White Images Research Articles

Superficial white matter microstructural imaging method based on time-space fractional-order diffusion

Objective. Microstructure imaging based on diffusion magnetic resonance signal is an advanced imaging technique that enables in vivo mapping of the brain’s microstructure. Superficial white matter (SWM) plays an important role in brain development, maturation, and aging, while fewer microstructure imaging methods address the SWM due to its complexity. Therefore, this study aims to develop a diffusion propagation model to investigate the microstructural characteristics of the SWM region. Approach. In this paper, we hypothesize that the effect of cell membrane permeability and the water exchange between soma and dendrites cannot be neglected for typical clinical diffusion times (20 ms < t < 80 ms). We then use SpinDoctor to simulate the diffusion magnetic resonance signals of real neurons and propose a time-space fractional-order diffusion model for SWM microstructure imaging. We evaluate the validity regime of our model using numerical simulations and compare the model parameters with several state-of-the-art methods. Main results. By analyzing the simulation signals of real neuronal cells as well as diffusion magnetic resonance data from the brains of fourteen healthy human subjects, we find that the time-space fractional-order diffusion model can be used to capture the structural complexity of the tissue, indirectly through the association of time fractional exponents with restricted diffusion and space fractional exponents with perfusion and membrane permeability. Significance. The results show that the diffusion propagation model can provide new insights into the tissue architecture of the SWM.

Multi-view Image Fusion Using Ensemble Deep Learning Algorithm For MRI And CT Images

Medical image fusions are crucial elements in image-based health care diagnostics or therapies and generic applications of computer visions. However, the majority of existing methods suffer from noise distortion that affects the overall output. When pictures are distorted by noises, classical fusion techniques perform badly. Hence, fusion techniques that properly maintain information comprehensively from multiple faulty pictures need to be created. This work presents Enhanced Lion Swarm Optimization (ESLO) with Ensemble Deep Learning (EDL) to address the aforementioned issues. The primary steps in this study include image fusions, segmentation, noise reduction, feature extraction, picture classification, and feature selection. Adaptive Median Filters are first used for noise removal in sequence to enhance image quality by eliminating noises. The MRIs and CT images are then segmented using the Region Growing–based k -Means Clustering (RKMC) algorithm to separate the images into their component regions or objects. Images in black and white are divided into image. In the white image, the RKMC algorithm successfully considered the earlier tumour probability. The next step is feature extraction, which is accomplished by using the Modified Principal Component Analysis (MPCA) to draw out the most informative aspects of the images. Then the ELSO algorithm is applied for optimal feature selection, which is computed by best fitness values. After that, multi-view image fusions of multi modal images derive lower-, middle-, and higher-level image contents. It is done by using Deep Convolution Neural Network (DCNN) and the Tissue-Aware Conditional Generative Adversarial Network (TAcGAN) algorithm, which fuses the multi-view features and relevant image features, and it is used for real-time applications. ELSO +EDL algorithm gives better results in terms of accuracy, Peak Signal-To-Noise Ratio (PSNR), and lower Root Mean Square Error (RMSE) and Mean Absolute Percentage Error (MAPE) when compared to other existing algorithms.

Label-Free Optical Technologies to Enhance Noninvasive Endoscopic Imaging of Early-Stage Cancers.

White light endoscopic imaging allows for the examination of internal human organs and is essential in the detection and treatment of early-stage cancers. To facilitate diagnosis of precancerous changes and early-stage cancers, label-free optical technologies that provide enhanced malignancy-specific contrast and depth information have been extensively researched. The rapid development of technology in the past two decades has enabled integration of these optical technologies into clinical endoscopy. In recent years, the significant advantages of using these adjunct optical devices have been shown, suggesting readiness for clinical translation. In this review, we provide an overview of the working principles and miniaturization considerations and summarize the clinical and preclinical demonstrations of several such techniques for early-stage cancer detection. We also offer an outlook for the integration of multiple technologies and the use of computer-aided diagnosis in clinical endoscopy.

Artificial Intelligence Versus Human Focus Group Rating of Facial Attractiveness.

Background: Many open-access artificial intelligence (AI)-based websites that rate facial attractiveness are available, but none have been compared with human focus group outcomes. Objective: To compare human and AI-based websites scoring of facial attractiveness of adult female white faces. Methods: A 40-photograph database of AI-generated adult, white, female, expressionless, and frontal-view facial images were scored by otolaryngology residents and five AI-based facial rating websites: prettyscale.com, attractivenesstest.com, face-score.com/en, hotchat3000.com, and beautyscoretest.com. Sample t-test and bivariate correlation were performed for statistical analyses. Results: The focus group of 24 otolaryngology residents consisted of 62.5% males and 58.3% white participants. There was a strong positive correlation between average human score and average AI score for each photo (Pearson's correlation 0.84, p < 0.01). The average human raters' scores were significantly lower than the average AI scores (5.0 ± 1.8 vs. 6.9 ± 0.9, p < 0.01). Thirty images (75.0%) had statistically higher scores from the AI websites versus the focus group. On correlation analysis, all AI-based websites individually had scores that positively correlate with the human scores (all p < 0.05). Conclusion: AI-based websites and human focus-group scoring of facial attractiveness of adult white female faces were significantly correlated with the AI ratings biased toward higher values, encouraging their cautious utilization in future research.

Effectiveness and visual performance assessment of anti-peeping films

Some private information inevitably becomes more visible when using wide-angle electronic devices in public places. Thus, some people use anti-peeping films to achieve privacy protection. To examine the effectiveness and its influence on the visual performance of the anti-peeping film, we investigated the color gamut, luminance, contrast ratio, Bhattacharyya coefficient and structural similarity characteristics of a smartphone without-film screen and with two-sided (left and right) of 25°, 30°, and 45° anti-peeping films, respectively, and four-sided (up, down, left, and right) anti-peeping films. The results show that the film reduces the screen luminance and contrast ratio except for the color gamut. At 0° viewing angle, the luminance decreases by 24.8%, 25.2%, 18.6%, and 53.8%, respectively. For the 25°, 30°, 45°, and 360° anti-peeping films, the luminance and contrast ratio decrease to 50% at 24°, 25°, 19°, and 20° horizontal viewing angles. The effectiveness of four-sided anti-peeping film is the best, followed by the two-sided ones, and the smaller the viewing angle, the better the effect. The Bhattacharyya coefficients of the white images’ pixel micrographs are 0.887, 0.896, 0.898, and 0.878, respectively, of which the smallest four-sided anti-peeping film has the greatest influence on screen clarity. Under the 0-40° viewing angle, the structural similarity continuously decreases, and the screen becomes blurred due to the anti-peeping film. This work investigates the optical characteristics of anti-peeping film and examines its influence on visual performance, and provides strategy for manufacturers and use guidance for users.

Dual-modality rigid endoscope for photoacoustic imaging and white light videoscopy.

There has been significant interest in the development of miniature photoacoustic imaging probes for a variety of clinical uses, including the in situ assessment of tumors and minimally invasive surgical guidance. Most of the previously implemented probes are either side viewing or operate in the optical-resolution microscopy mode in which the imaging depth is limited to . We describe a forward-viewing photoacoustic probe that operates in tomography mode and simultaneously provides white light video images. We aim to develop a dual-modality endoscope capable of performing high-resolution PAT imaging and traditional white light videoscopy simultaneously in the forward-viewing configuration. We used a Fabry-Pérot ultrasound sensor that operates in the 1500 to 1600nm wavelength range and is transparent in the visible and near infrared region (580 to 1250nm). The FP sensor was optically scanned using a miniature MEMs mirror located at the proximal end of the endoscope, resulting in a system that is sufficiently compact (10mm outer diameter) and lightweight for practical endoscopic use. The imaging performance of the endoscope is evaluated, and dual-mode imaging capability is demonstrated using phantoms and abdominal organs of an ex vivo mouse including spleen, liver, and kidney. The proposed endoscope design offers several advantages including the high acoustic sensitivity and wide detection bandwidth of the FP sensor, dual-mode imaging capability, compact footprint, and an all-optical distal end for improved safety. The dual-mode imaging capability also offers the advantage of correlating the tissue surface morphology with the underlying vascular anatomy. Potential applications include the guidance of laparoscopic surgery and other interventional procedures.

Longitudinal changes in diffusion tensor imaging in hemodialysis patients.

Hemodialysis patients have increased white matter and gray matter pathology in the brain relative to controls based on MRI. Diffusion tensor imaging is useful in detecting differences between hemodialysis and controls but has not identified the expected longitudinal decline in hemodialysis patients. In this study we implemented specialized post-processing techniques to reduce noise to detect longitudinal changes in diffusion tensor imaging parameters and evaluated for any association with changes in cognition. We collected anatomical and diffusion MRIs as well as cognitive testing from in-center hemodialysis patients at baseline and 1 year later. Gray matter thickness, white matter volume, and white matter diffusion tensor imaging parameters were measured to identify longitudinal changes. We analyzed the diffusion tensor imaging parameters by averaging the whole white matter and using a pothole analysis. Eighteen hemodialysis patients were included in the longitudinal analysis and 15 controls were used for the pothole analysis. We used the NIH Toolbox Cognition Battery to assess cognitive performance over the same time frame. Over the course of a year on hemodialysis, we found a decrease in white matter fractional anisotropy across the entire white matter (p < 0.01), and an increase in the number of white matter fractional anisotropy voxels below pothole threshold (p = 0.03). We did not find any relationship between changes in whole brain structural parameters and cognitive performance. By employing noise reducing techniques, we were able to detect longitudinal changes in diffusion tensor imaging parameters in hemodialysis patients. The fractional anisotropy declines over the year indicate significant decreases in white matter health. However, we did not find that declines in fractional anisotropy was associated with declines in cognitive performance.

Optical Asymmetric Cryptosystem Based on Dynamic Foveated Imaging and Bidimensional Empirical Mode Decomposition

In this paper, we propose an asymmetric cryptosystem based on dynamic foveated imaging and bidimensional empirical mode decomposition (BEMD). Firstly, a novel dynamic foveated imaging algorithm is developed to transform a plaintext image to a globally ambiguous and locally clear image. Then, the image is passed through a phase-truncated Fourier transform system to generate a white noise image. The resulting image is encoded using BEMD to produce an encrypted image. The proposed cryptosystem offers two distinct decryption methods, allowing the receiver to obtain a decrypted image from a specific frame or a combination of frames, depending on the unique keys. This encryption scheme significantly expands the key space and strengthens the system’s anti-iterative attack capability. Numerical simulation results demonstrate the effectiveness, security and robustness of the proposed cryptosystem.

Barriers for Veterans Accessing Outdoor Programming: An Examination of Website Representation

Background: Existing outdoor programs which market to veterans are likely to help veterans to get outdoors and to connect with one another. However, these programs appear to market to mostly White, male populations even though many veterans are female and racially diverse. Purpose: In order to consider barriers to outdoor program engagement for diverse and marginalized populations of veterans, the present study evaluated the website landing pages for outdoor programs serving veterans. This study explored the question: what racial and gender identities appear on the online marketing for veterans-serving outdoor programs? Methodology/Approach: In order to answer the research question, four coders independently coded visual data from imagery found on publicly-available websites and each organization's associated social media. Three hundred and six veteran-serving organizations providing outdoor programming were identified and coded. Findings/Conclusions: Image coding revealed that website landing pages for these programs presented a majority of White and male images, with a minority of images of women and racial minorities. Implications: The present study suggests that veteran-serving outdoor programs should consider how they present themselves in terms of their representation of diverse and marginalized identities in their marketing. Unless such program change their online materials, female veterans and veterans from racially diverse backgrounds and marginalized communities may continue to face significant barriers in accessing outdoor programming.

Aberrant interhemispheric structural and functional connectivity within whole brain in schizophrenia

ObjectiveSchizophrenia is a serious mental disorder whose etiology remains unclear. Although numerous studies have analyzed the abnormal gray matter functional activity and whole-brain anatomical changes in schizophrenia, fMRI signal fluctuations from white matter have usually been ignored and rarely reported in the literature. MethodsWe employed 45 schizophrenia subjects and 75 healthy controls (HCs) from a publicly available fMRI dataset. By combining the voxel-mirrored homotopic connectivity (VMHC) measure and fiber tracking method, we investigated the interhemispheric functional and structural connectivity within whole brain in schizophrenia. ResultsCompared to HCs, patients with schizophrenia exhibited significantly reduced VMHC in the bilateral middle occipital gyrus, precentral gyrus, postcentral gyrus and corpus callosum. Fiber tracking results showed the changes in structural connectivity for the bilateral precentral gyrus, and the bilateral corpus callosum, and the fiber bundles connecting bilateral precentral gyrus and connecting the bilateral corpus callosum passed through the posterior midbody, isthmus and splenium of mid-sagittal corpus callosum, which closely related to the interhemispheric integration of visual and auditory information. More importantly, we observed a negative correlation between averaged VMHC values in the postcentral gyrus and SAPS scores, and a positive correlation between the fractional anisotropy of fiber bundle connecting the bilateral precentral gyrus and Matrix Reasoning scores in schizophrenia. ConclusionOur findings provide a novel perspective of white matter functional images on understanding abnormal interhemispheric visual and auditory information transfer in schizophrenia.

Who said it? Native and non-native listeners' source memory for object-speaker associations

Memory for who said what can help native and non-native listeners identify critical information for use in conversations. In two experiments, source memory for object-speaker associations was tested for listeners varying in language proficiency. Native and non-native participants of English first heard speakers name objects with prototypical color (e.g., broccoli) before they had to indicate who had previously named each object. Speakers either consistently named objects of one color, or they randomly named objects of different colors. While both listener groups associated objects with speakers successfully, in Experiment 1, with black and white images, native listeners showed an advantage in source memory in the color consistent condition over the random condition, while non-native listeners had no such advantage. The finding for non-native listeners was replicated in Experiment 2 with color images. The results confirm the role of source memory representations in communication, but also suggest that encoding object-speaker association in L2 may be less sensitive to consistent patterns in the input.

Investigation of laser-induced contamination on dielectric thin films in MHz sub-ps regime

High-repetition rate diode-pumped sub-ps lasers are widely used in the industrial sector for high-quality material processing applications. However, for their reliable operation, it is crucial to study the power handling capabilities of the optical components used in these systems. The optical components, such as mirrors, gratings, dichroic filters, and gain media, are designed based on dielectric thin films. When subjected to high-intensity laser radiation, the phenomenon of laser-induced contamination (LIC) can lead to the growth of a nanometric, highly absorbent layer on an irradiated optical surface, which can result in transmission or reflection loss and eventual permanent damage. In this study, we investigate LIC growth on dielectric oxide thin films in an air environment irradiated by MHz sub-ps laser at 515 nm. We examine the effect of thin film deposition method, material, and thickness on LIC growth dynamics. The irradiated spots on the surface are inspected using multiple observation methods, including white light interference microscopy and fluorescence imaging. Our results show that the LIC growth dynamics depend on the laser intensity and irradiation time and can be affected by the thin film deposition method, material, and thickness. These findings could be used to inform the development of more resistant optical components, ensuring long-term reliable laser operation required for industrial applications. The study highlights the need for validating optical components using tests that closely mimic real-world applications and provides insight into the complex processes that lead to LIC.

Reimagining Color: Deep Learning-Based Black and White Image and Video Colorization

Reimagining Color: Deep Learning-Based Black and White Image and Video Colorization

Colorization of Black and White Images Using a Hybrid Deep Learning Framework

With the development of deep learning algorithms and their great success in the field of computer vision, the field of automatic image colorization has witnessed significant improvements in accuracy and realism. This study introduces a novel deep learning-based method for colorizing black and white photographs, utilizing the powerful feature extraction of the InceptionResNetV2 model and the generative capabilities of autoencoders. A custom data generator was developed for efficient preprocessing, data augmentation, and batch processing, enhancing memory usage and scalability. The system encodes grayscale images and extracts high-level features, which are then fused and decoded into two color channels, combined with the original luminance to recreate the image in the LAB color space. The method demonstrates strong performance with a PSNR of 22.8154 and a SSIM of 0.9097, showcasing its potential for applications like historical image restoration and media enhancement.

DCT-based White Blood Cell Image Enhancement for Recognition Using Deep Learning

DCT-based White Blood Cell Image Enhancement for Recognition Using Deep Learning

Automatic recognition of white blood cell images with memory efficient superpixel metric GNN: SMGNN.

An automatic recognizing system of white blood cells can assist hematologists in the diagnosis of many diseases, where accuracy and efficiency are paramount for computer-based systems. In this paper, we presented a new image processing system to recognize the five types of white blood cells in peripheral blood with marked improvement in efficiency when juxtaposed against mainstream methods. The prevailing deep learning segmentation solutions often utilize millions of parameters to extract high-level image features and neglect the incorporation of prior domain knowledge, which consequently consumes substantial computational resources and increases the risk of overfitting, especially when limited medical image samples are available for training. To address these challenges, we proposed a novel memory-efficient strategy that exploits graph structures derived from the images. Specifically, we introduced a lightweight superpixel-based graph neural network (GNN) and broke new ground by introducing superpixel metric learning to segment nucleus and cytoplasm. Remarkably, our proposed segmentation model superpixel metric graph neural network (SMGNN) achieved state of the art segmentation performance while utilizing at most 10000$ \times $ less than the parameters compared to existing approaches. The subsequent segmentation-based cell type classification processes showed satisfactory results that such automatic recognizing algorithms are accurate and efficient to execeute in hematological laboratories. Our code is publicly available at https://github.com/jyh6681/SPXL-GNN.

Artificial intelligence efficiently predicts gastric lesions, Helicobacter pylori infection and lymph node metastasis upon endoscopic images.

Medical images have been increased rapidly in digital medicine era, presenting an opportunity for the intervention of artificial intelligence (AI). In order to explore the value of convolutional neural network (CNN) algorithms in endoscopic images, we developed an AI-assisted comprehensive analysis system for endoscopic images and explored its performance in clinical real scenarios. A total of 6,270 white light endoscopic images from 516 cases were used to train 14 different CNN models. The images were divided into training set, validation set and test set according to 7:1:2 for exploring the possibility of discrimination of gastric cancer (GC) and benign lesions (nGC), gastric ulcer (GU) and ulcerated cancer (UCa), early gastric cancer (EGC) and nGC, infection of Helicobacter pylori (Hp) and no infection of Hp (noHp), as well as metastasis and no-metastasis at perigastric lymph nodes. Among the 14 CNN models, EfficientNetB7 revealed the best performance on two-category of GC and nGC [accuracy: 96.40% and area under the curve (AUC)=0.9959], GU and UCa (accuracy: 90.84% and AUC=0.8155), EGC and nGC (accuracy: 97.88% and AUC=0.9943), and Hp and noHp (accuracy: 83.33% and AUC=0.9096). Whereas, InceptionV3 model showed better performance on predicting metastasis and no-metastasis of perigastric lymph nodes for EGC (accuracy: 79.44% and AUC=0.7181). In addition, the integrated analysis of endoscopic images and gross images of gastrectomy specimens was performed on 95 cases by EfficientNetB7 and RFB-SSD object detection model, resulting in 100% of predictive accuracy in EGC. Taken together, this study integrated image sources from endoscopic examination and gastrectomy of gastric tumors and incorporated the advantages of different CNN models. The AI-assisted diagnostic system will play an important role in the therapeutic decision-making of EGC.

Analysis of Risk Factors for Urosepsis in Patients with Type 2 Diabetes Complicated by Upper Urinary Tract Stones: A Single-Centre Retrospective Study.

Urosepsis represents a complication of upper urinary tract stones (UUTSs) in patients with type 2 diabetes mellitus (T2DM), thus necessitating a comprehensive understanding of risk factors. This single-centre retrospective study aimed to analyse the risk factors for urosepsis in this patient population. Clinical data of patients with UUTS and T2DM admitted from January 2015 to January 2024 were collected and retrospectively analysed. Laboratory parameters, including white blood cell (WBC) count, serum creatinine, urine culture, C-reactive protein and imaging findings were assessed. Stepwise backward selection and logistic regression analysis was used to explore the risk factors of urosepsis. A total of 108 patients, including 56 patients complicated with urosepsis and 52 without urosepsis, were included. The urosepsis group exhibited significantly increased white blood cell count (15.75 ± 2.58 vs. 14.63 ± 2.76, p = 0.031), colony-forming units per millilitre in urine culture (5000.46 ± 1200.56 vs. 4570.13 ± 1000.24, p = 0.045), serum C-reactive protein levels (43.02 ± 12.36 vs. 38.54 ± 10.75, p = 0.047), presence of hydronephrosis (82.14% vs. 63.46%, p = 0.049) , ureteral stricture (46.43% vs. 25.00%, p = 0.034), prevalence of Gram-negative bacteria (85.71% vs. 67.31%, p = 0.042), antibiotic resistance (37.50% vs. 17.31%, p = 0.034), and empirical antibiotic use (62.50% vs. 40.38%, p = 0.035) compared with the non-urosepsis group. Gram-negative bacteria (odds ratio (OR) = 2.914, p = 0.027), antibiotic resistance (OR = 2.867, p = 0.022), renal hydronephrosis (OR = 2.648, p = 0.031), urethral stricture (OR = 2.600, p = 0.022) and antibiotic usage history (OR = 2.460, p = 0.023) exhibited significant OR values, whereas white blood cell (WBC) count demonstrated a moderate OR value (OR = 1.175, p = 0.034). These findings further underscore their potential to be reasonably predictive risk factors for urosepsis. This study identified various risk factors associated with urosepsis in patients with T2DM and UUTS. Laboratory parameters, imaging findings and urinary tract infection characteristics were found to be significant contributors to the development of urosepsis in this patient population.

Framework for Deep Learning Based Multi-Modality Image Registration of Snapshot and Pathology Images.

Multi-modality image registration is an important task in medical imaging because it allows for information from different domains to be correlated. Histopathology plays a crucial role in oncologic surgery as it is the gold standard for investigating tissue composition from surgically excised specimens. Research studies are increasingly focused on registering medical imaging modalities such as white light cameras, magnetic resonance imaging, computed tomography, and ultrasound to pathology images. The main challenge in registration tasks involving pathology images comes from addressing the considerable amount of deformation present. This work provides a framework for deep learning-based multi-modality registration of microscopic pathology images to another imaging modality. The proposed framework is validated on the registration of prostate ex-vivo white light camera snapshot images to pathology hematoxylin-eosin images of the same specimen. A pipeline is presented detailing data acquisition, protocol considerations, image dissimilarity, training experiments, and validation. A comprehensive analysis is done on the impact of pre-processing, data augmentation, loss functions, and regularization. This analysis is supplemented by clinically motivated evaluation metrics to avoid the pitfalls of only using ubiquitous image comparison metrics. Consequently, a robust training configuration capable of performing the desired registration task is found. Utilizing the proposed approach, we achieved a dice similarity coefficient of 0.96, a mutual information score of 0.54, a target registration error of 2.4 mm, and a regional dice similarity coefficient of 0.70.

Automatic Internal Stray Light Calibration of AMCW Coaxial Scanning LiDAR Using Black and White Image Patterns

Automatic Internal Stray Light Calibration of AMCW Coaxial Scanning LiDAR Using Black and White Image Patterns