Evolving Predictive Analytics in Healthcare: New AI techniques for real-time interventions

Description

A major use of practical predictive analytics in medicine has been in the diagnosis of current diseases, particularly through medical imaging. Now there is sufficient improvement in AI, IoT and data analytics to deal with real time problems with an increased focus on early prediction using machine learning and deep learning algorithms. With the power of artificial intelligence alongside the internet of 'medical' things, these algorithms can input the characteristics/data of their patients and get predictions of future diagnoses, classifications, treatment and costs. Evolving Predictive Analytics in Healthcare: New AI techniques for real-time interventions discusses deep learning algorithms in medical diagnosis, including applications such as Covid-19 detection, dementia detection, and predicting chemotherapy outcomes on breast cancer tumours. Smart healthcare monitoring frameworks using IoT with big data analytics are explored and the latest trends in predictive technology for solving real-time health care problems are examined. By using real-time data inputs to build predictive models, this new technology can literally 'see' your future health and allow clinicians to intervene as needed. This book is suitable reading for researchers interested in healthcare technology, big data analytics, and artificial intelligence. Cover Contents About the editors 1 COVID-19 detection in X-ray images using customized CNN model 1.1 Introduction 1.2 Related work 1.2.1 Key contributions and proposed work 1.3 Materials and methods 1.3.1 Feature extraction and selection 1.4 Results and discussion 1.5 Conclusion and future scope References 2 Introducing deep learning in medical diagnosis 2.1 Introduction 2.2 Literature survey 2.3 Overview of DL algorithms 2.3.1 Convolutional neural network 2.3.2 Recurrent neural network 2.3.3 Long short-term memory 2.3.4 Restricted Boltzmann machine 2.3.5 Deep belief networks 2.4 Proposed DL framework for neuro disease diagnosis 2.4.1 FAST-RCNN 2.4.2 Ten fully connected layer 2.5 Preprocessing of dataset 2.6 Implementation and results 2.7 Conclusion References 3 Intelligent approach for network intrusion detection system (NIDS) utilizing machine learning (ML) 3.1 Introduction 3.1.1 DoS and DDoS attacks 3.1.2 Man-in-the-middle (MitM) attack 3.1.3 Phishing and spear-phishing attacks 3.1.4 Password attack 3.1.5 Eavesdropping attack 3.1.6 Malware attack 3.2 Related work 3.3 Cloud computing 3.3.1 Machine learning 3.3.2 Exploratory data analysis 3.4 Results References 4 Classification methodologies in healthcare 4.1 Introduction 4.2 Classification algorithms 4.2.1 Statistical data 4.2.2 Discriminant analysis 4.2.3 Decision tree 4.2.4 K-nearest neighbor (KNN) 4.2.5 Logistic regression (LR) 4.2.6 Bayesian classifier 4.2.7 Support vector machine (SVM) 4.3 Parameter identification 4.3.1 Feature selection for classification 4.4 Real-time applications 4.4.1 Classification of patients based on medical record 4.4.2 Predictive analytics and diagnostic analytics based on medical records 4.4.3 Classification of diseases based on medical imaging 4.4.4 Mixed reality-based automation to help aid aging society 4.4.5 Tiny ML-based classification systems for medical gadgets 4.4.6 Classification systems for insurance claim management 4.4.7 Case study: Inspectra from Perceptra 4.4.8 Deep learning for beginners References 5 Introducing deep learning in medical domain 5.1 Introduction 5.1.1 DL in a nutshell 5.1.2 History of DL in the medical field 5.1.3 Benefits of DL in the medical domain 5.1.4 Challenges and obstacles of DL in the medical domain 5.1.5 Opportunities of DL in the medical field 5.2 DL applications in the medical domain 5.2.1 Drug discovery and medicine precision 5.2.2 Detection of diseases 5.2.3 Diagnosing patients 5.2.4 Healthcare administration 5.3 DL for medical image analysis 5.3.1 Medical image detection 5.3.2 Medical image recognition 5.3.3 Medical image segmentation 5.3.4 Medical image registration 5.3.5 Disease diagnosis and quantification 5.4 Conclusion References 6 Deep-stacked autoencoder for medical image classification 6.1 Introduction 6.2 Autoencoder 6.2.1 Stacked AE 6.2.2 Sparse AE 6.2.3 Convolutional AE 6.2.4 Deep AE 6.3 Proposed method 6.3.1 Representation learning using AE 6.3.2 Softmax layer 6.3.3 Support vector machine 6.3.4 K-nearest neighbor 6.3.5 Fine-tuning 6.3.6 Sparsity and regularization in AE 6.4 Results and discussions 6.4.1 Datasets 6.4.2 Evaluation metrics 6.4.3 Analysis of the simple AE 6.4.4 Effect of sparsity in AE 6.4.5 Effect of squeezing bottleneck in AE 6.4.6 Performance of deep stacked encoder 6.5 Conclusion References 7 Comparison of machine learning and deep learning algorithms for prediction of coronary heart disease 7.1 Introduction 7.1.1 Coronary heart disease (CHD) 7.1.2 ML and DL techniques 7.2 Related works 7.3 Materials and methods 7.3.1 Data preparation 7.3.2 Fixing the missing data issue 7.3.3 Data analysis 7.3.4 Feature selection 7.3.5 Balancing the dataset 7.3.6 Feature scaling 7.3.7 Methodology 7.3.8 Performance metrics 7.4 Results and discussion 7.5 Conclusion References 8 Revolution in technology-enabled healthcare: Internet of Things 8.1 IoT and healthcare information systems 8.2 Remote health monitoring and telehealth 8.2.1 PharmaIoT 8.2.2 Mobile applications for healthcare 8.2.3 Big data in healthcare 8.2.4 Challenges in MIoT 8.3 Wearables and medical devices 8.3.1 Activity trackers 8.3.2 Vital sign measurement 8.3.3 Smart jacket 8.3.4 Wire-based wearable devices 8.4 IoT in chronic diseases 8.5 IoT in emergency medical care 8.6 IoT and pregnancy care 8.7 IoT in eyecare 8.7.1 Visual acuity tester 8.7.2 Mobile imaging 8.8 Benefits of IoT in the healthcare system 8.9 Challenges with IoT in healthcare References 9 Smart healthcare monitoring framework using IoT with big data analytics 9.1 Introduction 9.2 Related work 9.3 Overview of IoT and big data 9.4 Data sources for healthcare 9.4.1 Electronic health records (EHR) data 9.4.2 Medical images data 9.4.3 Experimental data mining 9.4.4 Interactive data 9.4.5 Genomic data 9.5 Big data’s evolution in IoT 9.6 Recent trends in big data analytics and IoT 9.6.1 Specialized medical envisioning 9.6.2 Telehealth 9.6.3 Portable gadgets and the IoT 9.6.4 Biological IoT 9.7 Big data challenges in healthcare 9.7.1 Challenges relating to budgetary and economic considerations 9.7.2 Challenges relating to expertise 9.8 IoT challenges in healthcare 9.8.1 IoT and portable gadgets 9.8.2 Modes of communication in wearable devices 9.8.3 Smart healthcare monitoring frameworks 9.8.4 SHMS principles in the IoT 9.8.5 Implementation of SHMS with big data analytics 9.8.6 Proposed model 9.8.7 Case study 9.8.8 Performance evaluation of data analysis 9.9 Conclusion References 10 Experimental analysis and investigation of dementia detection framework using EHR-based variant LSTM model 10.1 Introduction 10.2 Related work 10.3 Materials and methods 10.3.1 EHR datasets 10.3.2 ML models 10.3.3 Approach to deep learning 10.3.4 Analysis of models 10.3.5 Proposed methodology 10.3.6 Model architecture 10.4 Dataset for the suggested method 10.4.1 Dataset pre-processing 10.4.2 Parameters of the CNN model 10.4.3 Parameters of the RNN model 10.4.4 Parameters of the LSTM model 10.5 Dementia detection and prediction model 10.6 Experimental results 10.7 Conclusion References 11 An intelligent agent-based distributed patient scheduling using token-based coordination approach: a case study 11.1 Introduction 11.1.1 Brief introduction to agent paradigm 11.1.2 Patient scheduling 11.1.3 Agent-based patient scheduling 11.2 Context of study and problem description 11.2.1 Application of agents in healthcare 11.2.2 Application of agents in scheduling 11.2.3 MAS toward coordination 11.3 Related work 11.3.1 Token as a coordination mechanism 11.3.2 Agent-based patient scheduling using token-based coordination 11.3.3 Algorithm for updating the nonlocal viewpoints of the resource 11.4 Model implementation and validation 11.4.1 Performance metrics 11.4.2 Comparison of results 11.5 Conclusion References 12 Internet of Things (IoT) for the efficient healthcare system 12.1 Introduction 12.2 Overview of IoT 12.3 Review of existing work 12.4 IoT architecture for Chikungunya and COVID-19 12.5 Conclusions and future scope References 13 Comprehension of melody representation and speed-up approaches for query by humming system 13.1 Introduction 13.2 Comparison with existing approaches 13.3 Experimental analysis of the proposed work 13.3.1 Mean reciprocal rank 13.3.2 Mean of average 13.3.3 Top X hit rate 13.3.4 Retrieval time 13.4 Approximation and envisioning of relations among performance appraisal metrics 13.4.1 Relevance analysis of mean reciprocal and mean of average rank 13.4.2 Synchronisation of accuracy and retrieval time with intersection point analysis 13.5 Conclusion References 14 Python for digital health solutions: elevated outcomes 14.1 Introduction 14.2 An overview of the evolution of the healthcare industry 14.2.1 A case study of Singapore 14.3 Python’s role in the healthcare industry 14.3.1 Healthcare data management 14.3.2 Healthcare simulations 14.3.3 Medical diagnosis, prognosis and treatment 14.3.4 Genomics and sequencing 14.3.5 A double-edged sword: the disadvantages of Python’s implementation 14.4 Conclusion Glossary References 15 IoT-enabled healthcare – a paradigm shift 15.1 Introduction 15.2 Architecture of IoT 15.3 IoT implementation in medical field 15.3.1 Architecture of medical IoT (MIoT) 15.3.2 Types of sensors used in MIoT 15.3.3 Tools and technologies used to implement MIoT 15.3.4 Functioning of healthcare system 15.4 IoT-enabled devices in healthcare 15.5 IoT technologies in medical field 15.6 Security challenges 15.6.1 Privacy and security 15.6.2 Data overloaded and accuracy 15.6.3 Outdated infrastructure 15.6.4 Cyber attack 15.7 Conclusion References 16 IoT-based cardiovascular prediction framework using deep learning algorithms 16.1 Introduction 16.1.1 Different types of CVDs 16.1.2 Intermediate risk factors of CVDs 16.1.3 Symptoms and prevention of CVDs 16.2 Related works 16.3 Introduction to deep learning 16.3.1 Deep learning vs. machine learning 16.3.2 Workflow of deep learning 16.3.3 Type of deep learning networks or algorithms 16.4 Proposed framework 16.4.1 Objectives of the proposed framework 16.4.2 Proposed framework 16.4.3 Methodologies 16.5 Discussion on experimental results 16.5.1 Hardware description 16.5.2 Dataset description 16.5.3 Selected features and evaluation parameters 16.5.4 Simulation results 16.6 Conclusion and future enhancement References 17 An intelligent approach using convolutional neural network (CNN) for early detection of melanoma and other skin diseases 17.1 Introduction 17.1.1 The skin 17.1.2 Anatomy of the skin 17.1.3 Problem statement 17.2 Scope of the project 17.2.1 Comprehensive analysis of related work 17.2.2 Dermatological disease detection using image processing and artificial neural network 17.2.3 Automatic detection and severity measurement of eczema using image processing 17.2.4 Skin cancer classification using deep learning and transfer learning 17.2.5 Dermatological classification using deep learning of skin image and patient background knowledge 17.3 Project requirements 17.3.1 Functional requirements 17.3.2 Non-functional requirements 17.3.3 Software requirements 17.4 Identification of alternative solutions and justification of selecting a solution 17.4.1 Acquisition of image 17.4.2 Classification types 17.4.3 CNN pre-trained model 17.4.4 Pre-processing of image 17.5 Application analysis 17.5.1 Model block diagram 17.5.2 Flowchart diagram 17.5.3 Use-case diagram 17.6 Details of the project implementation conforming to the proposal phase 17.6.1 Android mobile application front-end 17.6.2 Mobile application back-end development 17.6.3 Data preparation 17.6.4 Image processing for hair removal 17.6.5 Classification model building and training 17.7 Conclusion and future work References 18 Self-organizing deep learning approach for controlling movements of wheeled apparatus through corneal connotation 18.1 Introduction 18.2 Previous works 18.3 Methodology 18.4 Structural details 18.5 Conclusion References 19 Prediction of breast tumour outcome to chemotherapy using statistical MR images through deep learning approaches 19.1 Introduction 19.2 Materials and methods 19.2.1 Dataset 19.2.2 Neoadjuvant chemotherapy 19.2.3 MRI acquisition and parameters 19.2.4 Image processing 19.2.5 Data augmentation 19.3 CNN architectures 19.3.1 Single-input architecture 19.3.2 Multiple inputs architecture 19.4 Method evaluations 19.5 Results and discussion 19.6 Conclusion and future scope References 20 Risk analysis and prediction of cancer associated with Type II diabetes: a review 20.1 Introduction 20.2 Diabetes 20.2.1 Type I diabetes 20.2.2 Type II diabetes 20.3 Cancer 20.4 Related works 20.5 Performance analysis of existing methods 20.6 Conclusion and future work References Index Back Cover