Deep Learning and Applications

Efficiently forecasting the demands within a hospital’s Emergency Department (ED) is critical for optimal resource allocation and patient care management. This study focuses on leveraging deep learning techniques to predict various types of ED patient flows, facilitating informed decision-making by ED managers. The rising success of deep learning networks in modeling timeseries data makes them a compelling choice for patient flow forecasting. In this context, we investigate and compare seven deep learning models-Deep Belief Network (DBN), Restricted Boltzmann Machines (RBM), Long Short-Term Memory (LSTM), Gated Recurrent Unit (GRU), combined GRU and Convolutional Neural Networks (CNN-GRU), LSTM-CNN, and Generative Adversarial Network based on Recurrent Neural Networks (GAN-RNN)—to accurately forecast patient flow within a hospital’s emergency department. To enable traffic flow forecasting, a forecaster layer is introduced for each model. Real-world patient flow data spanning different ED services (biology, radiology, scanner, and echography) at Lille regional hospital in France serve as a case study to evaluate these models. Four effectiveness metrics are employed to assess and compare the forecasting methods. The outcomes demonstrate the superior performance of deep learning models in predicting ED patient flows compared to conventional shallow approaches like ridge regression and support vector regression. Significantly, the Deep Belief Network (DBN) stands out, achieving an averaged mean absolute percentage error of approximately 4.097.

It was only up till recent times that computer science and its were sufficient for the application in basic principles. With the in the field of artificial intelligence, the subset Deep learning is towards substantial research and advances, creating diverse We cannot consider deep learning to be an individual approach; it is a collective term which comprises fields from contrasting to be associated with the common spine—Deep learning. Basis for strong approach in deep learning lies in cognizance of the of deep learning. The implementations can be performed vastly in fields through implication of not just one but numerous algorithms achieving our goal. The architecture of deep learning has enhanced in previous years exponentially, and as per demand, the refinement of learning implying that the architecture is dynamic. A few of the most improvised architectures are mentioned below: 30Recurrent neural networks (RNNs) Long short-term memory (LSTM)/gated recurrent unit (GRU) Convolutional neural networks (CNNs) Deep belief networks (DBN) and deep stacking networks (DSNs) Open source software options for deep learning. The area of implementation for deep learning in problem solving is vast. Feed forward networks are very effective as well as recurrent networks can be a good source for the solution of the deep learning problems. The Framework for deep learning can be implemented in software packages for the useful creation of neural network. The framework needs an implementation on a standardized scale and hence needs industrial experts for the framework to be implemented. The entire framework is in simple terms based on the Diagnosis of the problem and further, evaluating the problem. It is evident that the architecture and framework of deep learning is vast and expanding its horizons to every field possible for implementation. Therefore deep learning architecture and framework would be vitalized, with step by step conception. The architecture would be simplified as well as illustrated. All the aforesaid architecture like Recurrent neural network, Long short term memory/gated recurrent unit, convolutional, Deep belief—deep stack as well as open source would be simplified as well as illustrated.

Deep learning, which is originated from an artificial neural network (ANN), is one of the major technologies of today’s smart cybersecurity systems or policies to function in an intelligent manner. Popular deep learning techniques, such as multi-layer perceptron, convolutional neural network, recurrent neural network or long short-term memory, self-organizing map, auto-encoder, restricted Boltzmann machine, deep belief networks, generative adversarial network, deep transfer learning, as well as deep reinforcement learning, or their ensembles and hybrid approaches can be used to intelligently tackle the diverse cybersecurity issues. In this paper, we aim to present a comprehensive overview from the perspective of these neural networks and deep learning techniques according to today’s diverse needs. We also discuss the applicability of these techniques in various cybersecurity tasks such as intrusion detection, identification of malware or botnets, phishing, predicting cyberattacks, e.g. denial of service, fraud detection or cyberanomalies, etc. Finally, we highlight several research issues and future directions within the scope of our study in the field. Overall, the ultimate goal of this paper is to serve as a reference point and guidelines for the academia and professionals in the cyber industries, especially from the deep learning point of view.

Effective forecasting of key features in hospital emergency department: Hybrid deep learning-driven methods

Comprehensive Study for Breast Cancer Using Deep Learning and Traditional Machine Learning

Deep learning (DL) architectures such as deep neural networks (DNN), deep belief networks (DBN), recurrent neural networks(RNN) and convolutional neural networks (CNN) have been applied to applications such as computer vision, speech recognition, natural language processing, audio recognition, social network filtering, machine translation, bioin-formatics, drug design, medical image analysis, material inspection and board game programs, in which has comparable performance than human experts. With the growing interest and research in the area of artificial neural network, deep neural network enable computers to get trained for error-free diagnosis to diseases like epilepsy. In literature, researchers carried out many mathematical models for pre-processing of EEG data and classification between seizure and seizure free signals or different 166types of network disorders. The introduction of various algorithms like machine learning deep learning, etc., in artificial intelligence, aids to classify the data with or without pre-processing and two class system. It is important to try multi-class time series classification of various brain activities (tumour, network disorders) using the sophisticated algorithms. In this chapter, different deep learning algorithms for multiclass, time series classification of different electrical activities in brain are discussed. The main focus is on the application of different RNN models in seizure classification of Electroencephalogram (EEG) signals. It is very important to interpret the 1D EEG signals and classify among different activities of brain for various diagnostic purpose. The fully interconnected hidden configuration of recurrent neural network (RNN) makes the model very dominant which enable to discover temporal correlations between far away events in the data. The training of RNN architecture when used in deep network is challenging because of vanishing/exploding gradient in deeper layer. This paper aims to perform multiclass time series classification of EEG signal using three different RNN techniques; simple Recurrent Neural Network, Long-Short Term Memory (LSTM) and GRUs. A comparative study between RNNs is done in terms of configuration, time taken and accuracy for EEG signals acquired from people having different pathological and physiological brain states. The accuracy and time taken for multilayer recurrent neural networks are determined for classification of EEG for five different classes using three different types of RNN networks, for 1 to 1024 units with 100 epochs and 5 different layers of 32 cells with 300 epochs, with a learning rate of 0.01. It has been observed that the number of layers increases the time complexity and provides constant accuracy for more than three layers. Further, it can be extended for the accuracy and time consumption for different batch sizes with different epochs to fix a proper network without over fitting the network.

In the recent era Deep learning is showing exemplary results in many application areas. Researches from different disciplines have incorporated deep learning in to their research to solve different interdisciplinary problems. Deep learning applications areas include Speech recognition, Natural Language Processing, Computer vision, Networking, Healthcare, IoT, Robotics, Agriculture, Remote sensing and many other areas. Thus, this paper contributes a review on various deep learning approaches which include Convolution Neural Network (CNN), Deep Neural Network (DNN), Auto-Encoder (AE), Recurrent Neural Network (RNN) enclosed with Gated Recurrent Units (GRU), Long Short-Term Memory (LSTM) and ConvLSTM, Deep Reinforcement Learning (DRL), Generative based Adversarial Network (GAN) and Deep Belief Network (DBN). For extraction of the most important features of an image various data extraction procedures have designed. CNN's tremendous learning capacity is based on the utilisation of several feature extraction stages that continuously learn from data. Some of the popular CNN architectures name as LeNet, AlexNet, ZFNet / Clarifai, VGGNET, GoogLeNet, ResNet, DenseNet, FractalNet and CapsuleNet. Recent works extended to a combination of two CNN models like Inception, ResNetV2 to attained different existing approaches consequences in deep learning. Identification of the significant objects, their parameters and relationships of an object images are required for image captioning. Syntactically and semantically correct sentences need to be generated. The intricacies and problems of picture captioning may be handled using deep learning approaches. A comparison between these models was also presented. We also discussed about different standard datasets which are utilized for executing and estimating the deep learning approaches.

Transportation Cyber-Physical Systems (TCPS) integrate the interaction between the connected transportation infrastructure, users, and computing and communication services to support the mobility of people and goods. TCPS is supported by many emerging technologies, among which Deep Learning (DL)-supported technology is a notable one. DL models closely mimic the functionality of human brains, and it can learn the relationship between inputs and outputs through data observations, thus facilitating different information extraction or decision-making functionalities of TCPS. In recent times, the availability of vast amounts of heterogeneous data from TCPS and the introduction of large-scale computing hardware have enabled the different applications of DL models. This chapter discusses early DL models, such as restricted Boltzmann machine, deep belief network, and deep Boltzmann machine, along with the recent ones including multilayer perceptron, autoencoder, convolutional neural network, recurrent neural networks, and deep reinforcement learning. The DL model development and testing considerations in the functioning of TCPS applications are discussed while pointing out the available hardware and software frameworks. Different types of cyberattacks that can occur on DL and the strategies protecting the DL against such cyberattacks are also presented. This chapter concludes with various applications of DL models in TCPS.

Algorithms based on deep learning are taking off as solutions to several problems in numerous fields. Numerous deep learning architectures like convolutional neural networks, recurrent neural networks, deep belief networks, artificial neural networks and deep neural networks have been used in fields such as audio recognition, machine vision, processing of natural language, recognition of speech, social network filtering, computer vision, translating languages and bioinformatics. One such application of deep learning architectures is object detection. Object detection in satellite and aerial imagery undertakes an important role in a long list of applications and has allured a lot of attention which resulted in exceptional progress in recent years. It has abundant uses in various fields such as surveillance, military, national security, urban planning, environmental applications, traffic control and forestry. Several issues like varying size, background and orientation of the target object make the automatic detection of target in satellite imagery a challenging issue. Recent research in this field has given better speed and accuracy to the detection of the target object. With the rise in the data being generated by the Earth Observation programs, it is becoming much more imperative to find better and efficient ways to exploit this data by improving upon the existing Deep Learning research. In this paper, we bring forth a wide-ranging analysis of the recent advancements in methods of object detection based on deep learning to recognize the objects in satellite imagery.KeywordsDeep learningObject detectionSatellite imageryRemote sensingYou Only Look Once (YOLO)You Only Look Twice (YOLT)Single Shot Detector (SSD)Convolutional Neural Networks (CNN)Faster Region-based Convolutional Neural Networks (Faster RCNN)Hybrid Deep Convolutional Neural Network (HDNN)Satellite Imagery Multiscale Rapid Detection with Windowed Networks (SIMRDWN)

In recent decades, deep learning (DL) has become a rapidly growing research direction, redefining the state-of-the-art performances in a wide range of techniques, such as object detection and speech recognition. In the aircraft design, dynamics, and control field, many works hinge on the information-rich data-driven approach, which includes the fusion-based prognostic and health management, the airliner's flight safety monitoring, intelligent sensing, and flight control systems development. While DL provides great potentials to solve these data-driven problems, a systematic review and discussion as to how the DL has been/can be used for these problems are still missing in relation to the rapidly developing and widely used DL techniques. In this article, we aim to address this urgent issue to provide a timely overview of the state-of-the-art for applying DL to the aircraft design, dynamics, and control field. In particular, we briefly introduce five representative DL methods, i.e., deep neural network, deep autoencoder, deep belief network, convolutional neural network, and recurrent neural network. Mathematical definitions for each method are presented, and illustrative applications are also discussed. We then review the existing DL-based works that have appeared in the aircraft design, dynamics, and control field. The review efforts are divided into two major groups, i.e., the own-ship aircraft modeling, wherein the works have been/can be implemented online for the aircraft design/dynamics/control, and other airplanes research works, wherein DL-based schemes provide offline monitoring of the aircraft operation. We then summarize the data sources and DL architectures. Referring to the experiences of DL research works/techniques development in other related fields, future opportunities, challenges, and potential solutions for implementing DL in the aircraft design, dynamics, and control field are also discussed.

Machine learning allows us to obtain useful information from raw data for quick and efficient solving of complex data-intensive tasks. As a sub-sector of artificial intelligence, machine learning explores study and construction of algorithms that make data-based predictions and are capable of shaping the learning process accordingly - such algorithms are far more effective than the technique of using strictly static program instructions. Machine learning algorithms are used in a wide variety of computational tasks, where it is difficult or infeasible to design and implement an explicit algorithm with decent performance. Deep learning is a branch of machine learning, based on a set of algorithms that model high-level abstractions in data by applying a depth graph with multiple processing layers, built from several linear or non-linear transformations. Research in this area is aimed at getting better representations and creating models for training on these representations from large-scale unlabeled data. Deep learning architectures such as deep neural networks, convolutional deep neural networks, deep belief networks and recurrent neural networks have been applied to fields including computer vision, speech recognition, natural language processing, sound recognition, and bioinformatics where they have produced cutting-edge results in a variety of cases. This article covers the concepts of machine learning, deep learning and image recognition. A specific example (with step-by-step explanation) of using deep learning for building an image recognition system with a neural network architecture is given. The resulting system provides ample opportunities to automate the technological processes and increase their efficiency. The concept of the system can be adapted to the tasks of a new type.

Big data are information assets characterized by high volume, velocity, variety, and veracity. Large, multilevel, and integrated datasets offer the promise of unlocking novel insights and accelerating breakthroughs. According to an increasing interest in artificial intelligence around the world, deep learning has attracted a great deal of public attention. Deep learning techniques increase learning capacity and provide a decision support system at scales that are transforming the future of health care. Every day, deep learning algorithms are used broadly across different industries. The latest advances in deep learning technologies provide new effective paradigms to obtain end-to-end learning models from intricate data. This chapter explains what deep learning is and why it is so important. We provide several examples where deep learning techniques find applications and what kind of problems can be solved using deep learning in medical imaging. This chapter also introduces a particular type of deep learning algorithm, including autoencoders, restricted Boltzmann machines, deep belief network, recurrent neural networks, and convolutional neural networks with practical variant case studies, which are at the basis of deep learning. KeywordsArtificial intelligenceDeep learning techniquesDecision support systemMedical imaging

Deep learning has gained increasing attention in automatic speech recognition, computer vision, natural language processing, drug discovery toxicology, audio recognition, bioinformatics, and automatic driving of vehicles due to its potential benefits such as feature extraction and data classification problems. It is an evolving research domain in diverse applications which increases the overall potential cost benefits for maintenance and refurbishes activities. Deep learning is an ubiquitous technology, and the machine learning algorithms assist in modelling high-level abstract view of data by means of processing layers which encompasses complex structures. The software tools in this area provide finer representations from massive volume of unlabeled data. The software in deep learning identifies patterns in the form of digital representation such as images, data, sound, etc. According to Gartner’s hype cycle, deep learning is on “permanent peak” since 2015, and HFS research survey states that 86% of respondents believe that the technology makes a huge business impact in the industry sector. It is a rapid growing domain with a set of powerful techniques and huge amount of computational power, where machine identifies objects and translates the recognized speech in real time. The main key aspects of deep learning are (i) models comprising several stages or layers of nonlinear information processing and (ii) methodologies for supervised or unsupervised learning for feature extraction at an abstraction level. The significant reasons for its popularity are the increased size of training data set, chip processing capabilities, and the recent advancement in signal processing and machine learning research. The deep learning techniques are effectually exploiting intricate nonlinear functions to acquire hierarchical and distributed feature representations in the perspective of utilizing both labeled and unlabeled data. The different deep learning methods and architectures such as convolutional deep neural network (CDNN), deep neural network (DNN), recurrent neural network (RNN), deep belief network (DBN), artificial neural network (ANN), and long short-term memory (LSTM) are discussed.

Objective: Electroencephalography (EEG) is very crucial for understanding the dynamic healthy and pathological complex processes in the brain. However, the manual analysis of the EEG signal is very complex, time-consuming, and depends on the expertise and experience of the users. Hence, nowadays research is conducted on automated EEG signal analysis using artificial intelligence and computer-aided technologies. This would allow fast and highly accurate results. The goal of this paper is to provide an extensive review of the EEG signal analysis using deep learning (DL).Methods: This systematic literature review of EEG processing using Deep Learning (DL) was achieved on Web of Science, PubMed, and Science Direct databases, resulting in 403 identified papers. All collected studies were analyzed based on main disorders studied, type of tasks performed, data source, stages of analysis, and DL architecture.Results: DL in EEG processing is promising in various research applications. It covered the common neurological disorders diagnosis such as epilepsy, movement disorder, depression, schizophrenia, autism, alcohol use, attention, memory, sleep, pain, etc. The main tasks covered by the included studies are detection and classification. The average range of data sources utilized by the included studies is 127 subjects with an EEG recording a total duration of 458 hours. In fact, we identified the use of a plethora of DL architecture for EEG analysis. 57% of papers used Convolutional Neural Networks (CNNs), whereas Recurrent Neural Networks (RNNs) were the architecture choice of about 12% of papers. Combinations of CNNs and Long Short-Term Memory (LSTM) were used in 13% of studies. Generative Adversarial Networks (GAN) and Autoencoder (AEs) were used in 5% and 4% of papers respectively. Restricted Boltzmann Machine (RBMs), Deep Belief Networks (DBNs), and other hybrid architectures appeared in 1% of studies.Conclusion: This systematic review showed that DL is a powerful tool to process, analyze, and interpret EEG data without requiring extraction steps. These intelligent models can allow self-learning from the training data. On the other hand, DL models need a lot of data to learn, while suffering from a lack of confidence due to their black-box nature. Hence, studies on transfer learning and Explainable Artificial Intelligence (XAI) could help in solving such issues. Big Data, Cloud Computing, the Internet of Things (IoT), and closed-loop technology can also help DL-based systems in achieving fast, and accurate processing of EEG recordings.

Deep learning uses artificial neural networks to recognize patterns and learn from them to make decisions. Deep learning is a type of machine learning that uses artificial neural networks to mimic the human brain. It uses machine learning methods such as supervised, semi-supervised, or unsupervised learning strategies to learn automatically in deep architectures and has gained much popularity due to its superior ability to learn from huge amounts of data. It was found that deep learning approaches can be used for big data analysis successfully. Applications include virtual assistants such as Alexa and Siri, facial recognition, personalization, natural language processing, autonomous cars, automatic handwriting generation, news aggregation, the colorization of black and white images, the addition of sound to silent films, pixel restoration, and deep dreaming. As a review, this paper aims to categorically cover several widely used deep learning algorithms along with their architectures and their practical applications: backpropagation, autoencoders, variational autoencoders, restricted Boltzmann machines, deep belief networks, convolutional neural networks, recurrent neural networks, generative adversarial networks, capsnets, transformer, embeddings from language models, bidirectional encoder representations from transformers, and attention in natural language processing. In addition, challenges of deep learning are also presented in this paper, such as AutoML-Zero, neural architecture search, evolutionary deep learning, and others. The pros and cons of these algorithms and their applications in healthcare are explored, alongside the future direction of this domain. This paper presents a review and a checkpoint to systemize the popular algorithms and to encourage further innovation regarding their applications. For new researchers in the field of deep learning, this review can help them to obtain many details about the advantages, disadvantages, applications, and working mechanisms of a number of deep learning algorithms. In addition, we introduce detailed information on how to apply several deep learning algorithms in healthcare, such as in relation to the COVID-19 pandemic. By presenting many challenges of deep learning in one section, we hope to increase awareness of these challenges, and how they can be dealt with. This could also motivate researchers to find solutions for these challenges.