Correction to: Prediction Modeling Methodology

Correction to: Chapter 8 in: P. Kubben et al. (eds.), Fundamentals of Clinical Data Science,10.1007/978-3-319-99713-1_8

Value-Based Health Care Supported by Data Science

The value agenda involves measuring outcomes that matter and costs of care to optimize patient outcomes per dollar spent. Outcome and cost measurement in the value-based health care framework, centered around a patient condition or segment of the population, depends on data in every step towards healthcare system redesign. Technological and service delivery innovations are key components of driving transformation towards high-value health care. The learning health system and network-based thinking are complementary frameworks to the value agenda. Health care and medicine exist in a data-rich environment, and learning about how data can be used to measure and improve value of care for patients is and increasingly essential skill for current and future clinicians.

Extracting Features from Time Series

Clinical data is often collected and processed as time series: a sequence of data indexed by successive time points. Such time series can be from sources that are sampled over short time intervals to represent continuous biophysical wave-(one word waveforms) forms such as the voltage measurements representing the electrocardiogram, to measurements that are sampled daily, weekly, yearly, etc. such as patient weight, blood triglyceride levels, etc. When analyzing clinical data or designing biomedical systems for measurements, interventions, or diagnostic aids, it is important to represent the information contained within such time series in a more compact or meaningful form (e.g., noise filtering), amenable to interpretation by a human or computer. This process is known as feature extraction. This chapter will discuss some fundamental techniques for extracting features from time series representing general forms of clinical data.

Data Sources

There are many sources that relevant data for clinical data science can originate from. The brief overview in this chapter highlights the most frequent sources, but is definitely not exhaustive. The goal of this chapter is to provide an introduction to the most common data sources and to familiarize the reader with basic terminology in this context, in order to more easily understand discussions in next chapters and in literature in general.

Preparing Data for Predictive Modelling

This is the first chapter of five that cover an introduction to developing and validating models for predicting outcomes for the individual patient. Such prediction models can be used for predicting the occurrence or recurrence of an event, or of the most likely value on a continuous outcome. We will mainly focus on the prediction of binary outcomes, such as the occurrence of a complication, recurrence of disease, the presence of metastases, remission, survival, etc. This chapter deals with the selection of an appropriate study design for a study on prediction, and on methods to manipulate the data before the statistical modelling can begin.

The EU’s General Data Protection Regulation (GDPR) in a Research Context

This chapter introduces the rational and regulatory mechanism underlying the EU data protection framework with specific focus on the EU’s General Data Protection Regulation (GDPR). It outlines the applicability of the research exemption included in the GDPR and discusses further or secondary use of personal data for research purposes.

Clinical Decision Support Systems

Clinical decision support (CDS) includes a variety of tools and interventions computerized as well as non- computerized. High-quality clinical decision support systems (CDSS), computerized CDS, are essential to achieve the full benefits of electronic health records and computerized physician order entry. A CDSS can take into account all data available in the EHR making it possible to notice changes outside the scope of the professional and notice changes specific for a certain patient, within normal limits. However, to use of CDSS in practice, it is important to understand the basic requirements of these systems.This chapter shows in what way CDSS can support the use of clinical data science in daily clinical practice. Moreover, it explains what types of CDSS are available and how such systems can be used. However, to achieve high-quality CDSS which is effective in use requires thoughtful design, implementation and critical evaluation. Therefore, challenges surrounding implementation of a CDSS are discussed, as well as a strategies to develop and validate CDSS.

Data at Scale

Pre-requisites to better understand the chapter: basic knowledge of major sources of clinical data.Logical position of the chapter with respect to the previous chapter: in the previous chapter, you have learned what the major sources of clinical data are. In this chapter, we will dive into the main characteristics of presented data sources. In particular, we will learn how to distinguish and classify data according to its scale.Learning objectives: you will learn the major differences between data sources presented in previous chapters; how clinical data can be classified according to its scale. You will get familiar with the concept of ‘big’ clinical data; you will learn which are the major concerns limiting ‘big’ data exchange.

Prediction Modeling Methodology

In the previous chapter, you have learned how to prepare your data before you start the process of generating a predictive model. In this chapter, you will learn how to make a predictive model using very common regression techniques and how to evaluate the performance of a model. In the next chapter we will then look at more advanced machine learning techniques that have become increasingly popular in recent years.

Diving Deeper into Models

Pre-requisites to better understand the chapter: knowledge of the major steps and procedures of developing a clinical prediction model.Logical position of the chapter with respect to the previous chapter: in the last chapters, you have learned how to develop and validate a clinical prediction model. You have been learning logistic regression as main algorithm to build the model. However, several different more complex algorithms can be used to build a clinical prediction model. In this chapter, the main machine learning based algorithms will be presented to you.Learning objectives: you will be presented with the definitions of: machine learning, supervised and unsupervised learning. The major algorithms for the last two categories will be introduced.