About Introduction to Biomedical Data Science
Data science has steadily increased in popularity over the past decade and involves all industries including healthcare and the biomedical sciences. The goal in healthcare is to use data science methods to improve medical quality and safety and reduce costs.
There is optimism that machine learning and artificial intelligence (AI) will be major drivers of predictive analytics, image, voice and text recognition. Recently, applied AI models have outperformed medical experts at classifying medical images - particularly in cardiology, dermatology, ophthalmology and radiology.
This textbook was written for anyone in the medical or informatics fields who feels they need a
stronger background in data science. Understanding the textbook content and data exercises do not require programming skills or higher math. Chapter exercises are based on healthcare data and supplemental YouTube videos are available in most chapters. The content begins with spreadsheet tips and tricks and ends with artificial intelligence.
Bill Hersh MD announced in February 2021 that OHSU will launch a new course on applied data science and machine learning for health and clinical informatics students using this textbook.
A special thanks to Ann Yoshihashi MD for her help with the publication of this textbook. If you have any questions, corrections or suggestions please use the Contact Us page.
Instructors: please register under the Register tab so you are eligible to download a PDF version of the book, PowerPoint slides, and an instructor manual.
Bios: Bob Hoyt MD and Bob Muenchen MS PSTAT
Below you will find a list of the textbook authors and a detailed table of contents.
1. Authors
Brenda Griffith
Technical Writer
Data.World
Austin, TX
Robert Hoyt MD, FACP, ABPM-CI, FAMIA
Associate Clinical Professor
Department of Internal Medicine
Virginia Commonwealth University
Richmond, VA
David Hurwitz MD, FACP, ABPM-CI
Associate CMIO
Allscripts Healthcare Solutions
Chicago, IL
Madhurima Kaushal MS
Bioinformatics
Washington University at St. Louis, School of Medicine
St. Louis, MO
Robert Leviton MD, MPH, FACEP, ABPM-CI, FAMIA
Assistant Professor
New York Medical College
Department of Emergency Medicine
Valhalla, NY
Karen A. Monsen PhD, RN, FAMIA, FAAN
Professor
School of Nursing
University of Minnesota
Minneapolis, MN
Robert Muenchen MS, PSTAT
Manager, Research Computing Support
University of Tennessee
Knoxville, TN
Dallas Snider PhD
Chair, Department of Information Technology
University of West Florida
Pensacola, FL
2. Table of Contents
OVERVIEW OF BIOMEDICAL DATA SCIENCE
-
Introduction
-
Background and history
-
Conflicting perspectives
-
the statistician’s perspective
-
the machine learner’s perspective
-
the database administrator’s perspective
-
the data visualizer’s perspective
-
-
Data analytical processes
-
raw data
-
data pre-processing
-
exploratory data analysis (EDA)
-
predictive modeling approaches
-
types of models
-
types of software
-
-
Major types of analytics
-
descriptive analytics
-
diagnostic analytics
-
predictive analytics (modeling)
-
prescriptive analytics
-
-
Putting it all together
-
Biomedical data science tools
-
Biomedical data science education
-
Biomedical data science careers
-
Importance of soft skills in data science
-
Biomedical data science resources
-
Biomedical data science challenges
-
Future trends
-
Conclusion
-
References
SPREADSHEET TOOLS AND TIPS
-
Introduction
-
basic spreadsheet functions
-
download the sample spreadsheet
-
-
Navigating the worksheet
-
Clinical application of spreadsheets
-
formulas and functions
-
filter
-
sorting data
-
freezing panes
-
conditional formatting
-
pivot tables
-
visualization
-
data analysis
-
-
Tips and tricks
-
Microsoft Excel shortcuts – windows users
-
Google sheets tips and tricks
-
-
Conclusions
-
Exercises
-
References
BIOSTATISTICS PRIMER
-
Introduction
-
Measures of central tendency & dispersion
-
the normal and log-normal distributions
-
-
Descriptive and inferential statistics
-
Categorical data analysis
-
Diagnostic tests
-
Bayes’ theorem
-
Types of research studies
-
observational studies
-
interventional studies
-
meta-analysis
-
-
Correlation
-
Linear regression
-
Comparing two groups
-
the independent-samples t-test
-
the wilcoxon-mann-whitney test
-
-
Comparing more than two groups
-
Other types of tests
-
generalized tests
-
exact or permutation tests
-
bootstrap or resampling tests
-
-
Stats packages and online calculators
-
commercial packages
-
non-commercial or open source packages
-
online calculators
-
-
Challenges
-
Future trends
-
Conclusion
-
Exercises
-
References
DATA VISUALIZATION
-
Introduction
-
historical data visualizations
-
visualization frameworks
-
-
Visualization basics
-
Data visualization software
-
Microsoft Excel
-
Google sheets
-
Tableau
-
R programming language
-
other visualization programs
-
-
Visualization options
-
visualizing categorical data
-
visualizing continuous data
-
-
Dashboards
-
Geographic maps
-
Challenges
-
Conclusion
-
Exercises
-
References
INTRODUCTION TO DATABASES
-
Introduction
-
Definitions
-
A brief history of database models
-
hierarchical model
-
network model
-
relational model
-
-
Relational database structure
-
Clinical data warehouses (CDWs)
-
Structured query language (SQL)
-
Learning SQL
-
Conclusion
-
Exercises
-
References
BIG DATA
-
Introduction
-
The seven v’s of big data related to health care data
-
Technical background
-
Application
-
Challenges
-
technical
-
organizational
-
legal
-
translational
-
-
Future trends
-
Conclusion
-
References
BIOINFORMATICS and PRECISION MEDICINE
-
Introduction
-
History
-
Definitions
-
Biological data analysis - from data to discovery
-
Biological data types
-
genomics
-
transcriptomics
-
proteomics
-
bioinformatics data in public repositories
-
biomedical cancer data portals
-
-
Tools for analyzing bioinformatics data
-
command line tools
-
web-based tools
-
-
Genomic data analysis
-
Genomic data analysis workflow
-
variant calling pipeline for whole exome sequencing data
-
quality check
-
alignment
-
variant calling
-
variant filtering and annotation
-
downstream analysis
-
reporting and visualization
-
-
Precision medicine - from big data to patient care
-
Examples of precision medicine
-
Challenges
-
Future trends
-
Useful resources
-
Conclusion
-
Exercises
-
References
PROGRAMMING LANGUAGES FOR DATA ANALYSIS
-
Introduction
-
History
-
R language
-
installing R & rstudio
-
an example R program
-
getting help in R
-
user interfaces for R
-
R’s default user interface: rgui
-
Rstudio
-
menu & dialog guis
-
some popular R guis
-
R graphical user interface comparison
-
R resources
-
-
Python language
-
installing Python
-
an example Python program
-
getting help in Python
-
user interfaces for Python
-
reproducibility
-
-
R vs. Python
-
Future trends
-
Conclusion
-
Exercises
-
References
MACHINE LEARNING
-
Brief history
-
Introduction
-
data refresher
-
training vs test data
-
bias and variance
-
supervised and unsupervised learning
-
-
Common machine learning algorithms
-
Supervised learning
-
Unsupervised learning
-
dimensionality reduction
-
reinforcement learning
-
semi-supervised learning
-
-
Evaluation of predictive analytical performance
-
classification model evaluation
-
regression model evaluation
-
-
Machine learning software
-
Weka
-
Orange
-
Rapidminer studio
-
Knime
-
Google tensorflow 2
-
honorable mention
-
summary
-
-
Programming languages and machine learning
-
Machine learning challenges
-
Machine learning examples
-
example 1 classification
-
example 2 regression
-
example 3 clustering
-
example 4 association rules
-
-
Conclusion
-
Exercises
-
References
ARTIFICIAL INTELLIGENCE
-
Introduction
-
definitions
-
-
History
-
Ai architectures
-
Deep learning
-
Image analysis (computer vision)
-
Radiology
-
Ophthalmology
-
Dermatology
-
Pathology
-
Cardiology
-
Neurology
-
Wearable devices
-
image libraries and packages
-
-
Natural language processing
-
NLP libraries and packages
-
text mining and medicine
-
speech recognition
-
-
Electronic health record data and AI
-
Genomic analysis
-
AI platforms
-
deep learning platforms and programs
-
-
Artificial intelligence challenges
-
general
-
data issues
-
technical
-
socio economic and legal
-
regulatory
-
adverse unintended consequences
-
need for more ML and AI education
-
-
Future trends
-
Conclusion
-
Exercises
-
References
BIOMEDICAL DATA SCIENCE RESOURCES