Efficacy Analysis in Clinical Trials an Update: Efficacy Analysis in an Era of Machine Learning

ISBN-13: 9783030199203 / Angielski / Miękka / 2020 / 304 str.

Ton J. Cleophas;Aeilko H. Zwinderman

Opis

Machine learning and big data is hot. It is, however, virtually unused in clinical trials. This is so, because randomization is applied to even out multiple variablesModern medical computer files often involve hundreds of variables like genes and other laboratory values, and computationally intensive methods are requiredThis is the first publication of clinical trials that have been systematically analyzed with machine learning. In addition, all of the machine learning analyses were tested against traditional analyses. Step by step statistics for self-assessments are includedThe authors conclude, that machine learning is often more informative, and provides better sensitivities of testing than traditional analytic methods do

Spis treści

Preface

Contents

Chapter 1

Traditional and Machine-Learning Methods for Efficacy Analysis

1.   Introduction

2.   The Principle of Testing Statistical Significance

3.   The T-Value = a Standardized Mean Result of a Study

4.   Unpaired T-Test

5.   Null-Hypothesis Testing of Three or More Unpaired Samples

6.   Three Methods to Test Statistically a Paired Sample

7.   Null-Hypothesis Testing of Three or More Paired Samples 

8.   Null Hypothesis Testing with Complex Data

9.   Paired Data with a Negative Correlation

10. Rank Testing

11. Rank Testing for Three or More Samples

12. Regression Analysis in the Efficacy Analysis of Clinical Trials

13. Predictors in Clinical Trials

14. Discrete and Discretized Data for Efficacy Analysis

15. Summary of Traditional Methods for Efficacy Analysis Applied in this Edition

16. Summary of Machine Learning Methods for Efficacy Analysis

17. Discussion

18. References

 

Chapter 2

Optimal-Scaling for Efficacy Analysis

1. Introduction

2. Example

3. Traditional Efficacy analysis

4. Optimal Scaling for Efficacy Analysis

5. Discussion

6. References

 

Chapter 3

Ratio-Statistic for Efficacy Analysis

1. Introduction

2. Data Example

3. Traditional Efficacy Analysis

4. Ratio-Statistic for Efficacy Analysis

5. Discussion

6. References

 

Chapter 4

Complex-Samples for Efficacy Analysis

1. Introduction

2. Data Example

3. Traditional Efficacy Analysis

4. Complex-Samples for Efficacy Analysis

5. Discussion

6. References

 

Chapter 5

Bayesian-Networks for Efficacy Analysis

1. Introduction

2. Data Example

3. Traditional Efficacy Analysis

4. Bayesian-Network for Efficacy Analysis

5. Discussion

6. References

 

Chapter 6

Evolutionary-Operations for Efficacy Analysis

1. Introduction

2. Data Example

3. Traditional Efficacy Analysis

4. Evolutionary-Operations for Efficacy Analysis

5. Discussion

6. References

 

Chapter 7

Automatic-Newton-Modeling for Efficacy Analysis

1. Introduction

2. Traditional Efficacy Analysis

    Dose-Effectiveness Study

    Time-Concentration Study

3. Automatic-Newton-Modeling for Efficacy Analysis

    Dose-Effectiveness Study

    Time-Concentration Study

4. Discussion

5. References

 

Chapter 8

High-Risk-Bins for Efficacy Analysis

1. Introduction

2. Traditional Efficacy Analysis

    The Fruit table

    The Snacks table

    The Fastfood table

    The Physicalactivities table

3. High-Risk-Bins for Efficacy Analysis

4. Discussion

5. References

 

Chapter 9

Balanced-Iterative-Reducing-Hierarchy for Efficacy Analysis

1. Introduction

2. Traditional Efficacy Analysis

    Example 1

    Example 2

3. Balanced-Iterative-Reducing-Hierarchy for Efficacy Analysis

    Example 1

    Example 2

4. Discussion

5. References

 

Chapter 10

Cluster-Analysis for Efficacy Analysis

1. Introduction

2. Data Example

3. Traditional Efficacy Analysis

4. Cluster Analysis for Efficacy Analysis

    1. Hierarchical cluster analysis

    2. K-means cluster analysis

    3. Density-based cluster analysis

5. Discussion

6. References

 

Chapter 11

Multidimensional-Scaling for Efficacy Analysis

1. Introduction

2. Traditional Efficacy Analysis

3. Multidimensional Scaling for Efficacy Analysis

    1. Proximity Scaling

    2. Preference Scaling

4. Discussion

5. References 

 

Chapter 12

Binary Decision-Trees for Efficacy Analysis

1. Introduction

2. Data Example with Binary Outcome

3. Traditional Efficacy Analysis

4. Decision-Trees for Efficacy analysis

5. Discussion

6. References

 

Chapter 13

Continuous Decision-Trees for Efficacy Analysis

1. Introduction

2. Data Example with a Continuous Outcome

3. Traditional Efficacy Outcome

4. Decision-Trees for Efficacy Analysis

5. Discussion

6. References

 

Chapter 14

Automatic-Data-Mining for Efficacy Analysis

1. Introduction

2. Data Example

3. Traditional Efficacy Analysis

4. Automatic-Data-Mining for Efficacy Analysis

    1. Step 1 open SPSS modeler

    2. Step 2 the distribution node

    3. Step 3 the audit node

    4. Step 4 the plot node

    5. Step 5. the web node

    6. Step 6 the type and c5.0 nodes

    7. Step 7 the output node

5. Discussion

6. References

 

Chapter 15

Support-Vector-Machines for Efficacy Analysis

1. Introduction

2. Data Example

3. Traditional Efficacy analysis

4. Support-Vector-Machines for Efficacy Analysis

    1. File reader node

    2. The nodes x-partitioner, svm learner, x-aggregator

    3. Error rates

    4. Prediction table

5. Discussion

6. References

 

Chapter 16

Neural-Networks for Efficacy Analysis

1. Introduction

2. Data Example

3. Traditional Efficacy Analysis

4. Neural-Networks Efficacy Analysis

5. Discussion

6. References

 

Chapter 17

Ensembled-Accuracies for Efficacy Analysis

1. Introduction

2. Data Example

3. Traditional Efficacy Analysis  

4. Ensembled-Accuracies for Efficacy Analysis

    1. Step 1 open SPSS modeler

    2. Step 2 the statistics file node

    3. Step 3 the type node

    4. Step 4 the auto classifier node

    5. step 5 the expert tab

    6. step 6 the settings tab

    7. step7 the analysis node

5. Discussion

6. References

 

Chapter 18

Ensembled-Correlations for Efficacy Analysis

1. Introduction

2. Example

3. Traditional Efficacy Analysis

4. Ensembled-Correlations for Efficacy Analysis

    1. Step 1 open SPSS modeler

    2. Step 2 the statistics file node

    3. Step 3 the type node

    4. Step 4 the auto numeric node

    5. Step 5 the expert node step

    6. Step 6 the settings tab

    7. Step 7 the analysis node

5. Discussion

6. References

 

Chapter 19

Gamma-Distributions for Efficacy Analysis

1. Introduction

2. Data Example

3. Traditional Efficacy Analysis

4. Gamma-Distributions for Efficacy Analysis

5. Discussion

6. References

 

Chapter 20

Validation with Big Data, a Big Issue

1. Introduction

2. Semantics of the Term Validation

3. Clinical Trial Validation

4. Diagnostic Test Validation

5. Big Data Validation

6. Big Data Jargon

7. Discussion

8. References

 

Index

O autorze

The authors are well-qualified in their field. Professor Zwinderman is past-president of the International Society of Biostatistics (2012-2015), and Professor Cleophas is past-president of the American College of Angiology (2000-2002). 

Professor Zwinderman is one of the Principle Investigators of the Academic Medical Center Amsterdam, and his research is concerned with developing statistical methods for new research designs in biomedical science, particularly integrating omics data, like genomics, proteomics, metabolomics, and analysis tools based on parallel computing and the use of cluster computers and grid computing.   

Professor Cleophas is a member of the Academic Committee of the European College of Pharmaceutical Medicine, that provides, on behalf of 22 European Universities, the Master-ship trainings  "Pharmaceutical Medicine" and "Medicines Development".  

From their expertise they should be able to make adequate selections of modern methods for clinical data analysis for the benefit of physicians, students, and investigators. The authors have been working and publishing together for 18 years, and their research can be characterized as a continued effort to demonstrate that clinical data analysis is not mathematics but rather a discipline at the interface of biology and mathematics.

The authors as professors and teachers in statistics at universities in The Netherlands and France for the most part of their lives, are concerned, that their students find regression-analyses harder than any other methodology in statistics. This is serious, because almost all of the novel methodologies in current data mining and data analysis include elements of regression-analysis, and they do hope that the current production "Regression Analysis for Starters and 2nd Levelers" will be a helpful companion for the purpose.

 

Five textbooks complementary to the current production and written by the same authors are 

Statistics applied to clinical studies 5th edition, 2012, 

Machine learning in medicine a complete overview, 2015, 

SPSS for starters and 2nd levelers 2nd edition, 2015, 

Clinical data analysis on a pocket calculator 2nd edition, 2016, 

Modern Meta-analysis, 2017

Regression Analysis in Medical Research, 2018 

all of them published by Springer