Details

Autor(en) / Beteiligte

• Chang, Anthony C.,

Titel

Intelligence-based cardiology and cardiac surgery : artificial intelligence and human cognition in cardiovascular medicine

Auflage

1st ed

Ort / Verlag

London, England : Academic Press,

Erscheinungsjahr

[2024]

Link zum Volltext

• Elsevier Books AutoHoldings

Beschreibungen/Notizen

• Includes bibliographical references and index.
• Front Cover -- Intelligence-Based Cardiology and Cardiac Surgery -- Intelligence-Based Cardiology and Cardiac Surgery -- Copyright -- Dedication -- Contents -- Contributors -- About the editors -- Foreword by Eric Topol -- Foreword by Ami Bhatt -- Preface -- Acknowledgments -- I - Basic concepts of data science and artificial intelligence -- 1 - Introduction to artificial intelligence for cardiovascular clinicians -- Basic concepts of artificial intelligence -- Definitions and concepts -- Artificial intelligence and the neurosciences -- History of artificial intelligence -- Key people and events -- Key epochs and movements -- History of artificial intelligence in medicine -- AI methodologies -- AI adoption -- Healthcare data and databases -- Cardiology data -- Healthcare data management -- Healthcare databases -- The data-to-intelligence continuum and AI -- Machine and deep learning -- Introduction to machine learning -- Classical machine learning and other types of learning -- Supervised learning -- Classification -- Regression -- Unsupervised learning -- Clustering -- Generalization (or dimension reduction) -- Semisupervised and self-supervised learning -- Ensemble learning -- Reinforcement learning -- Neural networks and deep learning -- Perceptron and multilayer perceptrons (MLP) -- Autoencoder neural network -- Generative adversarial network (GAN) -- Convolutional neural network (CNN) -- Recurrent neural network (RNN) -- Transfer learning -- Transformers -- Assessment of model performance -- Assessment methods -- Evaluation of regression models -- Evaluation of classification models -- Fundamental issues in machine and deep learning -- Model and data drift -- Model parameters and hyperparameters -- Imbalanced data set -- Interpretability and explainability -- Bias and variance trade-off -- Fitting of a model -- Curse of dimensionality.
• Correlation versus causation -- Machine versus deep learning -- Other key concepts and technologies in artificial intelligence -- Key concepts -- Bias and equity -- Complicated versus complex -- Healthcare cybersecurity -- Health economics -- Ethical considerations -- Education in clinicians -- Legal issues -- Overdiagnosis -- Regulatory issues -- Safety -- Key technologies -- Autonomous systems -- Blockchain -- Cloud technology -- Cognitive computing -- Digital twins -- Edge computing and embedded AI -- Extended reality -- Federated and swarm intelligence -- Foundation model -- Internet of things and everything (IoT and IoE) -- Knowledge graphs -- Low-shot learning -- Metaverse -- Monte Carlo simulation (MCS) -- Natural language processing (NLP) -- Robotics -- Robotic process automation (RPA) -- Human cognition and artificial intelligence in cardiology -- System 1 and system 2 thinking -- Uncertainty in biomedicine -- Clinician cognitive biases and heuristics -- Logical reasoning -- Evidence-based medicine -- Clinician perception/cognition -- Clinician-AI synergy: cognition-based AI -- Current status of AI in medicine and relevance to cardiovascular medicine -- The "why" for intelligence-based cardiology: sanctuary for the perfect storm -- Current areas of AI in medicine with relevance to cardiovascular medicine -- Medical imaging -- Extended reality -- Decision support -- Biomedical diagnostics -- Precision medicine -- Drug discovery -- Digital health -- Wearable technology -- Robotic technology -- Virtual assistance -- Current and future state of AI in cardiology -- Adoption of AI in cardiovascular medicine: the challenges ahead -- References -- II - Artificial intelligence in cardiovascular medicine -- A - Basic concepts of artificial intelligence in cardiology and cardiac surgery.
• 2 - Application of artificial intelligence in cardiovascular medicine and cardiac surgery -- Introduction -- Preview of the future -- Why is cardiology special for AI adoption? -- Current state of the art -- Data registries -- Augmented diagnostics, clinical decision support, risk prediction, and treatment -- Precision cardiology -- Potential opportunities for AI implementation in cardiology -- Future directions -- Challenges in implementation of AI in cardiology -- Strategies to overcome challenges in implementation of AI in cardiology -- Major takeaways -- Intelligence-based cardiac care -- References -- Further reading -- 3 - Data and databases in cardiovascular medicine and surgery -- Introduction -- Current state of the art -- Future directions -- Major takeaways -- References -- 4 - Data and databases for pediatric and adult congenital cardiac care -- Introduction -- Current state of the art -- Nomenclature -- Database -- Risk adjustment -- Verification of the completeness and accuracy of the data -- Collaboration across medical and surgical subspecialties -- Linking of databases and registries -- Longitudinal follow-up -- Assessment and improvement of quality -- Future directions -- Improving the discrimination and calibration of risk models with machine learning and other novel approaches -- Minimizing the burden to enter data and thereby minimizing the associated costs -- Documenting longitudinal outcomes consistently and effortlessly -- Measuring the value of healthcare -- Utilizing the massive amount of available data in our cardiac intensive care units with streaming analytics -- Decreasing the costs associated with research and the generation of new knowledge by conducting randomized trials within re ... -- Linking genotypic data to phenotypic data within registries.
• Utilizing pediatric and congenital cardiac databases to facilitate personalized precision medicine -- Major takeaways -- References -- 5 - Cognitive biases and heuristics in human cognition -- Introduction -- How do doctors make decisions? -- Our decision making environment -- Impact of culture on decision making -- Asking the right question -- Contrarians who seeing the unseen -- Utility theory and cognitive bias (Kahneman) -- The Sufi elephant and framing -- Other potential traps in decision making -- Evidence based medicine -- Solutions to heuristics, biases, and traps -- Solutions to improve decision-making -- Algorithms and decision trees to aid the decision-making process -- Major Takeaways -- References -- 6 - Spectrum bias in algorithms and artificial intelligence -- Introduction -- Current state of the art -- Spectrum bias and spectrum effect -- Examples of spectrum bias in non-AI diagnostic testing -- Spectrum bias in AI and machine learning algorithms -- Spectrum bias versus overfitting -- Identifying spectrum bias in existing studies -- Issues related to selection of disease spectrum -- Issues related to selection of controls -- Future directions -- Ways to reduce or prevent spectrum bias -- Describing and reporting the disease spectrum -- Subgroup analyses -- Major takeaways -- References -- 7 - Medical visual question answering -- Introduction -- Challenge 1: A trade-off between ambition and practicality -- Challenge 2: Classification versus generation VQA solvers -- Current state of the art -- General domain VQA -- Medical domain VQA -- Future directions -- Main takeaways -- References -- B - Artificial intelligence in cardiovascular areas -- 8 - Artificial intelligence and the electrocardiogram -- Introduction -- Current state of the art -- ECG in cardiac arrhythmia classification -- ECG in cardiovascular disease classification.
• ECG in cardiovascular disease risk prediction -- Utilization of ECG in noncardiac disease classification and prediction -- Future directions -- Transfer learning-ready deep learning models for ECG analysis -- Wearable technologies may help facilitate preventive care and address disparities in healthcare access -- Major takeaways -- References -- 9 - Artificial intelligence in electrophysiology -- Introduction -- Current state of the art -- Advancements in basic and computational science -- Emerging advancements in clinical practice -- ECG interpretation -- Atrial fibrillation -- Ventricular arrhythmia -- Cardiac resynchronization therapy -- Future directions -- Major takeaways -- References -- 10 - Artificial intelligence in echocardiography -- Introduction -- Current state of the art -- Machine learning in echocardiography -- Segmentation approaches to highlight objects and boundaries for cardiologists -- Regression analysis for predicting future outcomes -- Classification for automation of acquisition, diagnosis, and prognosis -- Future directions -- Major takeaways -- References -- Further reading -- 11 - Artificial intelligence in cardiac CT -- Introduction -- Current state of the art -- Artificial intelligence (AI) in CCT image preprocessing and quality improvement -- Artificial intelligence (AI) for cardiac structures segmentation -- Artificial Intelligence (AI) for coronary arteries segmentation and atherosclerosis characterization and functional assessment -- Functional assessment of CAD -- Outcome prediction -- Future directions -- Major takeaways -- References -- 12 - Artificial intelligence in cardiac MRI -- Introduction -- Background -- Current state of the art -- Deep learning for MR image acquisition and segmentation -- Deep ANNs for inverse problems in cardiac MRI -- Deep ANNs for cardiac MR image reconstruction.
• Deep learning for image generation.
• Intelligence-Based Cardiology and Cardiac Surgery: Artificial Intelligence and Human Cognition in Cardiovascular Medicine provides an especially timely multidisciplinary and comprehensive survey of artificial intelligence concepts and methodologies. It includes real-life applications in adult and pediatric cardiovascular medicine, spanning the life span from fetus to adult. Led by a senior cardiologist–data scientist and supported by renowned data scientists and cardiac clinicians with an ardent passion for artificial intelligence in cardiovascular medicine, the book provides a clinical interface between the medical and data science domains that is symmetric and realistic.
• Description based on print version record.