Sign language recognition method based on deep learning

Artificial Intelligence in the American Healthcare Industry: Looking Forward to 2030

Artificial Intelligence (AI) has become one of the most transformative foundational technologies of our time cutting across industry and function, while machine learning has become its engine of maturation. The rapid advancement of AI applications in different domains including computer vision, speech recognition, and natural language processing (NLP) has led to a need for highly accurate, efficient, scalable, and deployable systems in the real world. The paper discusses advances needed to advance AI applicability in these three major areas, especially using advanced machine-learning technologies to improve AI efficiency in terms of the performance, energy requirements, and computational scalability of AI systems In order to seek out the accuracy and speed of the computer vision systems, the study’s investigation studies the works related to the modernization of machine learning techniques namely deep learning, reinforcement learning, and transfer learning. Such advances are critical for real-time applications including ampere nous driving, facial recognizance, and medical image diagnosis. In parallel, studies on the architecture optimization for speech recognition models are being conducted to enhance feature extraction, reduce computation and latency while maintaining performance. Such improvements are crucial for voice assistants, transcription services and multilingual communication tools. For NLP, these translations shed light on the latest in transformer-based architectures, better pre-trained language models, and methods for reducing the size and inference time of models, all so that these AI services can be reasonable in chatbots, content summarization, and sentiment analysis, etc. This work presents a detailed insight into how machine learning techniques can offer massive energy savings that can be derived from the analysis of state-of-the-art architectures, finding bottlenecks, and comparing performance. In addition, it underscores recent advances in lightweight models, distributed training, model compression and hardware-aware optimization engage computational burden while preserving or improving predictive performance. This brings these approaches together to maintain that AI technologies can be consolidated more sustainably across edge devices, cloud platforms, and embedded systems.

Deep learning techniques are proved to be very effective in various domains, such as computer vision, pattern recognition, and natural language processing. With the rapid development of this technology, more and more deep models have been used in the recommendation system. However, only few attempts have been made in social-based recommender systems. This paper focuses on this issue and explores the use of deep neural networks for learning the interaction function from data. A novel recommendation model called SNHF (Social Neural Hybrid Filtering) is proposed. It combines collaborative and content-based filtering with social information in an architecture based on both models: (1) Generalized Matrix Factorization (GMF); and (2) Hybrid Multilayer Perceptron (HybMLP). Extensive experiments on two real-world datasets show that SNHF significantly outperforms state-of-the-art baselines and related work, especially in the cold start situation.

The therapeutic application of medicinal plants, or phytomedicine, has a long history and is still vital to the world's healthcare system. The field confronts obstacles despite its potential, such as uneven quality, poor clinical validation, safety issues, and variability in the content of bioactive compounds. These problems can be revolutionized by recent developments in artificial intelligence (AI), which offer tools for quick bioactive compound discovery, precise plant identification and authentication, pharmacological activity prediction modeling, and formulation optimization for individualized treatment. AI technologies, such as computer vision, machine learning, deep learning, and natural language processing, make it possible to combine traditional knowledge with contemporary pharmacological research, improving the safety, efficacy, and repeatability of herbal remedies. The main uses of AI in phytomedicine are highlighted in this paper, including quality control, toxicity and dosage prediction, integration with databases used in traditional medicine, and the identification of new herbal chemicals. Prominent case studies show how AI-driven methods might improve precision herbal therapy. Additionally, the paper highlights the significance of AI in transforming phytomedicine into a data-driven, scientifically verified, and customized therapeutic platform while discussing present limitations, ethical issues, and future potential.

Artificial intelligence-based suicide prevention and prediction: A systematic review (2019–2023)

Efficiency, lightweighting, and electrification endeavors have necessitated an increasing use of next-generation materials, dissimilar material joints, and unconventional stackups in automotive body-in-white assembly and resistance spot welding (RSW). These changes have brought forth novel engineering challenges which reduce the applicability of open-loop RSW control based on institutional knowledge and common RSW practices. Consequently, monitoring the RSW process, evaluating and tracking its outcomes, and leveraging massive data via analytics to derive feedback for closed-loop control have never been so crucial in automotive manufacturing. Despite these industry needs, statistical destructive and nondestructive tests still remain common. In recent years, advancements in computing, artificial intelligence (AI), and most notably deep learning (DL), have enabled the interpretation of data from nondestructive evaluation (NDE) at throughput and performance levels required by the automotive industry. DL continues to establish the state of the art in many computer vision and natural language processing applications, and consequently it continues to be increasingly studied and applied in NDE. In this work, we developed semantic segmentation approaches for post-process and in-process interpretation of ultrasonic data from RSW process monitoring. Algorithms were then developed that estimate measurements from the segmentation mask data leveraging prior knowledge about the weld. For both AI systems, correlation analyses were conducted between ultrasonically-measurable properties and physical measurements from cross-sectioning of dissimilar-material 2- and 3-sheet RSWs. This work demonstrates that AI can enhance, facilitate automation of, and provide new value to well-established NDE methodologies. AI and other smart manufacturing technologies enable new NDE use cases, enable innovative vehicle body designs, facilitate the use of novel joining technologies, improve vehicle safety, improve manufacturing efficiency, and reduce waste – consequently, they are poised to make NDE 4.0 a reality in automotive manufacturing.

With the rapid development of computer, computer vision technology is also making rapid progress. In this paper, deep neural network algorithm is used to improve the technology of computer vision, improve the visual effect, and at the same time, innovative algorithm structure, improve the identification of pattern recognition. Computer vision and pattern recognition is a very cutting edge technology that has given people very advanced tools for vision and recognition. At present, the technology can control more accurate image resolution and improve the ability of pattern recognition. This article mainly explains the process of using deep neural network algorithm to improve computer vision and pattern recognition from the internal mechanism, and reveals the working principle and internal mechanism of computer vision and pattern recognition technology application. Data analysis proves that the pattern recognition application established by deep neural network algorithm performs very well in the field of vision and pattern recognition.KeywordsDeep neural network algorithmComputer visionPattern recognitionTechnical application

Artificial Neural networks have been proved most efficient in Deep Learning mainly because of large number of datasets it can handle. The most widely used is the Convolutional Neural Network (CNN). It has been proved useful for computer vision, pattern recognition and Natural Language Processing (NLP). CNN is so vastly used, as, unlike traditional Neural Nets, it reduces number of parameters and focus more on domain specific features. There are various CNN architectures proposed, such as LeNet, AlexNet, GoogleNet. In this paper, we talk about structure of CNN and all the models of CNN which are proposed till date.

AI-based visual speech recognition towards realistic avatars and lip-reading applications in the metaverse

This comprehensive review paper delves into the intricate landscape of sign language recognition (SLR) technologies, addressing the pressing need for effective communication solutions for the global population grappling with hearing impairments. With a meticulous examination of the past two decades of SLR research, the paper categorically explores isolated and continuous manual and non-manual SLR, shedding light on diverse modalities and datasets. The review not only emphasizes the significance of sign language as a fully developed language but also underscores the potential of advanced technologies, such as deep learning and unsupervised learning, in revolutionizing SLR. The inclusion of diverse research projects, ranging from real-time Indian Sign Language conversion to text to the development of deep learning models for American Sign Language recognition, adds practical depth to the exploration. The paper's contributions extend to presenting a tabular format for enhanced comprehension, addressing key facets like modality and sign language types, and offering insightful guidance for the trajectory of this burgeoning field. Overall, this review serves as a valuable and timely resource, advocating for continued innovation to foster a more inclusive and interconnected world for individuals with hearing impairments.

The history of artificial intelligence (AI), from its conception to the creation of general AI (Gen-AI), is a fascinating story of human inventiveness, technical growth, and philosophical research. This article examines the historical milestones, major inventions, and transformational concepts that have influenced AI's trajectory. Beginning with early symbolic AI and rule-based systems, it investigates the shift to machine learning, highlighting discoveries in neural networks and deep learning that transformed disciplines such as computer vision and natural language processing. The introduction of generative models, such as GANs and VAEs, resulted in a considerable increase in AI capabilities, paving the path for Gen-AI. Unlike narrow AI, Gen-AI strives to imitate human-like intelligence and adaptability across a wide range of jobs, bringing serious ethical and philosophical concerns. This essay also looks at the current state of Gen-AI, its problems, and possible applications in healthcare, education, finance, and other areas. It finishes by picturing a future in which human-machine collaboration and ethical AI development are prioritized, highlighting the need of continual learning and responsible innovation.

Sign language recognition is an emerging field of computer vision and pattern recognition that enables deaf individuals to communicate with hearing people. With deep learning and advancements in computer vision, computers can automatically recognize hand shapes and motion dynamics, facial expressions, and body postures in sign language. Even though deep neural networks (DNNs) perform well on vision tasks, they consume excessive computational resources and have a large memory footprint. Thus, DNNs cannot be deployed on edge devices. Quantization is one of the most useful model compression techniques for overcoming the limitations of deploying these large models on devices with limited computing resources. This paper proposes an improved post-training quantization method, called the hybrid percentile-based outlier aware dynamic range quantization approach, which sets clipping thresholds layerwise to minimize quantization error. This hybrid quantization approach maintains FP32 precision for input weights, the final output classification layer, and activations, while quantizing only the weights in intermediate layers to achieve computational efficiency. Itworks well in layers with heavy-tailed weight distribution. Experimentation is conducted using benchmark deep learning models, including VGG16, ResNet50, EfficientNetB0, and SqueezeNet. The experiments use a widely used standard ASL Alphabet dataset. The proposed method is robust to outliers and achieves up to 10−11× reductions in model size and 11−122× inference speedups, with less than 1% accuracy drop relative to baseline models with 32-bit floating-point precisions. The findings show that the proposedhybrid quantization method accelerates the inference speed of deep learning models while maintaining optimal performance.

This scoping review explores the potential of Artificial Intelligence (AI) to improve pilot training, a critical component in maintaining aviation safety. Traditional training methods, while foundational, are limited by cost, access to real-world scenarios, and adaptability to individual learning needs (AI Labs, n.d.). AI technologies, such as machine learning, computer vision, and natural language processing, offer innovative tools to enhance the cognitive and decision-making capabilities of pilots, enabling personalized learning, real-time feedback, and adaptive simulation environments (Kabashkin et al., 2023; Eddins, 2024). A comprehensive review of academic databases was conducted using specific keywords related to AI and pilot performance. Studies included in this review focused on AI applications that directly impact pilot training efficacy. Data were extracted and qualitatively analyzed to assess the scope and quality of findings across selected literature.Key findings reveal a shift from instructor-led and simulator-based training to more dynamic, AI-enhanced platforms that simulate real-time stressors and decision-making contexts. Additionally, AI supports safety outcomes by powering predictive analytics to identify potential risks before they occur (Mylrea & Robinson, 2023). Despite these advances, challenges remain. Ethical issues such as algorithmic bias and data privacy need to be addressed. Technical limitations, including the demand for high-fidelity simulation and AI reliability, also hinder widespread adoption. Human factors, notably pilot trust in AI systems, significantly affect implementation success (Korentsides et al., 2024). This review concludes that while AI presents transformative opportunities for pilot training, its effective integration requires addressing existing limitations, updating regulatory frameworks, and exploring synergies with technologies like virtual and augmented reality. Future research should emphasize collaborative AI systems and ethical safeguards to ensure both technical effectiveness and pilot acceptance.

This introductory article opens the “Applications of AI in the Semiconductor Industry” section by giving a holistic overview of the development of artificial intelligence (AI) technologies applied to the industry. Historically, the semiconductor industry has utilised complex automation for many tasks and areas, especially in repetitive work and uniform processes. The high need for flexibility in manufacturing, increased diversification of products, complexity, and demand for more autonomous operations, including human-machine interaction, have led to a strong push towards using AI technologies in semiconductor manufacturing. AI technologies are applied in semiconductor product development, digitised product definition (DPD), knowledge management system for risk assessment and root cause analysis, image recognition for inspection and defect classification in front end (FE) and back end (BE) applications for anomaly detection in process chains. Deep learning (DL) and Machine Learning (ML) techniques have given a new stimulus to semiconductor industry research to address the unique challenges for semiconductor manufacturing as the technologies nods are evolving and the number of process parameters to be controlled is increasing. In the end, 106the article introduces the four contributions to this section, highlighting the use of AI, computer vision, neural networks (NNs) in various use cases in semiconductor manufacturing processes.

In today’s digital environment, the pattern is everywhere. It is present in many aspects of our daily life. Algorithms can be used to detect or mathematically observe a pattern physically. Vector feature values represent a pattern in the digital world. With the advent of artificial intelligence (AI) techniques in the recent era, there are so many machine learning (ML) and deep learning (DL) models that have been developed. ML is the branch of AI, which can perform tasks like data analysis, analytical model building, and classification without being explicitly programmed. DL is the subset of ML in AI. In DL, mostly artificial neural networks (ANNs) are utilized. It has the capability of work based upon unsupervised learning from the data that is unlabeled and unstructured. Using these DL and ML models, extract the meaningful features from the given image or video is known as pattern recognition (PR). PR is used in many engineering applications such as computer vision, natural language processing, character recognition, robotics, speech recognition, and so on. It is also used in many medical image processing applications and telemedicine. The present work discussed the pattern recognition problem and its various stages in detail. In addition to that, the application of deep learning and machine learning in pattern recognition is also explained briefly.