Kooperativer Bibliotheksverbund

UID:

Format: XVII, 544 p. 95 illus., 80 illus. in color. , online resource.

Edition: 1st ed. 2024.

ISBN: 9783031506901

Series Statement: ICSA Book Series in Statistics,

Content: This book discusses statistical methods and their innovative applications in precision health. It serves as a valuable resource to foster the development of this growing field within the context of the big data era. The chapters cover a wide range of topics, including foundational principles, statistical theories, new procedures, advanced methods, and practical applications in precision medicine. Particular attention is devoted to the interplay between precision health, big data, and mobile health research, while also exploring precision medicine's role in clinical trials, electronic health record data analysis, survival analysis, and genomic studies. Targeted at data scientists, statisticians, graduate students, and researchers in academia, industry, and government, this book offers insights into the latest advances in personalized medicine using advanced statistical techniques.

Note: Part I An Overview of Precision Health in the Big Data Era -- Overview of Precision Health: Past, Current, and Future -- A Selective Review of Individualized Decision Making -- Utilizing Wearable Devices to Improve Precision in Physical Activity Epidemiology: Sensors, Data and Analytic Methods -- Policy Learning for Individualized Treatment Regimes on Infinite Time Horizon -- Q-Learning Based Methods for Dynamic Treatment Regimes -- Personalized Medicine with Multiple Treatments -- Statistical Reinforcement Learning and Dynamic Treatment Regimes -- Part II New Advances in Statistical Methods of Precision Medicine and the Applications -- Integrative Learning to Combine Individualized Treatment Rules from Multiple Randomized Trials -- Adaptive Semi-supervised Learning for Optimal Treatment Regime Estimation with Application to EMR Data -- Estimation and Inference for Individualized Treatment Rules Using Efficient Augmentation and Relaxation Learning -- Subgroup Analysis Using Doubly Robust Semiparametric Procedures -- A Selective Overview of Fusion Penalized Learning in Latent Subgroup Analysis for Precision Medicine -- Part III Precision Medicine in Clinic Trials and the applications to EHR Data -- Mining for Health: A Comparison of Word Embedding Methods for Analysis of EHRs Data -- Adaptive Designs for Precision Medicine in Clinical Trials: A Review and Some Innovative Designs -- Maximum Likelihood Estimation and Design and Inference Considerations for Sequential Multiple Assignment Randomized Trials -- Precision Medicine Designs for Cancer Clinical Trials -- Part IV Precision Medicine in Survival Analysis and Genomic Studies -- Variant Selection and Aggregation of Genetic Association Studies in Precision Medicine -- Leveraging Functional Annotations Improves Cross-population Genetic Risk Prediction -- A Soft-Thresholding Operator for Sparse Time-Varying Effects in Survival Models -- Discovery of Gene-specific Time Effects on Survival -- Modeling and Optimizing Dynamic Treatment Regimens in Continuous Time.

In: Springer Nature eBook

Additional Edition: Printed edition: ISBN 9783031506895

Additional Edition: Printed edition: ISBN 9783031506918

Additional Edition: Printed edition: ISBN 9783031506925

Language: English

DOI: 10.1007/978-3-031-50690-1

URL: https://doi.org/10.1007/978-3-031-50690-1

URL: Volltext  (URL des Erstveröffentlichers)

URL: lizenzpflichtig

UID:

Format: 1 online resource (0 pages)

Edition: 1st ed. 2024.

ISBN: 3-031-50690-1

Series Statement: ICSA Book Series in Statistics,

Content: This book discusses statistical methods and their innovative applications in precision health. It serves as a valuable resource to foster the development of this growing field within the context of the big data era. The chapters cover a wide range of topics, including foundational principles, statistical theories, new procedures, advanced methods, and practical applications in precision medicine. Particular attention is devoted to the interplay between precision health, big data, and mobile health research, while also exploring precision medicine's role in clinical trials, electronic health record data analysis, survival analysis, and genomic studies. Targeted at data scientists, statisticians, graduate students, and researchers in academia, industry, and government, this book offers insights into the latest advances in personalized medicine using advanced statistical techniques.

Note: Part I An Overview of Precision Health in the Big Data Era -- Overview of Precision Health: Past, Current, and Future -- A Selective Review of Individualized Decision Making -- Utilizing Wearable Devices to Improve Precision in Physical Activity Epidemiology: Sensors, Data and Analytic Methods -- Policy Learning for Individualized Treatment Regimes on Infinite Time Horizon -- Q-Learning Based Methods for Dynamic Treatment Regimes -- Personalized Medicine with Multiple Treatments -- Statistical Reinforcement Learning and Dynamic Treatment Regimes -- Part II New Advances in Statistical Methods of Precision Medicine and the Applications -- Integrative Learning to Combine Individualized Treatment Rules from Multiple Randomized Trials -- Adaptive Semi-supervised Learning for Optimal Treatment Regime Estimation with Application to EMR Data -- Estimation and Inference for Individualized Treatment Rules Using Efficient Augmentation and Relaxation Learning -- Subgroup Analysis Using Doubly Robust Semiparametric Procedures -- A Selective Overview of Fusion Penalized Learning in Latent Subgroup Analysis for Precision Medicine -- Part III Precision Medicine in Clinic Trials and the applications to EHR Data -- Mining for Health: A Comparison of Word Embedding Methods for Analysis of EHRs Data -- Adaptive Designs for Precision Medicine in Clinical Trials: A Review and Some Innovative Designs -- Maximum Likelihood Estimation and Design and Inference Considerations for Sequential Multiple Assignment Randomized Trials -- Precision Medicine Designs for Cancer Clinical Trials -- Part IV Precision Medicine in Survival Analysis and Genomic Studies -- Variant Selection and Aggregation of Genetic Association Studies in Precision Medicine -- Leveraging Functional Annotations Improves Cross-population Genetic Risk Prediction -- A Soft-Thresholding Operator for Sparse Time-Varying Effects in Survival Models -- Discovery of Gene-specific Time Effects on Survival -- Modeling and Optimizing Dynamic Treatment Regimens in Continuous Time.

Additional Edition: ISBN 3-031-50689-8

Language: English

DOI: 10.1007/978-3-031-50690-1

UID:

Format: 1 online resource (0 pages)

Edition: First edition.

ISBN: 3-031-50690-1

Series Statement: ICSA Book Series in Statistics Series

Note: Intro -- Preface -- Contents -- Editors and Contributors -- About the Editors -- Contributors -- Chapter Reviewers -- Part I An Overview of Precision Health in the Big Data Era -- Precision Health: Past, Current, and Future -- 1 The Roadmap from Precision Medicine to Precision Health -- 2 Part I: An Overview of Precision Health in the Big Data Era (Second Chapter to Seventh Chapter) -- 3 Part II: New Advances in Statistical Methods of Precision Medicine and the Applications (Eighth Chapter to Twelfth Chapter) -- 4 Part III: Precision Medicine in Clinic Trials and the Applications to EHR Data (Thirteenth Chapter to Sixteenth Chapter) -- 5 Part IV: Precision Medicine in Survival Analysis and Genomic Studies (Seventeenth Chapter to Twenty-First Chapter) -- 6 Further Discussions -- A Selective Review of Individualized Decision Making -- 1 Single-Stage Decision Problems -- 1.1 Regression-Based Methods -- 1.1.1 Q-Learning and A-Learning -- 1.1.2 Weighted Loss Minimization -- 1.2 Direct-Search Methods -- 1.2.1 Inverse Probability Weighted Estimate (IPWE) -- 1.2.2 Augmented Inverse Probability Weighted Estimate (AIPWE) -- 1.3 Other Considerations -- 1.3.1 Risk-Averse Utilities -- 1.3.2 Generalizability -- 2 Multi-Stage Decision Problems -- 2.1 Regression-Based Methods -- 2.2 Direct-Search Methods -- 2.2.1 T -Stage IPWE -- 2.2.2 T -Stage AIPWE -- 3 Discussion -- Appendix: Proofs -- References -- Utilizing Wearable Devices to Improve Precision in Physical Activity Epidemiology: Sensors, Data and Analytic Methods -- 1 Introduction -- 1.1 Physical Activity Epidemiology -- 1.2 Wearable Devices -- 1.3 Motivating Studies -- 1.4 Analytic Challenges for Accelerometry -- 2 Accelerometry Data and Analytic Workflow -- 3 Accelerometry Processing, Summarization and Association Analysis -- 3.1 Standard Analytic Approaches. , 3.2 Raw Accelerometry-Based Metrics (AC Alternatives) -- 3.3 Machine Learned PA Behaviors Based on Raw Accelerometry (Beyond AC) -- 3.4 Summarization and Association Analysis: Flexible Functional Approaches -- 4 Application to the OPACH: A Novel Dose-Response Analysis of PA Intensity and Cardiovascular Health -- 4.1 Processing and Summarization: AAI vs. AC -- 4.2 Association with PA Intensity Categories: AAI vs. AC -- 4.3 Association with Continuous PA Intensity: An AAI-Based Functional Approach -- 5 Discussion -- References -- Policy Learning for Individualized Treatment Regimes on Infinite Time Horizon -- 1 Background and Challenges -- 2 Preliminaries -- 2.1 A Motivating Example -- 2.2 Notation -- 3 Q-learning for Finite Horizon -- 3.1 Additional Notation -- 3.2 Backward Induction -- 4 Temporal Difference Learning -- 4.1 On-Policy vs. Off-Policy Learning -- 4.2 On-Policy TD Learning -- 4.3 Generalization to Off-Policy Learning -- 5 Temporal Difference and Estimating Equations -- 6 Residual Gradient Algorithm -- 7 Proximal Temporal Consistency Learning -- 7.1 Temporal Consistency and Functional Space Embedding -- 7.2 A Proximal Bellman Operator with Sparse Policies -- 7.3 OhioT1DM Data Analysis -- 7.3.1 A Mobile Health Study: OhioT1DM -- 7.3.2 ProximalDTR Package and Analysis Results -- 8 Online Experiment and Policy Learning -- 9 Discussion -- References -- Q-Learning Based Methods for Dynamic Treatment Regimes -- 1 Introduction -- 2 Introduction to the Precision Medicine Framework -- 2.1 Notation and Potential Outcomes -- 2.2 DTRs and Optimal DTRs -- 2.3 Identification -- 3 Q-Learning for Precision Medicine -- 3.1 An Incomplete Conceptual Overview of Reinforcement Learning to Q-Learning -- 3.2 Q-Learning in the Context of Precision Medicine -- 4 Q-Learning for Optimal DTRs in the Finite Time Horizon Setting. , 4.1 Q-Learning for Optimal Single-Stage DTRs -- 4.2 Additional Considerations for the Single-Stage Setting -- 4.3 Q-Learning for Optimal Finite-Stage DTRs -- 4.4 Additional Considerations for the General Finite-Stage Setting -- 5 Q-Learning for Optimal DTRs in the Infinite Time Horizon Setting -- 6 An Illustrative Example -- 7 Summary/Conclusions -- References -- Personalized Medicine with Multiple Treatments -- 1 Introduction -- 1.1 Outcome-Weighted Learning for Binary Treatments -- 1.2 Augmented Outcome-Weighted Learning -- 1.3 Direct Learning for Binary Treatments -- 1.4 Multi-Category Classification -- 2 Personalized Medicine with Multiple Treatments -- 2.1 Sequential Outcome-Weighted Learning -- 2.2 Multi-Category Outcome-Weighted Margin-Based Learning -- 2.3 Multi-Category Direct Learning -- 2.4 Multi-Label Outcome-Weighted Deep Learning Algorithms -- 3 Summary -- References -- Statistical Reinforcement Learning and Dynamic Treatment Regimes -- 1 Introduction -- 2 Reinforcement Learning and Markov Decision Process -- 3 Offline (Batch) Reinforcement Learning -- 3.1 Policy Iteration -- 3.2 Value Iteration and Q Iteration -- 3.3 Temporal Difference Learning -- 3.4 Policy Gradient -- 4 Dynamic Treatment Regimes -- 5 Causality and RL -- References -- Part II New Advances in Statistical Methods of Precision Medicine and the Applications -- Integrative Learning to Combine Individualized Treatment Rules from Multiple Randomized Trials -- 1 Introduction -- 2 Methodology -- 2.1 Learning ITRs from a Single Trial -- 2.2 Integrative ITRs for Multiple Studies -- 2.2.1 Integrative Learning for High-Resolution ITRs Using Coarsened ITRs -- 2.2.2 Integrative Learning for Coarsened ITRs Using High-Resolution ITRs -- 2.3 Rationale of Integrative Learning -- 3 Simulation Study -- 3.1 Simulation Design. , 3.2 Simulations for Integrative Learning to Improve High-Resolution ITR -- 3.3 Simulations for Integrative Learning of Coarsened ITR -- 4 Real Data Application -- 5 Discussion -- Appendix -- Special Case: Integrative Learning to Analyze a Trial with Block-Wise Collection of Covariates -- Baysian Rules for Integrative Learning -- Simulation Studies of Sensitivity Analysis -- EMBARC Analysis with Blockwise Data Collection -- References -- Adaptive Semi-supervised Inference for Optimal Treatment Decisions with Electronic Medical Record Data -- 1 Introduction -- 2 Optimal Treatment Decision with EMR Data -- 3 Optimal Treatment Decision Using Labeled Data Only -- 4 Semi-supervised Learning for Optimal Treatment Decision -- 4.1 Nonparametric Imputation of the Q-Functions -- 4.2 Semi-nonparametric Imputation of the Q-Functions -- 5 Simulation Analysis -- 5.1 Data Generation -- 5.2 Simulation Results -- 6 Application to an EMR Study -- 7 Discussion -- Appendix 1: Proof of Theorem 1 -- Appendix 2: Proof of Theorem 2 -- References -- Estimation and Inference for Individualized Treatment Rules Using Efficient Augmentation and Relaxation Learning -- 1 Introduction -- 2 A Novel Learning Approach to Optimal Individualized Treatment Rules -- 2.1 Setup and Notation -- 2.2 Recap: Outcome Weighted Learning -- 2.3 Efficiency Augmentation and Relaxation Learning -- 3 Inference of the Estimated Linear ITR -- 3.1 Inference Under a Differentiable Surrogate Loss -- 3.2 Inference Under a Non-differentiable Surrogate Loss -- 4 Inference Under High-Dimensional Covariate -- 4.1 Split-and-Pooled De-correlated Score Test for Differentiable Surrogate Loss -- 4.2 Kernel-Smoothened De-correlated Score Test for Non-differentiable Loss -- 5 Simulations -- 5.1 The Advantage of the EARL vs. Q-Learning Approach -- 5.2 The Advantage of the EARL Using a Hinge Loss vs. a Logistic Loss. , 6 Real Data Example -- 7 Summary -- References -- Subgroup Analysis Using Doubly Robust Semiparametric Procedures -- 1 Introduction -- 2 The Method -- 2.1 Brief Review of the Doubly Robust Procedures -- 2.2 The Semiparametric Propensity Score and Outcome Models -- 3 Simulation Study and Application -- 3.1 Simulation Study -- 3.2 Application to a Smoking Cessation Study -- References -- A Selective Overview of Fusion Penalized Learning in Latent Subgroup Analysis for Precision Medicine -- 1 Introduction -- 2 A Generalized Heterogeneous Regression Model -- 3 Model Specifications -- 3.1 Subject-Specific Coefficients Regression Model -- 3.2 Subject-Specific Intercepts Regression Model -- 3.3 Subject-Specific Nonparametric Regression Model -- 4 A General Estimation Procedure -- 4.1 ADMM Algorithm with Concave Penalties -- 5 Simulation Studies -- 5.1 Two Subgroups Example -- 6 Real Data Application -- 7 Discussion -- References -- Part III Precision Medicine in Clinic Trials and the Applications to EHR Data -- Mining for Health: A Comparison of Word Embedding Methods for Analysis of EHRs Data -- 1 Introduction -- 2 Methods -- 2.1 Word Embedding Methods -- 2.2 Embedding of EHRs -- 3 Numerical Experiments -- 3.1 Building Disease Prediction Models -- 3.2 Experimental Set Up -- 3.3 Hyperparameters -- 3.4 Results -- 3.5 Computational Efficiency -- 4 Discussion -- References -- Adaptive Designs for Precision Medicine in Clinical Trials: A Review and Some Innovative Designs -- 1 Introduction -- 2 Mathematical Framework of Adaptive Randomization Designs -- 3 Some Popular Innovative Designs -- 3.1 Basket Design -- 3.2 Umbrella Design -- 3.3 Platform Design -- 3.4 Other Designs -- 4 Covariate-Adaptive Randomization Designs -- 5 Covariate-Adjusted Response-Adaptive Randomization Designs -- 5.1 Some Existing CARA Designs -- 5.2 New CARA Designs. , 5.2.1 Introduction of New Designs.

Additional Edition: ISBN 3-031-50689-8

Language: English