95% CI, 0.795-0.856]; P = .20) and specificity (0.925 [95% CI, 0.916-0.934] vs 0.930 [95% CI, 0.929-0.932]; P = .18) similar to the radiologist performance. The CEM+R model had significantly lower sensitivity (0.596 [95% CI, 0.466-0.717] vs 0.850 [95% CI, 0.766-0.923]; P < .001) and specificity (0.803 [95% CI, 0.734-0.861] vs 0.945 [95% CI, 0.936-0.954]; P < .001) than the radiologist in women with a prior history of breast cancer and Hispanic women (0.894 [95% CI, 0.873-0.910] vs 0.926 [95% CI, 0.919-0.933]; P = .004). This study found that the high performance of an ensemble deep-learning model for automated screening mammography interpretation did not generalize to a more diverse screening cohort, suggesting that the model experienced underspecification. This study suggests the need for model transparency and fine-tuning of AI models for specific target populations prior to their clinical adoption.

A novel secured Euclidean space points algorithm for blind spatial image watermarking

Digital raw images obtained from the data set of various organizations require authentication, copyright protection, and security with simple processing. New Euclidean space point’s algorithm is proposed to authenticate the images by embedding binary logos in the digital images in the spatial domain. Diffie–Hellman key exchange protocol is implemented along with the Euclidean space axioms to maintain security for the proposed work. The proposed watermarking methodology is tested on the standard set of raw grayscale and RGB color images. The watermarked images are sent in the email, WhatsApp, and Facebook and analyzed. Standard watermarking attacks are also applied to the watermarked images and analyzed. The finding shows that there are no image distortions in the communication medium of email and WhatsApp. But in the Facebook platform, raw images experience compression and observed exponential noise on the digital images. The authentication and copyright protection are tested from the processed Facebook images. It is found that the embedded logo could be recovered and seen with added noise distortions. So the proposed method offers authentication and security with compression attacks. Similarly, it is found that the proposed methodology is robust to JPEG compression, image tampering attacks like collage attack, image cropping, rotation, salt-and-pepper noise, sharpening filter, semi-robust to Gaussian filtering, and image resizing, and fragile to other geometrical attacks. The receiver operating characteristics (ROC) curve is drawn and found that the area under the curve is approximately equal to unity and restoration accuracy of [67 to 100]% for various attacks.

Application of Intelligent Analysis Technology of Football Video Based on Online Target Tracking Algorithm of Motion Characteristics in Football Training.

Because the scene of football video is fixed and simple, the events in football video, such as shooting and offside , also have clear semantics. At the same time, they have sufficient domain knowledge and have broad application prospects. The research framework of sports video intelligence analysis is usually regarded as a three-level framework, namely, the low-level feature layer, the middle-level key primitive generation layer, and the high-level event analysis layer. A MF_O2T (Moving Feature Online Target Tracking) algorithm is proposed. First, based on the marked first frame image, this algorithm extracts a set of standardized local images from the target areas of visible and infrared images as target convolution filters, removes the main color of the stadium by using HSV color space nonuniform quantization algorithm, and extracts the histogram of the main color of players in upper and lower blocks. Experimental results show that the algorithm designed in this paper has strong robustness, can better adapt to player tracking in different scenes of football video, and meets the real-time requirements of football training.

Hidradenitis suppurativa: Morphologic and vascular study of nodular inflammatory lesions by means of optical coherence tomography.

Hidradenitis suppurativa (HS) is an inflammatory disease characterized by a recurrent-remission trend and clinical lesions that range from asymptomatic to inflamed, deep-seated nodules with scarring and suppuration. The aim of our study was to identify morphologic and vascular features of HS nodules by means of dynamic optical coherence tomography (D-OCT) and to define if they are correlated to patient endotype and risk of disease progression. A set of standardized clinical pictures and D-OCT images were acquired from 57 inflammatory nodules of 40 patients affected by HS. A set of 20 clinical and D-OCT images were acquired from 20 healthy volunteers as a control group. The comparison of D-OCT features among HS and control group was analysed. The correlation between HS patient endotype and D-OCT features of the lesions was calculated. D-OCT enabled to identify vascular and morphological aspects characterizing HS nodular inflammatory lesions. In addition, several D-OCT features were significantly different among distinct disease endotypes. The characterization of HS nodular inflammatory lesions through D-OCT, corresponding to blood vessel dilation and inflammatory associated hyper-vascularization, may have important clinical consequences in the assessment of HS risk of progression, therapeutic decisions and treatment efficacy monitoring.

Estimation of optimized window size for hybridized kNN-random forest algorithm based image demosaicing

Image Demosaicing is gaining popularity in the field of image processing as it helps in identifying the missing elements by using already known value of surrounding pixels obtained from color filter array overlaid on bayer pattern. The proposed work aims at calculating the image quality metrics (Signal to Noise Ratio and Peak Signal to Noise Ratio) for finding the optimized window size by performing image demosaicing based on two machine learning algorithms i.e. hybrid kNN and random forest algorithm with 20 learning cycles. To pick out the best out of many possible window sizes (3×3, 5x3, 5x5, 7x5), a set of standard images from Kodak database have been taken into consideration. The proposed work of hybrid algorithm has been implemented in MATLAB tool. The findings of the proposed work show that 7x5 window size outperforms the other window sizes for almost all the images as well as from its counterpart kNN only.

Detection and Prediction of HMS from Drinking Water by Analysing the Adsorbents from Residuals Using Deep Learning

Contamination HM is an important issue associated with the environment, and it requires suitable steps for the reduction of HMs in water at an acceptable ratio. With modern technologies, this could be possible by enabling the carbon adsorbents to adsorb the pollutions via deep learning strategies. In this paper, we develop a model on detection and prediction of presence of HMs from drinking water by analysing the adsorbents from residuals using deep learning. The study uses dense neural networks or DenseNets to analyse the microscopic images of the residual adsorbents. The study initially preprocesses and extracts features using standardised procedure. The DenseNets are used finally for detection purpose, and it is trained and tested with standard set of microscopic images. The experimental results are conducted to test the efficacy of the deep learning model on detecting the HM composition. The results of simulation show that the proposed deep learning model achieves 95% higher rate of detecting the HM composition from the adsorption residuals than other methods.

Digital mammographic interpretation by UK radiographer mammographers: A JAFROC analysis of observer performance

IntroductionRadiologists utilise mammography test sets to bench mark their performance against recognised standards. Using a validated test set, this study compares the performance of radiographer readers against previous test results for radiologists. MethodsUnder similar test conditions radiographer readers were given an established test set of 60 mammograms and tasked to identify breast cancer, they were measured against their ability to identify, locate and give a confidence level for cancer being present on a standard set of mammographic images. The results were then compared to previously published results for radiologists for similar or the same test sets. ResultsThe 10 radiographer readers demonstrated similar results to radiologists and for lesion sensitivity were the highest scoring group. The study group score a sensitivity of 83; a specificity of 69.3 and lesion sensitivity of 74.8 with ROC and JAFROC scores of 0.86 and 0.74 respectively. ConclusionUnder test conditions radiographers are able to identify and accurately locate breast cancer in a range of complex mammographic backgrounds. Implications for practiceThe study was performed under experimental conditions with results comparable to breast radiologists under similar conditions, translation of these findings into clinical practice will help address access and capacity issues in the timely identification and diagnosis of breast cancer.

Locally optimum watermark decoder in NSST domain using RSS-based Cauchy distribution

It is important to consider the tradeoff between imperceptibility and robustness requirements in developing an image-watermarking technique, some statistical model-based image watermarking schemes have been designed in the past decade. The effectiveness of a statistical watermark decoder depends highly on the modeling of the transform-domain coefficients and the use of hypothesis testing. In this study, a multiplicative image watermarking scheme is proposed in the nonsubsampled Shearlet transform (NSST) domain, where NSST coefficients are modeled by ranked set sample (RSS) based Cauchy statistical distribution and locally most powerful (LMP) test criterion is applied. Digital image watermarking technique consists of two parts, namely, embedding and extracting. In the embedding process, to achieve relatively good imperceptibility and robustness, watermark data is inserted into the significant NSST directional subband that has the highest energy value, by modifying nonlinearly the significant NSST coefficients. In the extracting phase, a scheme is proposed for designing a blind NSST domain watermark decoder incorporating the RSS based Cauchy statistical distribution and LMP test criterion. Here, the RSS method is used to estimate the location parameter and shape parameter of Cauchy statistical distribution instead of traditional maximum-likelihood (ML) method, which can provide the Cauchy model with higher precision parameters. Experimental results on a set of standard test images show significant improvements in imperceptibility and robustness using the proposed method compared with the best known state-of-the-art approaches.

Deep Learning-Based Single Image Super-Resolution: An Investigation for Dense Scene Reconstruction with UAS Photogrammetry

The deep convolutional neural network (DCNN) has recently been applied to the highly challenging and ill-posed problem of single image super-resolution (SISR), which aims to predict high-resolution (HR) images from their corresponding low-resolution (LR) images. In many remote sensing (RS) applications, spatial resolution of the aerial or satellite imagery has a great impact on the accuracy and reliability of information extracted from the images. In this study, the potential of a DCNN-based SISR model, called enhanced super-resolution generative adversarial network (ESRGAN), to predict the spatial information degraded or lost in a hyper-spatial resolution unmanned aircraft system (UAS) RGB image set is investigated. ESRGAN model is trained over a limited number of original HR (50 out of 450 total images) and virtually-generated LR UAS images by downsampling the original HR images using a bicubic kernel with a factor × 4 . Quantitative and qualitative assessments of super-resolved images using standard image quality measures (IQMs) confirm that the DCNN-based SISR approach can be successfully applied on LR UAS imagery for spatial resolution enhancement. The performance of DCNN-based SISR approach for the UAS image set closely approximates performances reported on standard SISR image sets with mean peak signal-to-noise ratio (PSNR) and structural similarity (SSIM) index values of around 28 dB and 0.85 dB, respectively. Furthermore, by exploiting the rigorous Structure-from-Motion (SfM) photogrammetry procedure, an accurate task-based IQM for evaluating the quality of the super-resolved images is carried out. Results verify that the interior and exterior imaging geometry, which are extremely important for extracting highly accurate spatial information from UAS imagery in photogrammetric applications, can be accurately retrieved from a super-resolved image set. The number of corresponding keypoints and dense points generated from the SfM photogrammetry process are about 6 and 17 times more than those extracted from the corresponding LR image set, respectively.

Spatially Adaptive Image Denoising via Enhanced Noise Detection Method for Grayscale and Color Images

Keeping in view the variety of the applications, image denoising still remains the unexplored territory for the researchers. There are many pros and cons in existing denoising algorithms. The two prime cons of image denoising algorithms are (i) Over and under detection of noisy pixels (ii) Low performance at high noise levels. So, in order to overcome these existing issues, a spatially adaptive image denoising via enhanced noise detection method (SAID-END) is proposed for grayscale and color images. The denoising is achieved using a two-stage sequential algorithm, the first stage ensures accurate noise estimation by eliminating over and under detection of noisy pixels. The second stage performs image restoration by considering non-noisy pixels in estimation of the original pixel value. To enhance the accuracy while denoising high-density impulse noise and artifacts, both noise estimation and restoration stages are using a spatially adaptive window (window expands to spatially connected area), the size of the window depends upon the noise level in the vicinity of the reference noisy pixel. The two stages of the proposed method are referred to as (i) Enhanced adaptive noise detection (ii) Non-corrupted pixel sensitive adaptive image restoration. The proposed method is evaluated by two test steps to ensure its versatility and robustness. In the first step, the proposed method is tested on a wide standard data set of color and grayscale images affected by impulse noise and artifacts. The results of proposed method are compared with well-known methods compatible for denoising impulse noise and artifacts. In the second step, the results of proposed method are compared with the recent state of the art algorithms for traditional test images. The result shows that the proposed method outperforms the existing denoising methods when applied to grayscale and color images.

Blind Image Watermark Decoder in UDTCWT Domain Using Weibull Mixtures-Based Vector HMT

As one promising solution, digital watermarking has been proposed to resolve image copyright protection and content authentication, and has been applied successfully in many fields. Owing to their excellent description capability and invariance property, statistical models have become a popular tool for the image watermarking resulting in favorable trade-offs among imperceptibility, robustness and data payload. By modeling the robust undecimated dual tree complex Wavelet transform (UDTCWT) coefficient magnitudes with the Weibull mixtures based vector hidden Markov trees (HMT) and employing maximum likelihood (ML) test criterion, we propose a new image watermarking approach in UDTCWT domain in this paper. Our image watermarking approach consists of two parts, namely, embedding and extracting. In the embedding process, we compute the robust UDTCWT coefficient magnitudes with UDTCWT domain real/imaginary parts, and insert the digital watermark into the significant UDTCWT coefficient magnitude subband. In the extracting phase, robust UDTCWT coefficient magnitudes are firstly modeled by employing the Weibull mixture-based vector HMT, where the statistical properties of UDTCWT magnitudes are captured accurately. Then the expectation/conditional maximization (ECM) approach is introduced to estimate the statistical model parameters. Finally, an image watermark decoder for multiplicative watermarking is developed using the Weibull mixtures based vector HMT and ML test. The experiments show that the proposed method not only improves the imperceptibility, but also increases the robustness performance and outperforms state-of-the-art methods on a set of standard test images.

Spectral Graph Based Vertex-Frequency Wiener Filtering for Image and Graph Signal Denoising

In this article, we propose and develop a spectral graph based vertex varying Wiener filtering framework in the joint vertex-frequency domain for denoising of graph signals defined on weighted, undirected and connected graphs. To this end, we first extend the Zadeh time-frequency filter concept to graph signals and obtain vertex-frequency transfer function of the proposed Wiener filter by transforming its vertex varying impulse response that minimizes the mean square error between original and recovered signals. To facilitate the derived Wiener filter, we present a detailed derivation of a recently proposed graph Rihaczek vertex-frequency signal distribution (GRD) so as to match the structure of the proposed graph Zadeh filter, based on a graph translation operator defined by generalized convolution with a delta signal. We express the filter transfer function in terms of this graph transform of the original, noiseless signal. The form of the obtained Wiener filter is, interestingly, different than those of time-frequency Wiener filters prevalent in the classical signal processing. We also investigate the invertibility of the employed GRD. Assuming that the original graph signal of interest can be viewed as deterministic, we propose two algorithms to implement the proposed vertex-frequency Wiener filter from a single realization of a noisy, input signal. We derive mean and variance of the GRD of the noisy signal, since they are required in one of these algorithms. We apply proposed Wiener filter algorithms to denoise a standard set of images and three irregularly structured graph signals, and demonstrate their competitiveness with compared high performance denoising methods.

Image compression using explored bat algorithm by Renyi 2-d histogram based on multilevel thresholding

The main objective of the image compression is to extract meaningful clusters from a given image. A meaningful cluster is possible with perfect threshold values, which are optimized by assuming Renyi entropy as an objective function. Due to the equal distribution of energy over the entire 1-D histogram, it is computationally complex. In order to improve the visual quality of a reconstructed image, a 2-D histogram based multilevel thresholding is proposed to maximize the Renyi entropy using explored bat algorithm. Thus procured results are compared with other optimization techniques and these are incorporated. It is the first time, incorporating a weighted peak signal to noise ratio (WPSNR) and the visual PSNR (VPSNR) in the proposed method, because of the failure in measuring the visual quality of peak signal to noise ratio (PSNR). Experimental results are examined on a standard set of images, which are observed precisely and efficiently in the multilevel thresholding problem.