RSS 1.0Book Chapters
Recent Submissions
-
Walk Through Event Stream Processing Architecture, Use Cases and Frameworks SurveyNowadays events stream processing is one of the top demanding field(s) because of the business urgent need for ongoing real time analytics & decisions. Most business domains avail huge amount of data aiming to make use of each data point efficiently. Corporate(s) have a cloud of events vary from internal business transactions, social media feeds, IoT devices logs, ... etc. In this paper we would discuss state of the art event stream processing technologies using cloud of events by summarizing event stream processing definition, data flow architectures, common use cases, frameworks and architecture best practice. A final comparison is given for best practice and technology use based on data type.
-
The era of advanced machine learning and deep learning algorithms for malware detectionSoftware has been the essential element to computers in today's digital era. Unfortunately, it has experienced challenges from various types of malware, which are designed for sabotage, criminal money-making, and information theft. To protect the gadgets from malware, numerous malware detection algorithms have been proposed. In the olden days there were shallow learning algorithms, and in recent years there are deep learning algorithms. With the availability of big data for training of model and affordable and high-performance computing services, deep learning has demonstrated its superiority in many smart city applications, in terms of accuracy, error rate, etc. This chapter intends to conduct a systematic review on the latest development of deep learning algorithms for malware detection. Some future research directions are suggested for further exploration.
-
Brain stroke detection from computed tomography images using deep learning algorithmsStroke is one of the common causes of death worldwide. Stroke is the inability of a focus to be fed in the brain due to clogged or bleeding of the vessels feeding the brain. Because early stroke treatment and diagnosis are related to a favorable patient outcome, time is a critical aspect of successful stroke treatment. In this chapter, we examine the stroke classification from Brain Stroke CT Dataset, with deep learning architectures. In the experimental study, a total of 2501 brain stroke computed tomography (CT) images were used for testing and training. For this purpose, numerus widely known pretrained convolutional neural networks (CNNs) such as GoogleNet, AlexNet, VGG-16, VGG-19, and Residual CNN were used to classify brain stroke CT images as normal and as stroke. Several performance metrics such as accuracy (ACC), specificity (SPE), sensitivity (SEN), and F-score are used to evaluate the performances of the classifier. The best classification results are achieved by VGG-19 with ACC 97.06%, SEN 97.41%, SPE 96.49%, and F-score 96.95%.
-
Towards Sustainable Smart City via Resilient Internet of ThingsEvery day, 2.5 quintillion bytes of data are generated which is an unimaginable figure to human beings and even machine. To achieve the global smart city vision, automation, resilience, and sustainable development are crucial elements. This chapter focuses on resilient Internet of Things that links individuals and sensing devices which forms the foundation of data collection and provides ground truth of information. With the tremendous growth of primary data volumes and diversity in every domain, they have played an ever more crucial role in enabling researchers and enterprises to formulate processing and analysis methods to extract latent information from multiple data resources and to leverage a broad range of data management and analytics platforms. We have been witnessed the successful technology story of artificial intelligence in various applications. However, resilient and sustainable development has not yet fully integrated into artificial intelligence applications. It requires automated update and improvement of trained machine learning model with the ever-increasing data. This chapter is organised as follows. Firstly, a systematic review of the existing works of resilience and sustainability for smart city is presented. This is followed by a comparison of IoT solutions in software and hardware perspective. Various future research directions and conceptual study of smart city application are discussed.
-
Breast cancer detection from mammograms using artificial intelligenceBreast cancer is one of the fastest-growing forms of cancer in the world today. Breast cancer is primarily found in women, and its frequency has been gaining significantly in the last few years. The key to tackle the rising cases of breast cancer is early detection. Many studies have shown that early detection significantly reduces the mortality rate of those affected. Machine learning and deep learning techniques have been adopted in the present scenario to help detect breast cancer in an early stage. Deep learning models such as the convolutional neural networks (CNNs) are suited explicitly to image data and overcome the drawbacks of machine learning models. To improve upon conventional approaches, we apply deep CNNs for automatic feature extraction and classifier building. In this chapter, we have demonstrated thoroughly the use of deep learning models through transfer learning, deep feature extraction, and machine learning models. Computer-aided detection or diagnosis systems have recently been developed to help health-care professionals increase diagnosis accuracy. This chapter presents early breast cancer detection from mammograms using artificial intelligence (AI). Various models have been presented along with an in-depth comparative analysis of the different state-of-the-art architectures, custom CNN networks, and classifiers trained on features extracted from pretrained networks. Our findings have indicated that deep learning models can achieve training accuracies of up to 99%, while both validation and test accuracies up to 96%. We conclude by suggesting various improvements that could be made to existing architectures and how AI techniques could help further improve and help in the early detection of breast cancer.
-
Magnetic resonance imagining-based automated brain tumor detection using deep learning techniquesbrain tumor refers to an accumulation or growth of anomalous cells in the brain. The mass can either be benign (noncancerous) or malignant (cancerous). Cancerous brain tumors are the source of morbidity for which diagnosis and treatment require extensive resource allocation, experienced and skilled radiologists and doctors, and sophisticated diagnostic and therapeutic technology. Early detection of brain tumors is the basic requirement for the treatment of the patient. Manual detection of the brain tumor is an invasive process and hence is extremely risky. So, advancements in medical imaging techniques, such as magnetic resonance imagining (MRI), have proved to be an important tool in the early detection of brain tumors. Even with the advancements in medical imaging, it remains a very exigent task for radiologists. In many scenarios, the unavailability of a skilled radiologist or doctor can lead to improper diagnosis of the patient. Artificial intelligence and computer vision have successfully been used to achieve human-like accuracy in various image classification tasks. In this chapter, we propose various algorithms to detect the presence of brain tumors in MRI scans of the brain. We use various state-of-the-art convolutional neural networks and apply transfer learning to achieve this goal. We have also used machine learning algorithms that were trained on the embedding of the MRI scans, which were acquired through deep feature extraction, to detect the presence of abnormalities. Moreover, this chapter compares the performance of various models and techniques for automatic brain tumor detection using deep learning.
-
Breast tumor detection in ultrasound images using artificial intelligenceLeveraging artificial intelligence (AI) for categorizing breast tumors as malignant or benign from breast ultrasound images can provide an effective and relatively low-cost method for the diagnosis of breast cancer. Presently, many machine learning (ML) and deep learning (DL) algorithms have been used for early-stage breast cancer detection. AI algorithms have shown promising results in breast cancer detection tasks. The use of deep convolutional neural network approaches has provided solutions for the efficient analysis of breast ultrasound images. Convolutional neural network (CNN) models analyze the image data in multiple layers and extract features which helps in better feature extractions and better performance in comparison to the conventional ML algorithms. Apart from conventional learning algorithms, we use the transfer learning technique which uses knowledge from its previous training in another related problem set. In this chapter, we have demonstrated the use of DL models through transfer learning, deep feature extraction, machine learning models, and comparison of their performances.
-
A deep learning approach for COVID-19 detection from computed tomography scansThe classification of COVID-19 patients from chest computed tomography (CT) images is a very difficult task due to the similarities observed with other lung diseases. Based on various CT scans of COVID and non-COVID patients, the aim of this chapter is to propose a simple deep learning architecture and compare its diagnostic performance using transfer learning and several machine learning techniques that could extract COVID-19’s graphical features and classify them in order to provide a clinical diagnosis ahead of the pathogenic test, thus saving critical time for disease control. We also compare our approach and show that it outperforms various previous state-of-the-art techniques. We propose a deep learning architecture for transfer learning that is just a simple modification of eight new layers on the ImageNet pretrained convolutional neural networks (CNNs) which yielded us the best test accuracy of 98.30%, F1 score of 0.982, area under the receiver operating characteristic (ROC) curve of 0.982, and kappa value of 0.964 after training. Moreover, we use the proposed architecture for feature extraction and study the performance of various classifiers on them and were able to obtain the highest test accuracy of 91.75% with K-nearest neighbors. Also, we compare multiple CNNs and machine learning models for their diagnostic potential in disease detection and suggest a much faster and automated disease detection methodology. We show that smaller and memory efficient architectures are equally good compared to deep and heavy architectures at classifying chest CTs. We also show that visual geometry group (VGG) architectures are overall the best for this task.
-
Artificial intelligence-based retinal disease classification using optical coherence tomography imagesOptical coherence tomography (OCT) is a noninvasive imaging technology used to obtain high-resolution cross-sectional images of the retina. The layers within the retina can be differentiated and retinal thickness can be measured to aid in the early detection and diagnosis of retinal diseases and conditions. Notwithstanding the proven utility of OCT images, diagnosing large datasets of OCT images using the manual method still remains a challenge. In this chapter, we propose a deep learning-based approach, namely, the use of convolutional neural networks (CNN) and some pretrained image classification models on top of CNNs to get a proper and faster diagnosis of the OCT images. We also experiment with the features extracted using pretrained image classification models. Mainly three diseases—drusen, diabetic macular edema, choroidal neovascularization are addressed in this study. Our technique achieves an accuracy score of 0.9948 and an F1 score of 0.9948 on the test set. The outcomes suggest that the proposed model could serve as an analytic module to alert users and/or medical experts when retinal diseases are suspected.
-
Introduction to artificial intelligence techniques for medical image analysisAs the main goal of artificial intelligence (AI) is to provide inference from a sample, it employs statistics theory to develop mathematical models. When a model is constructed, its description and algorithmic solution for understanding must be competent. In some cases, the AI algorithm’s competency may be just as crucial as its classification accuracy. AI is applied in a variety of domains, such as anomaly detection, forecasting, medical signal/image analysis as a decision support component, and so on. The goal of this chapter is to assist scientists in selecting an acceptable AI approach and then guiding them in determining the best strategy by utilizing medical imaging. Furthermore, to introduce readers with the fundamentals of AI before digging into tackling real-world issues with AI methodologies. Machine learning, deep learning, and transfer learning are examples of basic ideas discussed. Topics relating to the various AI methodologies, such as supervised and unsupervised learning, will be covered. As a result, the key AI algorithms are discussed briefly in this chapter. Relevant PYTHON programming codes and routines are provided in each section.
-
Alzheimer’s disease detection using artificial intelligenceBiomedical data relevant to several diseases are generally employed to diagnose precise physiological or pathological conditions. The objective of biomedical image analysis is exact modeling by using Artificial Intelligence (AI) algorithms to diagnose different diseases. Alzheimer’s disease (AD) is one of the most widespread dementia forms influencing the elderly people. On-time diagnosis of Alzheimer’s disease is crucial to discover innovative methods for AD treatment. AI is an efficient approach for AD detection since it can be utilized as a Computer-aided decision support systems approach in medical procedures and play a critical role to detect changes in the brain images to identify AD. This chapter presents the recent studies and advances in AI used for medical image analysis and image processing in AD detection. The main focus is to have a consistent but easy and quick model for automated AD detection relied on the application of AI methods. Hence, the focus will be on AI techniques for AD detection from brain images. Moreover, some of the AI techniques, which were utilized for AD detection is overviewed. Then a simple AD detection approach using deep learning models will be presented. The results show that CNN achieved a testing accuracy of 95.70% accuracy and a validation accuracy of 99.71% for the diagnosis of AD from brain MRI scans. The chapter will be completed with a review of the current state-of-the-art, a discussion of new trends and open challenges for potential investigation.
-
Deep learning approaches for the cardiovascular disease diagnosis using smartphoneOne of the most important subjects for societies is human health services, which aims to determine the appropriate, accurate and robust diagnosis of the disorder for patients to get the adequate treatment as quickly as possible. Since this diagnosis is always a challenging process, support from other areas such as statistics and computer science are needed for healthcare. Biomedical signals relevant to several diseases are recorded from the human body and are generally employed to diagnose physiological or pathological conditions. The objective of biomedical signal analysis is exact modeling by using machine learning techniques for the diagnosis of diseases. This chapter explains how deep learning approaches are utilized in disease diagnosis. An automated diagnosis of cardiovascular diseases (CVDs) based on deep learning approaches is also presented as a case study. Atrial fibrillation (AFib) is one of the most common chronic and relapsing heart arrhythmias. Mechanocardiography (MCG) through which translational and rotational precordial chest movements are monitored is an effective approach for the detection of CVDs. MCG information obtained from cardiac patients using a smartphone's multidimensional built-in inertial sensors. The aim is to identify AFib episodes employing a smartphone MCG (or sMCG). Hence, this book chapter deals with applications of deep learning for the diagnosis of human diseases. In addition, this chapter focuses on current methods relevant to the utilization of deep learning techniques employed for cardiac abnormality detection, in order to discover remarkable patterns, make non-trivial assessments and make use of smartphone sensors effective in decision making. Hence, this chapter will assist researchers to explore the applicability of artificial intelligence approaches in their particular specialties for disease diagnosis and treatment.
-
Advanced pattern recognition tools for disease diagnosisMachine learning (ML) uses statistical theory to create models from data samples. Using the predictive and statistical models, computers can clean and curate the data, interpret and predict the outcomes of certainties (or uncertainties) with precise accuracy. Of course, the interpretation of the produced results and algorithmic solution designed for each problem needs to be fine-tuned and proficient for the target problem. Biomedical images relevant to different diseases are recorded from a body and are generally employed to diagnose precise physiological or pathological conditions. The objective of biomedical image analysis is exact modeling by using pattern recognition and computer vision to diagnose diseases by employing ML techniques. This chapter explains how artificial intelligence (AI) and ML techniques are utilized in disease diagnosis. An automated COVID-19 diagnosis approach based on deep feature extraction is also presented. After extracting features using deep transfer learning (DTL), the X-ray images are fed into the shallow ML model to diagnose COVID-19 from X-ray images. With chest X-ray, a patient can be identified as a potential COVID-19 patient and can be quarantined. X-ray equipment are already accessible in most hospitals, and already digitized. Since X-ray images are high dimensional data, a Convolutional Neural Network based feature extraction via transfer learning models are appropriate for the diagnosis of COVID-19. It may help an inpatient environment where the existing programs find it difficult to determine whether to keep the patient inward with other patients or separate them. This technique will also help classify patients with high COVID-19 risk who need to repeat testing with a false negative RT-PCR
-
Automated detection of colon cancer using deep learningColon cancer is a type of cancer that affects the large population. In the beginning, small ploys are formed in the large intestine, which if left untreated become cancer. Detection of colon cancer in its early stages reduces the risk of life to large extent and makes treatment easier and reduces the cost of treatment. The traditional way of detection of colon cancer is very time-consuming and often can be wrong if not done by a skilled person. The development of deep learning has enabled us to detect colon cancer accurately using histopathological images. In this chapter different pretrained models and techniques are used. The accuracy of results is very good; using ResNet50 99.8% accuracy is achieved on the test data which is very good. Using these techniques, the time for detection of colon cancer can be reduced significantly.
-
Lung cancer detection from histopathological lung tissue images using deep learningLung cancer is a disease in which the growth of cells in the lung goes out of control. This disease can be lethal if the treatment to stop the growth of cells is not given to the patient in its early stages. Hence, it is very crucial to correctly recognize lung cancer in less time. Using the traditional method where each tissue is observed by a medical practitioner is time-consuming as well as error-prone; moreover, the practitioner should be very skilled. All these problems can be solved by using automated methods to detect lung cancer. In this chapter different deep learning models and techniques are used to detect lung cancer using histopathological images. The accuracy achieved by these models is very high and takes negligible time to give the results. Using a pretrained ResNet model combined with a support vector machine accuracy of 98.57% is achieved on the test data.
-
Artificial intelligence based Alzheimer’s disease detection using deep feature extractionAlzheimer’s disease (AD) is an acute brain disease that affects neural functions and destroys the memories and abilities of human beings. AD causes severe chronic, progressive, and irreversible cognitive declination and brain damage. It is one of the most common forms of dementia that affects the elderly. Early identification of AD is critical for developing new treatment options. Artificial intelligence (AI) is an excellent tool for detecting AD since these methods are used in clinical settings as a computer-aided diagnosis (CAD) system and play an important role in detecting alterations in brain images for AD detection. This chapter discusses the recent methods and developments in medical image analysis and image processing for AD detection using AI. The primary objective of this chapter is the development of easy-to-implement methods that promote early AD detection based on deep feature extraction methods. We developed a deep feature extraction methodology with machine learning approaches to achieve a good performance in AD detection. Furthermore, some of the techniques that were used by previous researchers are reviewed. A discussion on the existing state-of-the-art methods, a review of emerging trends, and future research problems will round up the chapter
-
Diagnosis of breast cancer from histopathological images with deep learning architecturesBreast cancer is one of the most common cancer types among women worldwide. If not treated in earlier stages, it may be fatal. Therefore early diagnosis of breast cancer can minimize the human life risk. Mammograms and ultrasound imaging technologies play a crucial role to detect intraductal papillomas. However, the determination process of intraductal papillomas requires histopathological image analysis which may be mostly time-consuming, subjective, and tedious if carried out manually by the experts. To cover issue, computer-aided diagnosis (CAD) systems came into consideration. However, earlier CAD systems could not achieve significant improvement in the diagnosis process and their usage of them did not become widespread for more than a decade. Since deep learning has made so many significant advances in a wide variety of image applications, CAD systems that use its principles perform as well as the experts in stand-alone mode, and even perform better when used in support mode. In this chapter, we utilized various deep learning architectures for the detection process of breast cancer on the invasive ductal carcinoma (IDC) dataset which is one of the most popular and remarkable datasets in this field. According to the results, the pretrained VGG16 and MobileNet architectures obtain the best detection performance, reaching nearly 92% classification accuracy.
-
Artificial intelligence-based skin cancer diagnosisThe first melanoma tumor has affected millions of people across the globe and taken many human lives. It can be diagnosed in its early stage, therefore it becomes very important to detect it before it becomes lethal. The melanoma skin cancer can be detected from the images of tumor by applying various techniques of deep learning. Medical science has progressed to a large extent in recent times. Its progress can be catalyzed further with the help of technology such as artificial intelligence or deep learning. In the first stage of our study, we used CNN (convolutional neural network) and transfer learning for differentiating between normal and melanoma tumors. In the next stage, features of the image were extracted from different pretrained models and then these features were passed through global average pooling layer and a classifier was put on top of it. In the first stage, that is, end-to-end learning, MobileNet architecture achieved the highest F1 score of 0.8014. In feature extraction technique, the model in which features were extracted from MobileNet architecture and XGBoost was used as a classifier achieved the highest F1 score of 0.818
-
Skin cancer classification model based on hybrid deep feature generation and iterative mRMRChapter 4 develops a hybrid deep feature extraction model based on five pre-trained deep learning models and an ImRMR based feature selection model for skin cancer classification.
-
Brain hemorrhage detection using computed tomography images and deep learningBrain hemorrhage is one of the most serious medical diseases, requiring immediate treatment through posttraumatic healthcare. For this life-threatening disease, immediate care involves an urgent diagnosis. Intracranial bleeding is frequently associated with severe headaches and loss of consciousness. When a patient shows these symptoms, expert radiologists examine computed tomography (CT) images of the patient’s brain to locate and diagnose the type of bleeding. On the other hand, the manual examination performed by radiologists is complicated and time-consuming, naturally and unnecessarily delaying the intervention. In this chapter, we examined hemorrhage classification from CT images dataset, with deep learning architectures. In the experimental study, a total of 200 brain CT images were used as test and train. For this aim, different convolutional neural networks such as ResNet-18, EfficientNet-B0, VGG-16, and DarkNet-19 were used to classify brain CT images as normal and as hemorrhage. The accuracy (ACC), sensitivity (SEN), specificity (SPE), and F-score were used as the performance metrics for the classifier performances. The best classification results were ACC 83.50%, SEN 82%, SPE 85%, F-score 83.20%, and MCC 65% with DarkNet-19, respectively.