相关参考文献
注意:仅列出部分参考文献,下载原文获取全部文献信息。
Article
Computer Science, Artificial Intelligence
Bineeth Kuriakose et al.
Summary: Navigation assistance is crucial for people with vision impairments, but existing navigation assistants lack portability and convenience. To address this, we propose DeepNAVI, a smartphone-based navigation assistant that leverages deep learning to provide detailed obstacle information and offer audio mode for user convenience.
EXPERT SYSTEMS WITH APPLICATIONS
(2023)
Article
Computer Science, Software Engineering
Wenjie Chen et al.
Summary: In this paper, an intelligent guide system based on a smartphone is proposed. Videos of tactile paving captured by the phone camera are processed and analyzed, and guide messages are sent to the user in the form of sound or vibration. A fine-tuned MobileNet model is used for feature extraction, and Single Shot MultiBox Detector (SSD) is employed for TWSI detection. The user's position is determined using the Score Voting algorithm, and corresponding guide information is provided. The system achieves high accuracy and real-time performance, and can replace bumpy tactile paving with flat stickers or paint with TWSI patterns.
JOURNAL OF SYSTEMS AND SOFTWARE
(2023)
Review
Engineering, Electrical & Electronic
Bineeth Kuriakose et al.
Summary: This paper provides a comprehensive review of navigation solutions for the visually impaired, highlighting the lack of core features and the need to consider humanitarian conditions in system design. It also offers recommendations for future navigation system development.
IETE TECHNICAL REVIEW
(2022)
Article
Architecture
Hosam Elhegazy et al.
Summary: This research focuses on composite flooring systems for multi-story buildings and the application of value engineering techniques in the construction industry, using data from RSMeans from 1997 to 2019. A simple computer model is designed to recommend the optimal composite flooring system for multi-story buildings during the preliminary design stage.
JOURNAL OF ASIAN ARCHITECTURE AND BUILDING ENGINEERING
(2022)
Article
Ergonomics
Daniela Salinas-Toro et al.
Summary: The study found that during the coronavirus pandemic, an increase in VDT hours among subjects working remotely was associated with a significant increase in ES symptoms and a high prevalence of DED.
INTERNATIONAL JOURNAL OF OCCUPATIONAL SAFETY AND ERGONOMICS
(2022)
Review
Pharmacology & Pharmacy
Pablo Argueso
Summary: Mucins are a type of ancient glycoproteins that provide viscoelastic, lubricating, and hydration properties to wet surfaced epithelia. They form a barrier with selective permeability to protect underlying tissues. In the eye, mucins are expressed, processed, and spatially distributed based on specific organ requirements to protect against environmental insults while allowing the passage of light. The adaptation of human ocular surface epithelia includes the production of a thin and watery tear film containing soluble mucin. Transmembrane mucins play a significant role in the tear film, occupying up to one-quarter of its total thickness. Understanding the properties and functions of ocular mucins is crucial for developing applications with potential biological and clinical impact.
ADVANCED DRUG DELIVERY REVIEWS
(2022)
Article
Cardiac & Cardiovascular Systems
Seyedeh Maryam Zekavat et al.
Summary: Using machine learning to assess retinal vasculature, this study found that low retinal vascular density and fractal dimension were associated with increased risk for various diseases, and identified novel genes linked to angiogenesis and inflammation.
Article
Ophthalmology
Swati Upadhyaya et al.
Summary: The study aimed to evaluate the sensitivity and specificity of a portable non-mydriatic fundus camera for assessing glaucoma. The results showed high sensitivity and specificity of the camera for diagnosing glaucoma remotely, suggesting it as an effective tool for community outreach programs.
Article
Computer Science, Artificial Intelligence
V. Desika Vinayaki et al.
Summary: This study proposes a novel framework for detecting and classifying diabetic retinopathy (DR) lesions. The framework includes four stages: pre-processing, segmentation, feature extraction, and classification. Experimental results show that the proposed framework exhibits superior performance in DR detection and classification.
NEURAL PROCESSING LETTERS
(2022)
Article
Ophthalmology
Sneha Singh et al.
Summary: This study found that conjunctival epithelial cells are susceptible to SARS-CoV-2 infection, with the beta VOC variant being more infectious and evoking a stronger antiviral and inflammatory response compared to the parental strain.
Article
Ophthalmology
Monireh Mahjoob et al.
Summary: Visual acuity significantly reduces under mental load, and this reduction is related to alterations in responses within the visual cortex.
OPHTHALMIC AND PHYSIOLOGICAL OPTICS
(2022)
Article
Automation & Control Systems
Michal Adamik et al.
Summary: This article presents a novel robotic system for creating realistic pencil drawings based on image evolution, showing that machine creativity can be achieved through procedural image generation using genetic algorithms, and that aesthetic and fast drawings can be produced even with simple line segment approximations.
ROBOTICS AND AUTONOMOUS SYSTEMS
(2022)
Article
Computer Science, Artificial Intelligence
V. Santhosh Kumar Tangudu et al.
Summary: A newly proposed model for COVID-19 detection using chest X-ray images achieves high accuracy and improves training time. Early detection of infection is crucial for survival, and this model shows promising results.
Review
Endocrinology & Metabolism
Karis Little et al.
Summary: Type 2 diabetes is associated with complications like diabetic retinopathy and cognitive decline, with chronic hyperglycemia and insulin resistance playing crucial roles. Understanding these common pathological mechanisms may help in identifying T2D patients at risk of cognitive decline.
TRENDS IN ENDOCRINOLOGY AND METABOLISM
(2022)
Article
Environmental Sciences
Na Yu et al.
Summary: This study investigated the effects of button features on user performance in smart home interfaces. The results showed that button size and graphics/text ratio significantly influenced user performance in both age groups, while icon style only affected the senior group. Elderly participants preferred larger buttons with larger texts and skeuomorphic icons, while young participants preferred equal-sized graphics and text.
INTERNATIONAL JOURNAL OF ENVIRONMENTAL RESEARCH AND PUBLIC HEALTH
(2022)
Review
Ophthalmology
Thales Antonio Cabral de Guimaraes et al.
Summary: Age-related macular degeneration (AMD) is the leading cause of irreversible blindness in the developed world, with promising treatment prospects for wet AMD but challenges remain for dry AMD. Various therapeutic approaches are being investigated, including drugs with antioxidative properties, complement cascade inhibitors, gene therapy, etc., with promising results from early clinical trials.
BRITISH JOURNAL OF OPHTHALMOLOGY
(2022)
Article
Chemistry, Multidisciplinary
Darius Plikynas et al.
Summary: This paper proposes a method of utilizing social outsourcing for indoor navigation, where volunteers record and maintain indoor route databases to assist blind and visually impaired individuals in accessing convenient indoor navigation information.
APPLIED SCIENCES-BASEL
(2022)
Article
Chemistry, Multidisciplinary
Yazed Alsaawy et al.
Summary: Traffic congestion is a global challenge, and this paper proposes a comprehensive framework for solving the problem. The framework consists of four layers and includes various technologies and algorithms for efficient and reliable traffic management.
APPLIED SCIENCES-BASEL
(2022)
Article
Ophthalmology
Barbara Burgos-Blasco et al.
Summary: The study suggests that recovered COVID-19 patients may experience changes in retinal layers, with an increase in global RNFLT, a decrease in macular RNFL, and an increase in ganglion cell layer thickness. Patients with anosmia and ageusia showed more pronounced alterations in these retinal layers.
EUROPEAN JOURNAL OF OPHTHALMOLOGY
(2022)
Article
Ophthalmology
John M. Nesemann et al.
Summary: The study found that visually impaired participants were more likely to self-report disabilities in other domains, suggesting potential unintended benefits of public health programs designed to reduce visual impairment.
Article
Computer Science, Information Systems
Usman Masud et al.
Summary: The goal of this research is to create a simple, convenient, and cost-effective wearable system for visually impaired individuals. By utilizing hardware components such as Raspberry-Pi, camera, ultrasonic sensor, and Arduino, the system aims to assist visually impaired people in their daily routines through obstacle detection and scenes classification.
Article
Computer Science, Information Systems
Sha Zhu et al.
Summary: This article focuses on the energy consumption of AIoT tasks and proposes an optimization method for task scheduling that utilizes intelligent edge computing. The proposed method improves energy efficiency compared to other strategies, as demonstrated through extensive simulation-based performance evaluation.
IEEE INTERNET OF THINGS JOURNAL
(2022)
Article
Architecture
Won-Kee Hong et al.
Summary: This study introduces artificial neural networks (ANNs) for solving structural engineering problems and designing ductile concrete beams, which demonstrate a learning and memory capability similar to human brains. ANNs are developed based on large structural design datasets for engineering applications rather than structural mechanics. The study aims to develop design of doubly reinforced concrete beams based on ANN, presenting a reverse design with design charts for engineers to solve problems accurately and rapidly.
JOURNAL OF ASIAN ARCHITECTURE AND BUILDING ENGINEERING
(2022)
Article
Computer Science, Information Systems
Rabbia Mahum et al.
Summary: This study aims to detect glaucoma at early stages using deep learning-based feature extraction. Retinal fundus images are utilized for training and testing, with image pre-processing and segmentation used to extract the region of interest and derive features of the optic disc. Multi-class classifiers are employed for classification. Experimental results demonstrate the high accuracy of the proposed model for early detection of glaucoma.
Article
Computer Science, Cybernetics
Ruben Alcaraz Martinez et al.
Summary: Statistical charts play a crucial role in conveying information, but pose challenges for individuals with low vision and color vision deficiency. This research aims to create a heuristic set of indicators to evaluate the accessibility of statistical charts for these user profiles, expanding accessibility requirements in the field.
UNIVERSAL ACCESS IN THE INFORMATION SOCIETY
(2022)
Article
Engineering, Biomedical
N. Deepa et al.
Summary: This study proposes a method for AD classification based on optimized VGG-16 architecture and arithmetic optimization algorithm. It involves pre-processing, segmentation, and classification. The CAT12 toolkit is used for image processing, and image enhancement techniques are applied to improve contrast levels. Experimental results demonstrate superior classification performance compared to other methods.
BIOMEDICAL SIGNAL PROCESSING AND CONTROL
(2022)
Article
Ophthalmology
Takahiko Izumi et al.
Summary: The choroid in eyes with CSC has a steeper slope around the posterior pole, and UWF-OCT can be used to evaluate abnormalities of choroidal structures in eyes with CSC from the posterior pole to the periphery.
GRAEFES ARCHIVE FOR CLINICAL AND EXPERIMENTAL OPHTHALMOLOGY
(2022)
Article
Urology & Nephrology
Katherine R. Tuttle et al.
Summary: Diabetic kidney disease is a common cause of kidney failure worldwide, but screening and treatment rates are low. Effective therapies exist but are underutilized due to high drug costs and low reimbursement rates. Data on clinical and cost-effectiveness are needed to gain support. The pharmaceutical industry should lower drug prices to improve access. Personalized multidisciplinary models and community outreach efforts are essential for optimal care. Patient preferences should shape implementation strategies.
CLINICAL JOURNAL OF THE AMERICAN SOCIETY OF NEPHROLOGY
(2022)
Article
Computer Science, Artificial Intelligence
Sreetama Mukherjee et al.
Summary: This paper introduces a hybrid style transfer model for music composer styles, focusing on three composers from the Maestro dataset and achieving high accuracy in classification tasks.
EXPERT SYSTEMS WITH APPLICATIONS
(2022)
Article
Clinical Neurology
Laura C. Donaldson et al.
Summary: This study investigated whether the pattern of visual field (VF) loss in patients with chiasmal compression predicted the appearance of the causative lesion on neuroimaging. The results showed that junctional scotoma (JXS) was more often seen with larger lesions and when there was compression of both the prechiasmatic optic nerve (ON) and ON-OC junction. It also emphasized the importance of macular optical coherence tomography (OCT) in the evaluation of patients with lesions involving the optic chiasm.
JOURNAL OF NEURO-OPHTHALMOLOGY
(2022)
Review
Medicine, General & Internal
Roger Chou et al.
Summary: This study found limited direct evidence on glaucoma screening, showing no association with benefits. Screening tests can identify persons with glaucoma and treatment was associated with a lower risk of glaucoma progression, but evidence of improvement in visual outcomes, quality of life, and function remains lacking.
JAMA-JOURNAL OF THE AMERICAN MEDICAL ASSOCIATION
(2022)
Article
Ophthalmology
Shiho Mochida et al.
Summary: This study identified the pathogenesis of peripheral visual field defects in high myopia and investigated the association between focal lamina cribrosa defects and high myopia-specific peripheral visual field defects. The findings suggest that there is an important relationship between these two factors and that microperimetry is a reliable method for evaluating peripheral visual field defects in highly myopic patients.
JAPANESE JOURNAL OF OPHTHALMOLOGY
(2022)
Article
Computer Science, Artificial Intelligence
Mohammed Abdulla Salim Al Husaini et al.
Summary: Breast thermography supported by deep convolutional neural networks is expected to contribute significantly to early detection of breast cancer. Different recent deep learning methods were investigated for identifying breast disorders. Results showed that Inception V4 and MV4 had high accuracy with color images, while Inception V3 outperformed V4 and MV4 with grayscale images. Increasing the number of layers may not necessarily improve the performance.
NEURAL COMPUTING & APPLICATIONS
(2022)
Article
Clinical Neurology
Valerie Cochen De Cock et al.
Summary: This study investigated the safety and efficacy of subcutaneous night-time apomorphine infusion in treating insomnia in patients with Parkinson's disease. The results showed that this treatment significantly improved sleep disturbances and had an overall good safety profile.
Article
Computer Science, Artificial Intelligence
Shukai Wu et al.
Summary: This study utilizes deep neural networks for flexible artistic creation, synthesizing face portraits by using reference photos and line art drawings. It focuses on learning feature representations for the line domain and the reference domain, and employs attention mechanisms and spatial correspondence to generate more pleasing images.
Article
Ophthalmology
Richard F. Spaide et al.
Summary: This study evaluated the degenerative findings in the premacular vitreous and cistern formation using an enhanced vitreous imaging method. The results showed that the vitreous had stereotypic patterns of degeneration with specific orientations of fibers in the macular region. The formation of vitreous syneresis and synchysis seemed to be related to the organization architecture of the vitreous and the effects of eye motion over time.
RETINA-THE JOURNAL OF RETINAL AND VITREOUS DISEASES
(2022)
Article
Computer Science, Artificial Intelligence
Ali H. Al-Timemy et al.
Summary: This study introduces a novel method based on corneal topographic maps for improving the diagnosis accuracy of keratoconus (KCN) using deep network classifiers and transfer learning strategy. The results show the significant improvement in accuracy by utilizing ensemble classifiers, with networks learning features that align with clinical knowledge.
COGNITIVE COMPUTATION
(2022)
Review
Chemistry, Analytical
Siguo Bi et al.
Summary: With the rise of artificial intelligence and the Internet of Things, smart devices have become interconnected and widely applied in various fields. In the field of librarianship, the era of Smart Libraries powered by AI and IoT has arrived. This research provides a comprehensive overview of AI and IoT technologies in three fundamental aspects: smart service, smart sustainability, and smart security. It also highlights the future trends of smart libraries.
Article
Construction & Building Technology
Thibault Marzullo et al.
Summary: The study introduces an open-source building performance simulation test bed ACTB, which interfaces high-fidelity building models with advanced controllers implemented in Python. Leveraging tools like an external clock and REST API, ACTB provides key performance indicators for evaluating the effectiveness of control strategies.
JOURNAL OF BUILDING PERFORMANCE SIMULATION
(2022)
Article
Computer Science, Information Systems
Lin Jia et al.
Summary: This article investigates how to coordinate edge and cloud resources for cost-efficient continuous learning in AIoT applications. A cost-efficient optimization framework is designed and analyzed, leveraging Lyapunov optimization theory, to make online decisions for dynamically arrived new data samples.
IEEE INTERNET OF THINGS JOURNAL
(2022)
Editorial Material
Oncology
Harrison Phillips et al.
Article
Medicine, General & Internal
Muhammad Mujahid et al.
Summary: Pneumonia is a leading cause of death in infants and elderly people, and accurate diagnosis systems are needed. This study uses X-ray images for pneumonia detection and experiments with pre-trained convolutional neural networks and ensemble models. The results show that the model using Inception-V3 and CNN achieves the highest accuracy and recall score.
Article
Computer Science, Artificial Intelligence
Waleed Al Shehri et al.
Summary: This study investigates the adoption of assistive technology in Saudi universities and identifies the factors influencing its acceptance. The findings reveal that limited awareness of visual disability and assistive technology, as well as psychological sensitivity of disabled users, are the primary factors affecting acceptance. The study also provides recommendations for overcoming the barriers that hinder the acceptance of assistive techniques by Saudi students with visual disabilities.
PEERJ COMPUTER SCIENCE
(2022)
Article
Multidisciplinary Sciences
Jahanzaib Latif et al.
Summary: In this paper, a novel two-phase ODGNet for glaucoma diagnosis is proposed, which incorporates visual saliency maps and shallow CNN for accurate OD localization. The experimental results demonstrate that the proposed ODGNet achieves high predictive performance and accuracy.
SN APPLIED SCIENCES
(2022)
Article
Computer Science, Hardware & Architecture
Sakib Shahriar
Summary: This article introduces the research on generating artworks using generative adversarial networks (GANs). It includes a comprehensive survey of recent works using GANs for generating visual arts, music, and literary text, as well as a performance comparison and description of various GAN architectures. The article also highlights some of the key challenges in GAN-based art generation and offers recommendations for future work.
Article
Multidisciplinary Sciences
Waqar Ahmad et al.
Summary: This paper proposes a Generative Adversarial Network (GAN) based architecture for mapping low-resolution medical images to high-resolution images. The architecture utilizes progressive upscaling and an additional loss term to extract rich details and achieve superior accuracy on four different medical image datasets.
SCIENTIFIC REPORTS
(2022)
Article
Multidisciplinary Sciences
Joon Hyung Yeo et al.
Summary: The study evaluated the efficacy of a wearable smartphone-based low vision aid in patients with visual impairment, showing significant improvements in visual function and positive impact on the quality of life in younger patients.
SCIENTIFIC REPORTS
(2022)
Review
Chemistry, Multidisciplinary
Leila Mohammadpour et al.
Summary: With the advancement and widespread use of Internet applications in recent years, the need for securing Internet networks has increased. Intrusion detection systems (IDSs) employing artificial intelligence (AI) methods, particularly deep learning (DL) algorithms such as convolutional neural networks (CNNs), play a vital role in ensuring network security. However, there is a lack of comprehensive surveys on CNN-based IDS schemes.
APPLIED SCIENCES-BASEL
(2022)
Article
Multidisciplinary Sciences
Jordan C. Koone et al.
Summary: This article introduces a method of printing graphical data into tactile graphics using the lithophane effect. This method allows both blind and sighted individuals to interpret the data through touch or vision.
Article
Ophthalmology
Wolfgang Radner et al.
Summary: This study evaluates the print quality and accuracy of sizing of Landolt ring near vision charts and finds notable deficiencies, suggesting that they are not suitable as reference tests for the calibration of reading charts.
Article
Ophthalmology
Stefano Ferrari et al.
Summary: By analyzing corneal tissues from asymptomatic donors with SARS-CoV-2 RNA detected in postmortem nasopharyngeal swab tests, this study found that the presence of SARS-CoV-2 in corneas is more likely due to prolonged detection of RNA rather than active viral replication. There is a very low risk of infectivity and transmission through corneal transplantation.
BMJ OPEN OPHTHALMOLOGY
(2022)
Article
Computer Science, Interdisciplinary Applications
Tingyi Tian et al.
Summary: This study developed a human-computer interactive health detector for real-time emotion recognition using EEG images. By training the ResNet-50 model, the emotional state of patients induced by environmental stimuli can be accurately analyzed and identified. The proposed model outperforms other algorithm models in terms of accuracy and stability.
COMPUTER METHODS AND PROGRAMS IN BIOMEDICINE
(2022)
Article
Virology
Madeline M. Schwarz et al.
Summary: Infection with Rift Valley fever virus (RVFV) can lead to eye damage, including blurred vision, uveitis, and retinitis. Using a rat model, this study experimentally defined the viral tropism of RVFV in the posterior segment of the eye and characterized the virally-mediated ocular inflammation. The findings provide a foundation for the development of vaccines and treatments to protect against adverse ocular outcomes.
JOURNAL OF VIROLOGY
(2022)
Article
Environmental Sciences
Urszula Zaluska et al.
Summary: People with sensory disabilities face various barriers to travel, which significantly constrain their activities. It is important to provide training for service staff to meet the specific needs of people with different types of disabilities.
INTERNATIONAL JOURNAL OF ENVIRONMENTAL RESEARCH AND PUBLIC HEALTH
(2022)
Article
Mathematical & Computational Biology
Chen Yue et al.
Summary: This study develops an accurate method based on a generative adversarial network (GAN) to address the current issue of micro vessel segmentation in retinal images. By incorporating SE-ResNet, dilated convolution, and attention mechanism, the proposed method achieves higher accuracy in segmenting retinal vessels compared to state-of-the-art methods.
COMPUTATIONAL INTELLIGENCE AND NEUROSCIENCE
(2022)
Article
Chemistry, Multidisciplinary
Roseline Oluwaseun Ogundokun et al.
Summary: With advancements in pose estimation techniques, human posture detection has gained attention in various applications. This study proposes a new three-phase model that integrates CNN transfer learning, image data augmentation, and hyperparameter optimization (HPO) to address the prevalent issues of overfitting and poor performance in neural network models. Experimental results demonstrate that the proposed model achieves good classification performance in human posture detection.
APPLIED SCIENCES-BASEL
(2022)
Article
Computer Science, Hardware & Architecture
G. S. R. Satyanarayana et al.
Summary: This study proposes a vehicle detection technique using convolutional neural networks without the need for labeled vehicle datasets. By training the CNN with road marks as the background, accurate vehicle detection can be performed under various conditions without the requirement of a GPU.
Article
Computer Science, Hardware & Architecture
Cong Hua et al.
Summary: This paper proposes a circular coding technique for visual localization that achieves high accuracy while maintaining real-time performance. By projecting building outlines onto a unit circle to generate database coding circles and extracting vertical corner lines from street-view images for query coding circles, the system simplifies the localization task and enables efficient retrieval from the coding circle database. Experimental results demonstrate the system's fast speed and high accuracy, making it suitable for various applications.
Proceedings Paper
Computer Science, Interdisciplinary Applications
Selina Feitl et al.
Summary: This study proposes the use of surface haptic feedback on an electrostatic display to present tactile floor plans for people with visual impairments. The results show that electrostatic haptic feedback enables individuals to recognize and understand graphic elements like rooms and a floor plan, although it requires more time and cognitive workload compared to a more expensive dot matrix display.
PROCEEDINGS OF THE 15TH INTERNATIONAL CONFERENCE ON PERVASIVE TECHNOLOGIES RELATED TO ASSISTIVE ENVIRONMENTS, PETRA 2022
(2022)
Proceedings Paper
Computer Science, Artificial Intelligence
Danielle Nyakam Nya et al.
Summary: This paper proposes a model-free control method for inventory control in supply chains and demonstrates its efficiency and promising future through concrete numerical simulations and comparative studies, even in the presence of various disturbances.
2022 8TH INTERNATIONAL CONFERENCE ON CONTROL, DECISION AND INFORMATION TECHNOLOGIES (CODIT'22)
(2022)
Article
Computer Science, Theory & Methods
Ahmed Yassine Chakor et al.
Summary: Intelligent monitoring provides a clear understanding of computer network behavior and reduces troubleshooting time caused by hardware or software failures. Automation and predictive maintenance are crucial for successful intelligent monitoring. This paper proposes a machine learning-based method for prediction and decision support to proactively anticipate interventions.
APPLIED COMPUTER SYSTEMS
(2022)
Article
Engineering, Electrical & Electronic
Mahnoosh Tajmirriahi et al.
Summary: The study introduces a novel GAN model (DDFA-GAN) for generating more realistic OCT images, with a focus on Fourier domain similarity in the structural design. By jointly training two discriminators, this model is proven feasible with a limited number of training data.
IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT
(2022)
Article
Computer Science, Information Systems
Kun Xia et al.
Summary: An intelligent blind guide system (IBGS) was introduced in this study, which utilizes speech recognition to achieve traffic light recognition, obstacle avoidance, payment, and navigation functions. The system uses WIFI instead of Bluetooth and incorporates Internet of Things technology for information interconnection. Additionally, a new speech recognition framework was proposed to improve efficiency by suppressing non-critical attention. The system showed promising results in various tests.
Article
Computer Science, Artificial Intelligence
Stefan Feuerriegel et al.
Summary: The article highlights the role of artificial intelligence in supporting managers and discusses the organizational and technical hurdles that need to be addressed. It offers a first step in overcoming these challenges by identifying core factors that affect effective decision delegation to AI.
NATURE MACHINE INTELLIGENCE
(2022)
Article
Mathematical & Computational Biology
Chen Yue et al.
Summary: This paper proposes an improved generative adversarial network model for retinal blood vessel segmentation, which utilizes attention mechanisms and dense connection modules to enhance segmentation performance. The experimental results demonstrate that this method achieves high accuracy in retinal blood vessel pixel segmentation on three public datasets.
MATHEMATICAL BIOSCIENCES AND ENGINEERING
(2022)
Article
Biotechnology & Applied Microbiology
Jin Huang et al.
Summary: In this study, a lightweight architecture called BM-Net was proposed for analyzing breast cancer WSIs. With the use of data augmentation and focal loss techniques, high performance was achieved, showing great potential in breast cancer detection.
BIOENGINEERING-BASEL
(2022)
Article
Engineering, Electrical & Electronic
Yuk-Ming Tang et al.
Summary: In the context of global economic slowdown, smart warehouse and manufacturing management have become important topics, as enterprises can achieve economies of scale and balance supply and demand through big data analytics and optimization algorithms. While many frameworks have been developed for various fields, they mainly focus on product logistics and do not address the specific manufacturing issues in the cyclical industry.
Review
Management
Yasaman Mashayekhy et al.
Summary: This paper explores the impact of IoT technologies on inventory management in supply chains and discusses the trend and potential opportunities of applying IoT to inventory management in the Industry 4.0 era. The study finds that this topic is receiving increasing attention in various industries, with a growing number of articles published in relevant journals. The conclusion suggests that upgrading the supply chain to an integrated supply chain 4.0 is beneficial, as traditional inventory replenishment policies are not responsive enough to new technologies and cannot cope well with IoT systems.
Article
Computer Science, Theory & Methods
Ayoub Skouta et al.
Summary: This paper presents a modified CNN UNet architecture for identifying retinal hemorrhages in fundus images. By training the proposed UNet using GPU and the IDRiD dataset, it effectively segments and detects potential areas of retinal hemorrhages. The experimental results show that this method achieves high sensitivity, specificity, and accuracy in segmenting bleeding areas, reducing the efforts of ophthalmologists significantly.
JOURNAL OF BIG DATA
(2022)
Article
Computer Science, Information Systems
Abdullah A. Asiri et al.
Summary: This study proposes a ResNet-50 feature extractor based on a multilevel deep convolutional neural network for reliable brain tumor image segmentation. By classifying and detecting 2043 MRI patients, better average results are obtained compared to existing methods. This modified architecture could be an important tumor diagnosis system.
CMC-COMPUTERS MATERIALS & CONTINUA
(2022)
Proceedings Paper
Computer Science, Artificial Intelligence
Chaichana Suedumrong et al.
Summary: This research aims to automatically diagnose different stages of diabetic retinopathy using a deep learning neural network approach. The design and implementation of GPU-accelerated deep convolutional neural networks are presented to classify high-resolution retinal images into five stages of the disease, achieving an accuracy of 71.65%.
PROCEEDINGS OF THE FUTURE TECHNOLOGIES CONFERENCE (FTC) 2021, VOL 2
(2022)
Article
Engineering, Electrical & Electronic
Emanuele Cardillo et al.
Summary: This study introduces an electronic travel aid for blind and visually-impaired individuals utilizing a millimeter wave radar and a traditional white-cane, capable of warning users of potential obstacles, distinguishing between human and nonhuman targets, and accurately recognizing human presence in complex environments.
IEEE JOURNAL OF ELECTROMAGNETICS RF AND MICROWAVES IN MEDICINE AND BIOLOGY
(2022)
Article
Construction & Building Technology
P. R. Boyce
Summary: Light can have significant impacts on human health, affecting various aspects from eyestrain and headaches to circadian disruption and sleep enhancement. However, the effects of lighting on health can be positive or negative, and are influenced by a multitude of factors beyond just lighting itself. It is crucial to carefully consider the individual differences and complexities involved in assessing the true benefits of lighting for human health.
LIGHTING RESEARCH & TECHNOLOGY
(2022)
Article
Anthropology
Christian Horn et al.
Summary: Rock art carvings, especially those from the Nordic Bronze Age in southern Scandinavia, contain valuable quantitative data that can help understand social structures and ideologies of the era. By training models to locate and classify image objects, new avenues for research on rock art have been opened. This interdisciplinary undertaking has led to important reflections on archaeology, digital humanities, and artificial intelligence.
JOURNAL OF ARCHAEOLOGICAL METHOD AND THEORY
(2022)
Article
Ergonomics
Marwa M. Zalat et al.
Summary: The study revealed a high prevalence of computer vision syndrome (CVS) among university staff members, with symptoms including dryness, headache, worsening vision, and difficulty focusing on near vision. Females, those with longer work duration, frequent smartphone users, and individuals who spend most of their screen time during the day and night were at higher risk for CVS. Visual ergonomics and preventive measures application, such as correct screen level, regular screen cleaning, appropriate lighting, and use of eye drops, were significantly associated with lower CVS prevalence.
INTERNATIONAL JOURNAL OF OCCUPATIONAL SAFETY AND ERGONOMICS
(2022)
Review
Rehabilitation
Andrius Budrionis et al.
Summary: This paper provides an overview of recent research on smartphone-based electronic travel aid (ETA) systems for visually impaired individuals, which offer potential cost-effective and reasonable solutions for better orientation and navigation. The study found limited interest among researchers in combining haptic interfaces and computer vision capabilities in smartphone-based ETAs, as well as a lack of attempts to employ state-of-the-art computer vision methods and evaluate existing navigation solutions. This highlights a major mismatch between user needs and academic development in the field.
ASSISTIVE TECHNOLOGY
(2022)
Review
Ophthalmology
Faris Almutairi et al.
Summary: Congenital stationary night blindness (CSNB) is a group of rare disorders of the retina, with an increasing number of genes being identified as causes. In Saudi Arabia, recessive mutations in TRPM1 and CABP4 are most common, associated with high myopia or hyperopia. Four novel mutations were discovered, and fundus albipunctatus was described for the first time in two patients, caused by mutations in RDH5 and RPE65. No cases of dominantly inherited CSNB were found.
ACTA OPHTHALMOLOGICA
(2021)
Review
Clinical Neurology
Vejay N. Vakharia et al.
Summary: Surgery for drug-refractory mesial temporal lobe epilepsy can provide long-term seizure remission, but can lead to postoperative visual field defects (VFDs) and driving license ineligibility. Various modifications to surgical approaches have been made to minimize damage to the optic radiation, with advancements in microsurgical techniques introducing minimally invasive alternatives such as laser interstitial thermal therapy and stereotactic radiosurgery. Intraoperative use of optic radiation tractography and interventional MRI can help reduce VFDs and improve outcomes for patients undergoing surgery.
CURRENT OPINION IN NEUROLOGY
(2021)
Article
Computer Science, Hardware & Architecture
Xiang Wang et al.
Summary: This paper provides a comprehensive review on recent deep learning methods for multi-view stereo, categorizing them into depth map based and volumetric based methods. It reviews representative methods in detail and summarizes widely used datasets and metrics for evaluation, while also presenting insightful observations and challenges for future research directions.
Article
Computer Science, Interdisciplinary Applications
Tae Keun Yoo et al.
Summary: The study demonstrated the improvement of applying deep learning (DL) in diagnosing rare ophthalmic diseases through few-shot learning (FSL) using generative adversarial network (GAN). Synthetic images helped extract rare disease features and the DL model showed significant enhancement in OCT diagnosis, outperforming traditional models and reducing diagnosis delay.
MEDICAL & BIOLOGICAL ENGINEERING & COMPUTING
(2021)
Article
Computer Science, Interdisciplinary Applications
Yuhui Ma et al.
Summary: In this paper, a novel medical image quality enhancement method StillGAN is proposed, which outperforms traditional methods and other deep learning methods in experiments. By introducing local structure and illumination constraints to learn both overall characteristics and local details, StillGAN is suitable for different medical image datasets, contributing to improve the efficiency of automated screening, analysis and diagnosis of medical images.
IEEE TRANSACTIONS ON MEDICAL IMAGING
(2021)
Article
Telecommunications
J. Raja et al.
Summary: This study utilizes deep learning techniques to identify glaucoma from fundus images, achieving high accuracy through computer analysis, providing a new approach for early prevention and treatment of glaucoma.
WIRELESS PERSONAL COMMUNICATIONS
(2021)
Article
Engineering, Industrial
Dominic Loske et al.
Summary: This study explores the application of artificial intelligence in retail logistics, utilizing methods such as fuzzy DEA to demonstrate the efficiency advantages of AI-based routing in route planning. The results suggest that the combination of AI with methods like AHP is effective in improving logistics efficiency and can lead to customer-oriented logistics systems.
INTERNATIONAL JOURNAL OF PRODUCTION ECONOMICS
(2021)
Article
Ophthalmology
Eren Ekici et al.
Summary: The severity and prevalence of central visual field loss varied among different glaucomatous optic disc phenotypes. Glaucomatous eyes with focal ischemic (FI) and myopic glaucoma (MY) optic disc phenotypes are more likely to have 10-2 visual field loss.
AMERICAN JOURNAL OF OPHTHALMOLOGY
(2021)
Article
Engineering, Multidisciplinary
Silky Goel et al.
Summary: Diabetes is rapidly growing in India with over 72 million patients. Prolonged diabetes can cause diabetic retinopathy, leading to severe damage to blood vessels and neurons in the eyes. Deep learning techniques have shown high accuracy in classifying the severity of this disease.
MATHEMATICAL PROBLEMS IN ENGINEERING
(2021)
Article
Optics
Jingwen Zhou et al.
Summary: The proposed infrared and visible image fusion method directly uses VGG-19 for feature extraction to avoid the problem of insufficient detail extraction caused by decomposing the source image. Experimental results show that the method performs better in visual quality and noise reduction, while also outperforming some traditional or popular fusion methods in objective evaluation.
Article
Computer Science, Information Systems
Ahsan Bin Tufail et al.
Summary: This study utilized various 3D-CNN architectures based on DL to classify diabetic retinopathy, incorporating artificial data augmentation methods for improved performance. The findings revealed that in binary classification, the best performance was achieved by training with combined augmentation methods, while in multiclass classification, the model trained without augmentation performed the best.
WIRELESS COMMUNICATIONS & MOBILE COMPUTING
(2021)
Article
Engineering, Electrical & Electronic
Andres A. Diaz-Toro et al.
Summary: The article introduces a stereo vision-based system that assists blind people to navigate safely in unfamiliar environments, by sensing the world, segmenting the floor, fusing local 2D grids, creating global occupancy 2D grids, among other methods. Experimental results show high performance and participant acceptance in various conditions.
JOURNAL OF SENSORS
(2021)
Article
Engineering, Biomedical
Xi Xu et al.
Summary: In this study, a Transfer Induced Attention Network (TIA-Net) was proposed for automatic glaucoma detection, leveraging fundus features learned from similar ophthalmic data. By integrating channel-wise attention and maximum mean discrepancy, discriminative features fully characterizing glaucoma-related deep patterns were extracted, leading to improved accuracy and sensitivity.
BIOMEDICAL ENGINEERING ONLINE
(2021)
Article
Mathematical & Computational Biology
Yu Chen et al.
Summary: This study introduces a method based on generative adversarial networks to address the issue of data imbalance in diabetic retinopathy, leading to improved accuracy of DR grading models.
COMPUTATIONAL INTELLIGENCE AND NEUROSCIENCE
(2021)
Article
Computer Science, Theory & Methods
Lucas Rodrigo-Salazar et al.
Summary: The Internet of Things aims to integrate smart devices into daily life, improving the efficiency of systems in various domains. However, the field of assistive technology remains underexplored, presenting opportunities for further research and development.
Article
Computer Science, Hardware & Architecture
Fangyu Shi et al.
Summary: This study established an ASD painting database, analyzed the unique characteristics of paintings by children with ASD, and trained a classifier using extracted features as a potential screening tool for autism spectrum disorder.
Article
Computer Science, Information Systems
Mukhriddin Mukhiddinov et al.
Summary: Individuals with visual impairments face challenges in daily activities, but a smart glass system using computer vision and deep learning techniques can help them move independently. The system includes modules for low-light image enhancement, object recognition with audio feedback, salient object detection, and text-to-speech and tactile graphics.
Article
Chemistry, Analytical
Parvathaneni Naga Srinivasu et al.
Summary: Deep learning models, specifically MobileNet V2 and LSTM, have been proposed for efficient classification of skin diseases, showing higher accuracy and effectiveness compared to other state-of-the-art models. The use of a grey-level co-occurrence matrix helps assess diseased growth progression, with the proposed method outperforming others with over 85% accuracy.
Article
Pediatrics
Hakan Ozturk et al.
Summary: The study found that during the pandemic, excessive use of smartphones, tablets, and computers in children may lead to increased tendency for dry eyes and myopia, as well as axial elongation in the eyes.
JOURNAL OF PEDIATRIC RESEARCH
(2021)
Article
Nida Nasir et al.
Revue d'Intelligence Artificielle
(2021)
Review
Mathematical & Computational Biology
S. Ramaneswaran et al.
Summary: This paper proposes a hybrid Inception v3 XGBoost model for the classification of acute lymphoblastic leukemia (ALL) from microscopic white blood cell images. Experimental results show that the proposed model outperforms other methods and achieves a weighted F1 score of 0.986. Additionally, using an XGBoost classification head instead of a softmax classification head improves classification performance for this dataset across different CNN backbones.
COMPUTATIONAL AND MATHEMATICAL METHODS IN MEDICINE
(2021)
Article
Computer Science, Information Systems
Saman Fouladi et al.
Summary: The novel coronavirus disease (COVID-19) has infected over 141 million people worldwide since April 20, 2021. In this study, models for classifying Chest CT Images of COVID-19 were proposed without the need for pre-trained networks and data augmentation, achieving high accuracies.
COMPUTER COMMUNICATIONS
(2021)
Article
Ophthalmology
Katherine K. Weise et al.
Summary: Pre-participation vision screening was effective in identifying athletes meeting minimum visual acuity requirements, but not in detecting athletes who could benefit from improved vision or those with sport-specific significant eye findings. Most athletes tested had improved visual acuity or significant findings after comprehensive eye examination.
OPTOMETRY AND VISION SCIENCE
(2021)
Article
Ophthalmology
Katherine K. Weise et al.
Summary: The study aimed to determine an objective list of vision-related conditions that may benefit from clear and tinted football helmet visor use in athletes.
OPTOMETRY AND VISION SCIENCE
(2021)
Article
Genetics & Heredity
Dian Hong et al.
Summary: This study constructed an automatic facial recognition model for identifying children with GSs, which showed high accuracy and outperformed human experts in screening for GSs.
ORPHANET JOURNAL OF RARE DISEASES
(2021)
Article
Chemistry, Analytical
Wejdan L. Alyoubi et al.
Summary: Diabetic retinopathy (DR) is a disease caused by diabetes complications, which can lead to blindness. This study introduces a fully automatic diagnosis system using deep learning models to accurately classify DR images and localize affected lesion areas. The combination of two models achieves a high accuracy rate of 89%.
Proceedings Paper
Computer Science, Theory & Methods
Eman Naser Karajah et al.
Summary: Covid-19, a newly discovered coronavirus, has been officially declared a pandemic by the World Health Organization with no specific treatment available. This research aims to use deep learning and transfer learning techniques to diagnose Covid-19 infection faster through X-ray images, achieving a high accuracy of 99.7% by modifying the VGG 16 model.
2021 INTERNATIONAL CONFERENCE ON PROMISING ELECTRONIC TECHNOLOGIES (ICPET 2021)
(2021)
Proceedings Paper
Engineering, Biomedical
Samantha Johnson et al.
Summary: The research team developed TATUM, a mechanism that enables independent communication using tactile ASL, and demonstrated a high recognition rate in validation studies, showing the potential to support deaf and deaf-blind individuals in communicating via visual and tactile ASL.
2021 43RD ANNUAL INTERNATIONAL CONFERENCE OF THE IEEE ENGINEERING IN MEDICINE & BIOLOGY SOCIETY (EMBC)
(2021)
Proceedings Paper
Engineering, Biomedical
Ioan Sima et al.
Summary: The study utilized the Inception V3 deep learning model to classify colorectal polyps and transfer the training weights through transfer learning, achieving satisfactory results compared to other works and human experts.
2021 IEEE 21ST INTERNATIONAL CONFERENCE ON BIOINFORMATICS AND BIOENGINEERING (IEEE BIBE 2021)
(2021)
Article
Computer Science, Information Systems
Ozer Can Devecioglu et al.
Summary: This study introduces a network model Self-ONNs for early detection of glaucoma in fundus images, showing superior performance and significant reduction in computational complexity. Compared to traditional deep Convolutional Neural Networks, Self-ONNs demonstrate better results in glaucoma detection.
Article
Computer Science, Information Systems
Esther Parra-Mora et al.
Summary: Epiretinal membrane (ERM) is an eye disease affecting 7% of the world population, with higher incidence in people over 75. This study shows that deep learning, transfer learning, and optimized neural network structures can achieve high accuracy in automatic detection of ERM using optical coherence tomography imaging.
Article
Computer Science, Information Systems
Khwaja Tahseen Ahmed et al.
Summary: This article presents a method of retrieving visual features from flexible image datasets and obtaining accurate results through a combination of deep convolutional neural networks and content-based image retrieval schemes.
Article
Humanities, Multidisciplinary
Rachel Hutchinson et al.
Summary: The use of audio descriptive guides in museums can enhance memorability for sighted visitors, while also providing important access provisions for blind and partially sighted individuals.
MUSEUM MANAGEMENT AND CURATORSHIP
(2021)
Article
Computer Science, Information Systems
V Sudha et al.
Summary: Diabetic Retinopathy (DR) is an eye disease caused by diabetes, which can lead to blindness. Early detection and automated diagnosis using deep learning techniques can accurately classify and label the severity of DR.
CMC-COMPUTERS MATERIALS & CONTINUA
(2021)
Article
Computer Science, Information Systems
Masum Shah Junayed et al.
Summary: Cataract, a common eye disorder, can lead to vision distortion and blindness if not detected and treated timely. A novel deep neural network named CataractNet is introduced for automatic cataract detection in fundus images, outperforming state-of-the-art approaches with an average accuracy of 99.13%.
Article
Computer Science, Software Engineering
Hu Xinghong et al.
ACM TRANSACTIONS ON GRAPHICS
(2019)
Article
Biochemistry & Molecular Biology
Daniel S. Kermany et al.
Article
Multidisciplinary Sciences
Naoto Shibata et al.
SCIENTIFIC REPORTS
(2018)
Article
Psychology, Biological
Jeffrey D. Wammes et al.
QUARTERLY JOURNAL OF EXPERIMENTAL PSYCHOLOGY
(2016)