Vision Research Research Articles

Best research Ever

Dr Rebekka Heitmar provides a round up of key clinical research from the 26th Annual Meeting of the European Association for Vision and Eye Research (Ever) 2023

Utilizing a composite citation index for evaluating clinical ophthalmology research: insights into gender, nationality, and self-citation among top ophthalmology researchers.

To compare the performance of a composite citation score (c-score) and its six constituent citation indices, including H-index, in predicting winners of the Weisenfeld Award in ophthalmologic research. Secondary objectives were to explore career and demographic characteristics of the most highly cited researchers in ophthalmology. A publicly available database was accessed to compile a set of top researchers in the field of clinical ophthalmology and optometry based on Scopus data from 1996 to 2021. Each citation index was used to construct a multivariable model adjusted for author demographic characteristics. Using area under the receiver operating curve (AUC) analysis, each index's model was evaluated for its ability to predict winners of the Weisenfeld Award in Ophthalmology, a research distinction presented by the Association for Research in Vision and Ophthalmology (ARVO). Secondary analyses investigated authors' self-citation rates, career length, gender, and country affiliation over time. Approximately one thousand unique authors publishing primarily in clinical ophthalmology/optometry were analyzed. The c-score outperformed all other citation indices at predicting Weisenfeld Awardees, with an AUC of 0.99 (95% CI: 0.97-1.0). The H-index had an AUC of 0.89 (95% CI: 0.83-0.96). Authors with higher c-scores tended to have longer career lengths and similar self-citation rates compared to other authors. Sixteen percent of authors in the database were identified as female, and 64% were affiliated with the United States of America. The c-score is an effective metric for assessing research impact in ophthalmology, as seen through its ability to predict Weisenfeld Awardees.

MMV_Im2Im: an open-source microscopy machine vision toolbox for image-to-image transformation.

Over the past decade, deep learning (DL) research in computer vision has been growing rapidly, with many advances in DL-based image analysis methods for biomedical problems. In this work, we introduce MMV_Im2Im, a new open-source Python package for image-to-image transformation in bioimaging applications. MMV_Im2Im is designed with a generic image-to-image transformation framework that can be used for a wide range of tasks, including semantic segmentation, instance segmentation, image restoration, image generation, and so on. Our implementation takes advantage of state-of-the-art machine learning engineering techniques, allowing researchers to focus on their research without worrying about engineering details. We demonstrate the effectiveness of MMV_Im2Im on more than 10 different biomedical problems, showcasing its general potentials and applicabilities. For computational biomedical researchers, MMV_Im2Im provides a starting point for developing new biomedical image analysis or machine learning algorithms, where they can either reuse the code in this package or fork and extend this package to facilitate the development of new methods. Experimental biomedical researchers can benefit from this work by gaining a comprehensive view of the image-to-image transformation concept through diversified examples and use cases. We hope this work can give the community inspirations on how DL-based image-to-image transformation can be integrated into the assay development process, enabling new biomedical studies that cannot be done only with traditional experimental assays. To help researchers get started, we have provided source code, documentation, and tutorials for MMV_Im2Im at [https://github.com/MMV-Lab/mmv_im2im] under MIT license.

Alterations in glutamate and nitric oxide metabolism in the RdS mice

Aims/Purpose: Nitric oxide (NO) is an essential signalling molecule, with gaseous nature. NO is synthesized by the conversion of L‐arginine to L‐citrulline. This reaction is catalysed by the enzyme nitric oxide synthase (NOS). There are three NOS isoforms, but nNOS reflects most of the production of NO in the retina. Although NO is employed in the retina to maintain ocular function, it has also been related to different visual diseases. Increased levels of NO have been associated with ocular diseases involving increased glutamate concentrations, ischaemia, oxidative stress and inflammation (such as retinitis pigmentosa (RP)). The main purpose of this work was to determine glutamate retinal concentrations at different stages of the retinal degeneration that occurs in the rds mice and to study possible NO metabolism alterations.Methods: Rds mice were treated in accordance with the ARVO statement for the use of animals in ophthalmic and vision research. Mice were sacrificed at different postnatal days (11, 17, 21 and 28). Retinal glutamate concentrations were determined by high pressure liquid chromatography (HPLC). The expression of different enzymes related with NO metabolism (nNOS, guanylate cyclase (GC) and arginase) was determined by histochemistry.Results: Glutamate as well as nNOS was observed to be increased in the retina of rds mice mainly between the peak of rods and cones death. GC was observed to be decreased in the rds mice compared to control ones and no significant changes were observed in arginase expression when we compared rds and control mice.Conclusions: There are alterations in the expression of glutamate, nNOS and GC in rds mice retina compared to control mice. RP is a disease with no effective treatment to date and NO metabolism may be a new target in RP treatment.

DECODING OF THE EXTRAOCULAR MUSCLES ACTIVATIONS BY COMPLEXITY-BASED ANALYSIS OF ELECTROMYOGRAM (EMG) SIGNALS

The analysis of extraocular muscles’ activation is crucial for understanding eye movement patterns, providing insights into oculomotor control, and contributing to advancements in fields such as vision research, neurology, and biomedical engineering. Ten subjects went through the experiments, including normal watching, blinking, upward and downward movements of eyes, and eye movements to the left and right while their electromyogram (EMG) signals were recorded. We analyzed the complexity of recorded EMG signals using fractal theory, sample entropy, and approximate entropy (ApEn). The results showed that the techniques are able to decode the changes in the complexity of EMG signals between different eye movements. In other words, we can use these methods to study extraocular muscle activations in different conditions.

Basic concepts of exercise physiology for the vision and eye research community

The epidemiology of the second half of the 20th century has been marked by a transition from infectious to non‐communicable diseases. Instead of having a unique etiological cause, the non‐communicable diseases are believed to develop by the temporal confluence of several risk factors on a given individual. Risk factors of genetic origin are considered non‐modifiable nowadays. In contrast, behavioural risk factors, including lifestyle, are considered modifiable, and, therefore the preventive and adjuvant interventions aimed to reduce the risk or modify the natural course of non‐communicable diseases are based in lifestyle changes.Since the seminal work of Jeremy Morris in the 50's (the so‐called “London Bus Study”) epidemiological evidences have been accumulating over time on the potential role of a reduced physical activity in every domain of daily life as a risk factor for several non‐modifiable diseases. This is aggravated by the progressive trend of a reduced physical activity performed by the worldwide population during the last decades. The relationship of physical inactivity and/or sedentarism with several systemic diseases is clearly proven nowadays, and the underlying pathophysiological explanations are starting to be understood.In contrast what it happens in other organs or systems the potential effects of physical activity on ocular physiology and/or pathophysiology are still almost unknown. The design and development of future studies focused on these topics would need that the eye and vision research community is familiarized with: a) the current terminology on the field of physical activity and exercise; b) the hormetic properties of exercise, and consequently, the differences among acute responses and chronic adaptations to it; and, c) the potential physiological mechanisms ‐ranging from classical neuroendocrine responses to the recent concept of exerquines‐ that may explain them. All of them will be presented and discussed at this talk.

An Overview of Future Development Methods of Infectious Bronchitis Vaccines

Vaccines are major disease-controlling tools in human and animal practices. Vaccination controlled or even eradicated a long list of diseases worldwide. Classic viral vaccines are usually composed of live or inactivated whole viruses and have been produced for many years. However, they are unsuccessful, especially in persistent infections, fast-evolving viruses, complex and compound antigens, and emerging agents. Novel vaccine development technologies such as DNA, protein, or viral vector vaccines have revolutionized vaccine development and opened a new horizon for study and research in vaccine research and production vision. Newly-developed vaccines, or even most traditional ones, are based on new technologies, especially in human diseases where cost and complications in production can be ignored. However, in animal health, especially for commercial poultry production, the cost of development, simplicity, and mass application of large-scale production cannot be overlooked. In recent years, the significance of producing novel vaccines has been highlighted in parallel with technological advancement, especially with emerging novel variants of infectious bronchitis viruses—serious poultry pathogens for years. In this review, we will introduce some studies on novel vaccine development techniques and investigate the results of those vaccines in the protection of chickens and their clinical outcomes.

Uncertainty Modeling for Gaze Estimation.

Gaze estimation is an important fundamental task in computer vision and medical research. Existing works have explored various effective paradigms and modules for precisely predicting eye gazes. However, the uncertainty for gaze estimation, e.g., input uncertainty and annotation uncertainty, have been neglected in previous research. Existing models use a deterministic function to estimate the gaze, which cannot reflect the actual situation in gaze estimation. To address this issue, we propose a probabilistic framework for gaze estimation by modeling the input uncertainty and annotation uncertainty. We first utilize probabilistic embeddings to model the input uncertainty, representing the input image as a Gaussian distribution in the embedding space. Based on the input uncertainty modeling, we give an instance-wise uncertainty estimation to measure the confidence of prediction results, which is critical in practical applications. Then, we propose a new label distribution learning method, probabilistic annotations, to model the annotation uncertainty, representing the raw hard labels as Gaussian distributions. In addition, we develop an Embedding Distribution Smoothing (EDS) module and a hard example mining method to improve the consistency between embedding distribution and label distribution. We conduct extensive experiments, demonstrating that the proposed approach achieves significant improvements over baseline and state-of-the-art methods on two widely used benchmark datasets, GazeCapture and MPIIFaceGaze, as well as our collected dataset using mobile devices.

Vision and eye research

We are all, no matter whether we are basic scientists or ophthalmologist, whether it is question of discovering of novel cellular pathways, druggable targets, new therapeutical tools or technological innovations, protecting of our visual system, treating diseases threatening its function or replacing these functions, working for vision. Good vision is one of the key elements of good health‐related quality of life (HRQoL). It is a factor of working ability and self‐activity. It is important for individual development and education, understanding of the world. Poor vision on the other hand is related to decreased HRQoL, increased risk of accidents, increased use of healthcare services, early retirement, loss of productivity and problems for independent initiative. In this presentation I will describe the relationships of visual acuity, HRQoL, use of healthcare services, and socioeconomical cost based on data of nationwide health surveys and healthcare registers. The results clearly demonstrate the significant role of vision for individuals and societies. I hope, they further motivate us to work harder in our laboratories and clinics and hopefully also convince funding organizations and governments to understand, how by supporting our efforts, it is possible not only to increase the welfare but also save money in our societies.

Logistic Regression Based Model for Pain Intensity Level Detection from Biomedical Signal

Hospitals today make the effort to assess acute pain with self-report assessments such as the quantitative pain the level of intensity Index and visual input index. However, because these techniques rely on patient input, they are imprecise. Thus, an objective, statistical approach to ongoing pain monitoring is necessary. In computer vision research, identifying pain intensity is a difficult challenge to solve. However, current subjective pain evaluation is unreliable because it heavily relies on the patient's response. In order to improve the standardization of pain evaluation, automated pain identification using physiological data might provide essential objective information. In the present study, we provide an objective pain recognition approach based on physiological signals that can extract novel features from electromyography (EMG), electrodermal activity (EDA), and electrocardiogram (ECG) data that have not previously been utilized for pain recognition. Utilizing the Bio-Vid Heat Pain Database (Part A) for evaluation and clinical validation, the proposed machine learning logistic regression-based method performs significantly better than previous techniques recorded in the literature for both the electrodermal activity (EDA) and fusion approaches, with average performances of 82.36% and 83.20% for the binary classification experiment that discriminates within the baseline and the pain tolerance level (T0 vs. T4). Our result shows that it outperforms most of the previously proposed methods in related works.

Remembering Peter Schiller, principled pioneer of vision research

Remembering Peter Schiller, principled pioneer of vision research

Comparison of retinal parameters between rhesus and cynomolgus macaques.

Nonhuman primates are important research models for basic vision research, preclinical pathogenesis, and treatment studies due to strong similarities in retinal structure and function with humans. We compared retinal parameters between 10 healthy normal rhesus macaques (Macaca mulatta) and 10 cynomolgus macaques (Macaca fascicularis) by optical coherence tomography and electroretinography. The Heidelberg Spectralis® HRA+OCT and Roland multifocal electrophysiometer were used to analyze retinal morphology, multifocal electroretinograms (mfERGs), and full-field electroretinograms (ff-ERGs). Mean retinal thickness was lowest in the central fovea of macaques and did not differ significantly between species, but the retinal thicknesses of the nerve fiber ganglion cell layer and the inner plexiform layer were significantly different. The amplitude density of the N1 wave was lower in rhesus macaques than in cynomolgus macaques in ring and quadrant areas. Dark-adapted 3.0 oscillatory potentials (reflection of amacrine cell activity) and light-adapted 30-hz flicker ERG (a sensitive cone-pathway-driven response) waveforms of the ff-ERG were similar in both species, while the times to peaks in dark-adapted 0.01 ERG (the rod-driven response of bipolar cells) and dark-adapted 3.0 ERG (combined rod and cone system responses) as well as the implicit times of the a- and b-waves in light-adapted 3.0 ERG (the single-flash cone response) were substantially different. This study provides normative retinal parameters for nonhuman primate research on basic and clinical ophthalmology, as well as a reference for researchers in the appropriate selection of rhesus or cynomolgus macaques as models for ophthalmology studies.

Optimization study of anomaly detection algorithm in machine vision inspection technology

Abstract In recent years, target detection algorithms based on machine vision have been a hotspot in computer vision research. The You Only Look Once (YOLO) algorithm, as an excellent target detection algorithm, has played an important role in improving detection speed and accuracy with the improvement of the network architecture in its development process. This paper introduces the concept of integrated learning to the YOLOv5 network architecture, incorporating deformable convolution and attention mechanisms. It also chooses the focal EIOU loss function to replace the GIOU loss function, thereby addressing the issue of localization loss, prioritizing abnormally detected targets, and enhancing the detection efficiency of these abnormal targets. Finally, we examine the practical value of the improved YOLOv5 algorithm by testing its performance and applying it to real-world anomaly detection. The results show that the improved YOLOv5 model outperforms the original YOLOv5 model in terms of performance and practical application advantages. In terms of performance, the classification accuracy of sea_person and earth_person in the improved YOLOv5 model is 37% and 26%, respectively, which is a significant performance gain overall. In actual application tests, the proposed method is more accurate than the traditional method. The accuracy is significantly higher.

Crafting Expressive Faces: A Comparative Exploration of Facial Animation Techniques in VToonify and AnimeGan

Abstract: This review paper presents an overview and comparative analysis of two cutting-edge technologies: VToonify and AnimeGAN, designed to enhance portrait video style transfer and photo animation, respectively. VToonify introduces a novel high-resolution portrait video style transfer approach, offering users enhanced control over the transformation process. On the other hand, AnimenGAN is a lightweight Generative Adversarial Network (GAN) designed specifically for photo animation with a focus on generating anime-style outputs. This review delves into the underlying methodologies and technical principles employed by both VToonify and AnimeGAN. We discuss the crucial features that set these approaches apart from traditional methods and their respective strengths in terms of controllability, high-resolution output, and efficiency. Furthermore, the paper investigates each technology's key challenges and potential areas for future improvements. The review highlights the practical applications of VToonify and AnimeGAN in the realm of creative content generation, multimedia, and visual storytelling. Moreover, we explore real-world use cases and evaluate the impact of these technologies on various industries, including entertainment, advertising, and social media, Through a comprehensive analysis, this paper aims to provide readers with an informed understanding of the state-of-the-art in portrait video style transfer and photo animation. By combining insights from VToonify and AnimeGAN, this review contributes to advancing research in computer vision, deep learning, and artistic content creation.

Suspicious Activity Detection

Abstract: Unpredictable activities is guessing a person's physical location or joint activity based on images or videos. The project will require the use of neural networks to detect human activities from CCTV images. Human behavior is one of the fundamental problems in computer vision and has been studied for over 15 years. This is important because the number of applications that can benefit from the search function is huge. For example, it is used in applications such as human prediction, video analysis, animal analysis and behavior understanding, language recognition, human relations-computer, and characters are less powerful. While low-cost depth sensors have limitations such as being limited to indoor use, their low-noise data and low depth make it difficult to estimate humans from depth images. Therefore, we plan to use neural networks to solve these problems. Analyzing human activity through image analysis is an active area of research in image processing and computer vision. Thanks to visual monitoring, human activities in public places such as bus stops, train stations, airports, banks, shopping malls, schools, parking lots and roads can be monitored in order to prevent attacks, theft, accidents, illegal parking, violence, fights and chain events. purse snatching etc. illegal and other activities. It is very difficult to monitor public places regularly, so there is a need for intelligent video surveillance that can monitor people's activities in real time and divide them into tasks, normal and abnormal activities; and can generate an alarm. Often research is done with images rather than video. Also, none of the published articles attempt to use CNNs to detect suspicious activity.

Effects of low-vision rehabilitation on reading speed and depression in age-related macular degeneration: A systematic review and meta-analysis

Age-related macular degeneration (AMD) is a prevalent eye disease, which can lead to vision loss, impacting daily functioning. This study aimed to evaluate the effect of low-vision rehabilitation techniques such as magnifiers, telescopes, and microperimeter feedback, on improving reading speed and reducing depression in AMD patients. This study is a systematic review and meta-analysis. A systematic search of databases MEDLINE (OVID), EMBASE (OVID), and CINHAL (EBSCO), gray literature including ClinicalTrials.gov and ProQuest Dissertations and Theses Global (ProQuest), Conferences of The Association for Research in Vision and Ophthalmology, American Academy of Ophthalmology, and Canadian Ophthalmological Society were done from inception to 27 July 2022. Fixed-effect and random-effect models were applied to account for heterogeneity between studies. Publication bias was checked through funnel plots. A total of 33 studies (2611 subjects) were included;14 studies were included in the meta-analysis (1123 subjects) and 19 were included in the qualitative analysis. We found a non-significant increase in reading speed (effect size [ES] = 0.02, 95% confidence interval [CI] = [−0.18, 0.23]) for studies comparing patients who received no low-vision rehabilitation, versus the intervention group of AMD patients who received low-vision rehabilitation. We also found a significant decrease in reading speed (ES = −0.92, 95% CI = [−1.46, −0.39]) between studies evaluating the difference in reading speed at baseline visits compared to the follow-up visits in AMD patients who received low-vision rehabilitation. Low-vision rehabilitation was also found to result in a decrease in depression severity when compared to control groups, as well as from initial assessment to subsequent follow-up visits. Individuals with AMD who engaged in low-vision rehabilitation demonstrated improved reading speeds compared to their own baseline and to those who did not receive any interventions. Furthermore, these interventions also contributed to a reduction in symptoms of depression.

Fairness in Deep Learning: A Survey on Vision and Language Research

Despite being responsible for state-of-the-art results in several computer vision and natural language processing tasks, neural networks have faced harsh criticism due to some of their current shortcomings. One of them is that neural networks are correlation machines prone to model biases within the data instead of focusing on actual useful causal relationships. This problem is particularly serious in application domains affected by aspects such as race, gender, and age. To prevent models from incurring unfair decision-making, the AI community has concentrated efforts on correcting algorithmic biases, giving rise to the research area now widely known as fairness in AI . In this survey paper, we provide an in-depth overview of the main debiasing methods for fairness-aware neural networks in the context of vision and language research. We propose a novel taxonomy that builds upon previous proposals but is tailored for deep learning research to better organize the literature on debiasing methods for fairness. We review all important neural-based methods and evaluation metrics while discussing the current challenges, trends, and important future work directions for the interested researcher and practitioner.

TransCNNLoc: End-to-end pixel-level learning for 2D-to-3D pose estimation in dynamic indoor scenes

Accurate localization in GPS-denied environments has always been a core issue in computer vision and robotics research. In indoor environments, vision-based localization methods are susceptible to changes in lighting conditions, viewing angles, and environmental factors, resulting in localization failures or limited generalization capabilities. In this paper, we propose the TransCNNLoc framework, which consists of an encoding–decoding network designed to learn more robust image features for camera pose estimation. In the image feature encoding stage, CNN and Swin Transformer are integrated to construct the image feature encoding module, enabling the network to fully extract global context and local features from images. In the decoding stage, multi-level image features are decoded through cross-layer connections while computing per-pixel feature weight maps. To enhance the framework’s robustness to dynamic objects, a dynamic object recognition network is introduced to optimize the feature weights. Finally, a multi-level iterative optimization from coarse to fine levels is performed to recover six degrees of freedom camera pose. Experiments were conducted on the publicly available 7scenes dataset as well as a dataset collected under changing lighting conditions and dynamic scenes for accuracy validation and analysis. The experimental results demonstrate that the proposed TransCNNLoc framework exhibits superior adaptability to dynamic scenes and lighting changes. In the context of static environments within publicly available datasets, the localization technique introduced in this study attains a maximal precision of up to 5 centimeters, consistently achieving superior outcomes across a majority of the scenarios. Under the conditions of dynamic scenes and fluctuating illumination, this approach demonstrates an enhanced precision capability, reaching up to 3 centimeters. This represents a substantial refinement from the decimeter scale to a centimeter scale in precision, marking a significant advancement over the existing state-of-the-art (SOTA) algorithms. The open-source repository for the method proposed in this paper can be found at the following URL: github.com/Geelooo/TransCNNloc.

DYNet: A Printed Book Detection Model Using Dual Kernel Neural Networks.

Target detection has always been a hotspot in image processing/computer vision research, and small-target detection is a frequently encountered problem in the field of target detection. With the continuous innovation of target detection technology, people always hope that the detection of small targets can reach the real-time accuracy of large-target detection. In this paper, a small-target detection model based on dual-core convolutional neural networks (CNN) is proposed, which is mainly used for the intelligent detection of books in the production line of printed books. The model is mainly composed of two modules, including a region prediction module and suspicious target search module. The region prediction module uses a CNN to predict suspicious region blocks in a large context. The suspicious target search module uses a different CNN from the above to find tiny targets in the predicted region blocks. Comparative testing of four small book target samples using this model shows that this model has better book small-target detection accuracy compared to other models.

HiroPoseEstimation: A Dataset of Pose Estimation for Kid-Size Humanoid Robot

Pose estimation is a field of computer vision research that involves detecting, associating, and tracking data points on body parts. It is used for health monitoring, sign language understanding, human gesture control, elderly activities, sports, and humanoid robot pose estimation. The anatomy of a humanoid robot is similar to a human, which forms the basis for utilizing humanoid robot pose estimation. The Humanoid League is a major domain of the RoboCup competition, featuring soccer matches between humanoid robots. Pose estimation is used to measure the robot’s performance. Nevertheless, there have not been many research done on this subject. A new dataset model needs to be developed to solve the proposed problem. This work introduces HiroPoseEstimation, a kid-size humanoid robot dataset with several types of robots used in various poses based on movements in a soccer game. It is evaluated with both bottomup and top-down approaches using keypoint mask R-CNN and single-stage encoder-decoder model. Both methods demonstrate good performance on the proposed dataset.