Predicting Heart Failure. Группа авторов
Algorithms for Early Diagnosis of Heart Diseases
14 8 Applications of Machine Learning for Predicting Heart Failure
15 9 Machine Learning Techniques for Predicting and Managing Heart Failure
17 11 Commercial Non-Invasive and Invasive Devices for Heart Failure Prediction: A Review
19 13 Future Techniques and Perspectives on Implanted and Wearable Heart Failure Detection Devices
20 Index
List of Illustrations
1 Chapter 1Figure 1.1 Images for boundary detection in...Figure 1.2 Supervised and unsupervised machine learning.
2 Chapter 2Figure 2.1 Supine, left lateral decubitus...Figure 2.2 ECG. E denotes electrodes placed...Figure 2.3 Major waves of a single ECG signal...Figure 2.4 Treadmill assessment.Figure 2.5 Evolution of rubber-based stethoscopes.Figure 2.6 A digital stethoscope.Figure 2.7 The echocardiography method. The...Figure 2.8 Formation of plaques in the...Figure 2.9 Plaque formation in the heart arteries.
3 Chapter 3Figure 3.1 Statistics of the number of publications...Figure 3.2 A schematic representation of some...Figure 3.3 Application of wearable/implantable...
4 Chapter 4Figure 4.1 Implant sensor which can monitor...Figure 4.2 Implantable cardioverter defibrillators.Figure 4.3 Implantable biosensor underneath...Figure 4.4 Biosensor sensor insertion procedure.Figure 4.5 Different types of wireless energy transfer.Figure 4.6 Block diagram of microwave wireless power transfer.Figure 4.7 Rectenna (rectifying antenna)...Figure 4.8 Inductive mutual coupling.Figure 4.9 A schematic description of typical...Figure 4.10 Ultrasound wireless power transfer.Figure 4.11 Block diagram of the biosensor implant...Figure 4.12 Simple block diagram of inductive...Figure 4.13 Schematic of bidirectional...Figure 4.14 Simplified ADS schematic of resonant link.Figure 4.15 Schematic circuit of transmitting section.Figure 4.16 Block diagram of the demodulation module.Figure 4.17 Block diagram of the inductive...Figure 4.18 Output voltage.Figure 4.19 Block diagram of the data transfer structure.Figure 4.20 Inductive coupling between two...Figure 4.21 Average ambient RF power.Figure 4.22 The penetration depth of...Figure 4.23 Typical WiFi energy harvesting to power-up implant chip.Figure 4.24 Fractal antenna.Figure 4.25 Array of fractal Koch antennae patch.Figure 4.26 S-parameter.Figure 4.27 3D radiation pattern plots.Figure 4.28 Directivity of the proposed antenna.Figure 4.29 Active impedance.Figure 4.30 Block diagram of powering-up the implanted chip.Figure 4.31 Power-up pump circuit.Figure 4.32 Charge pump stage voltage.
5 Chapter 6Figure 6.1 Relationship between statistical models and AI.Figure 6.2 Relationship between AI and ML techniques.Figure 6.3 Linear and non-linear decision boundaries and surfaces.Figure 6.4 Comparing ML and deep learning.Figure 6.5 A perceptron and a simple ANN with interconnected perceptrons.Figure 6.6 Steps in building AI algorithms in cardiology.
6 Chapter 7Figure 7.1 Building the classifier phase.Figure 7.2 Using the classifier for prediction.Figure 7.3 Bar plot of target variable.Figure 7.4 Bar plot for the relationship between target and sex.Figure 7.5 Bar plot for the relationship between target, sex, and age.Figure 7.6 Histogram of age variables classified based on sex and target.Figure 7.7 Histogram of resting blood pressure...Figure 7.8 ROC for all models over the Cleveland data set.Figure 7.9 ROC for all models over the Statlog data set.
7 Chapter 8Figure 8.1 Overall architecture of HF diagnosis/prediction using ML models.
8 Chapter 9Figure 9.1 Prevalence of HF worldwide...Figure 9.2 NYHA classification. Based...Figure 9.3 Data description for HF diagnosis.Figure 9.4 Features for HF diagnosis.Figure 9.5 Data description for HF severity...Figure 9.6 Removed features after the application of the preprocessing step.Figure 9.7 Optimal set of features for severity estimation.Figure 9.8 Optimal set of features for adverse events prediction.
9 Chapter 10Figure 10.1 Various types of CVD and their origin.Figure 10.2 Various types of sensing mechanism for CVD.Figure 10.3 Different types of ELISA techniques.Figure 10.4 Wearable potentiometric ion sensor.Figure 10.5 DNA based magnetic biosensor.Figure 10.6 Role of AI in CVD treatment.Figure 10.7 Overview of the ML workflow. Reproduced...Figure 10.8 (a) Block diagram of the integrated DL...Figure 10.9 (a) Different left ventricular...
10 Chapter 11Figure 11.1 ECG reading using a Holter...Figure 11.2 General workflow of MCT devices.Figure 11.3 Hand-to-leg bioimpedance analysis system.Figure 11.4 General components of smart wearable...Figure 11.5 Overall working of remotely monitored invasive devices.
11 Chapter 12Figure 12.1 Schematic representation of the pathophysiology...
12 Chapter 13Figure 13.1 The system architecture of the CardioMEMS system [25].Figure 13.2 Cardiac pacemaker (left) [31] and CRT-D (right) [32].Figure 13.3 Left atrial pressure (LAP) sensing device:...Figure 13.4 Implantable cardiac devices.Figure 13.5 Different types of wearable devices...Figure 13.6 Concept of PPG which uses a light-emitting diode...Figure 13.7 Different signal processing and machine...Figure 13.8 Omron HeartGuide [44].Figure 13.9 Smart wearable devices for arrhythmia detection.Figure 13.10 Using machine learning algorithms in...Figure 13.11 Proposed framework to use selfies to screen patients with CAD.Figure 13.12 Concept of machine learning in...
List of Tables
1 Chapter 1Table 1.1 HF types and their outcomes.Table 1.2 Literature summary for artificial...
2 Chapter 3Table 3.1 Specific biomarkers for the evaluation of CVDs.Table 3.2 Latest biosensor applications for cardiovascular research.
3 Chapter 4Table 4.1 Comparison of different energy sources.
4 Chapter 5Table 5.1 Summary of trials...
5 Chapter 6Table 6.1 ML techniques.
6 Chapter 7Table 7.1 Heart failure data set with 14 attributes.Table 7.2 Statistical description of Cleveland data set.Table 7.3 Statistical description of Statlog data set.Table 7.4 Performance measures using the Cleveland data set.Table 7.5 Performance measures using the Statlog data set.
7 Chapter 8Table 8.1 ML paradigms and how they can help tackle HF.
8 Chapter 9Table 9.1 State of the art of ML for HF diagnosis.Table 9.2 State of the art of ML for HF severity estimation.Table 9.3 State of the art of ML for HF rehospitalization prediction.Table 9.4 State of the art of ML for HF mortality prediction.Table 9.5 State of the art of ML for HF prediction of adverse events.Table 9.6 HF diagnosis classification results, unbalanced class.Table 9.7 HF diagnosis classification results, balanced classTable 9.8 Results of the NYHA class detection model.Table 9.9 Results of the adverse events prediction model.
9 Chapter 11Table 11.1 BMI value ranges and associated health risks.Table 11.2 Non-invasive devices used in...Table 11.3 Normal range values of different pressure parameters.Table 11.4 Categories, device, company name...
10 Chapter 12Table