Data Analysis in Medicine

Master 2019/2020

Type: Elective course (Data Science)

Area of studies: Applied Mathematics and Informatics

Delivered by: School of Data Analysis and Artificial Intelligence

Where: Faculty of Computer Science

When: 1 year, 1, 2 module

Mode of studies: offline

Instructors: Olga Rebrova

Master’s programme: Data Science

Language: English

ECTS credits: 4

Contact hours: 48

Full Syllabus

Abstract

Data analysis in medicine is a growing field, where information sciences meet modern clinical applications. The main goal of this class is to introduce HSE students to the broad spectrum of medical data analysis problems and applications, and to provide the students with the basic skills necessary for conduction professional medical data analysis. Medical data analysis is a field that bridges math, computer science, and medicine. Yet – and unfortunately – medical data analysis is still not offered by most universities, and the lack of this training is becoming more and more apparent. Today’s graduates have to be more prepared for solving the fundamental medical data analysis problems and for doing applied medical data analysis. This course is aimed at providing our students with knowledge, which could boost their careers. The course is based on the most recent medical data analysis developments and international standards. While the choice of medical data analysis, its problems and projects already defines the novelty of this class, we are trying to do our best to provide our students with the most up-to-date learning experience. The students work with real clinical data, solving real clinical problems. That is, unlike more conventional science classes, we prepare our students to solve real-world problems by working on these problems in the class. Independent, creative work is emphasized. Instead of “following the script” we accept the fact that several optimal solutions may be possible in most clinical projects. Critical thinking is emphasized. The goal of each class project is to develop a solution that can be used in real life – with noisy data, imperfect practices, human errors, diverse equipment.

Learning Objectives

To develop knowledge of concepts underlying medical research.
To develop practical skills needed in modern medical data analysis.

Expected Learning Outcomes

Student is acquainted with content of the course and main problems of the field.
Student is able to compare independent and dependent groups using parametric and non-parametric tests and confidence intervals for appropriate statistics.
Student is able to compare independent and dependent groups using statistical tests and confidence intervals for appropriate statistics.
Student knows basics of trial design and statistical aspects of the trial protocol. Student is able to allocate cases into groups using different types of randomisation.
Student knows the basics of statistical estimation of medical tests accuracy and is able to cakcuate point and interval estimates for appropriate parameters.
Student knows the idea of meta-analysys and indirect comparisons and is able to perform the basic calculations.
Student knows the specific issues of machine learning in medicine and is able to develop regression and classification models to solve medical probems.
Student is able to apply factor analysis or K-means or closed descriptions to solve medical problem.
Student knows the purposes and approaches to develop multi-criteria models.
Student knows the idea of clinical economical analysis.

Course Contents

Topic 1. Introduction to data analysis in medicine
Evidence based medicine, health technology assessment, role of statistical analysis, mathematical models as medical technologies
Topic 2. Basics for quantitative data analysis in medicine
Distributions, parametric and nonparametric descriptive statistics, hypothesis testing, confidence intervals, comparison of independent and dependent samples (parametric and nonparametric tests), comparison of 2 and more samples, correlation, multiple comparisons problem.
Topic 3. Basics for qualitative data analysis in medicine.
Descriptive statistics, hypothesis testing, confidence intervals, comparison of independent and dependent samples, comparison of 2 and more samples, association.
Topic 4. Experimental trials of treatments.
Design of trials, planning clinical trials of treatments, types of systematic bias, effect measures, data analysis, international guidelines for reporting.
Topic 5. Studies of diagnostic and screening tests.
Design of trials, planning trials of diagnostic and screening tests, types of systematic bias, effect measures, data analysis, international guidelines for reporting.
Topic 6. Meta-analysis. Indirect comparison.
Purposes, systematic review methodology, meta-analysis procedure, publication bias, heterogeneity, models for meta-analysis and indirect comparisons, sensitivity analysis, international guidelines for reporting.
Topic 7. Machine learning for medicine.
Specific problems of machine learning in medicine, decision support systems development approaches, general issues, regression models, classification models.
Topic 8. Data mining for medicine.
Factor analysis, K-means, closed descriptions.
Topic 9. Multicriteria decision analysis for medicine.
Purposes, approaches, stages, methods of weighting, models.
Topic 10. Clinical economic analysis.
Purposes, methods (cost-effectiveness, etc.), modeling (decision tree, Markov model).

Assessment Elements

Home work
Intermediate test
Exam
Class attendance

Interim Assessment

Interim assessment (2 module)
0.2 * Class attendance + 0.4 * Exam + 0.2 * Home work + 0.2 * Intermediate test

Bibliography

Recommended Core Bibliography

Pardalos P. M., Coleman T. F., Xanthopoulos P. (ed.). Optimization and Data Analysis in Biomedical Informatics. – Springer Science & Business Media, 2012.
Riffenburgh, R. H. (2006). Statistics in Medicine (Vol. 2nd ed). Amsterdam: Academic Press. Retrieved from http://search.ebscohost.com/login.aspx?direct=true&site=eds-live&db=edsebk&AN=248877
Riffenburgh, R. H. (2012). Statistics in Medicine (Vol. 3rd ed). Amsterdam: Academic Press. Retrieved from http://search.ebscohost.com/login.aspx?direct=true&site=eds-live&db=edsebk&AN=465058

Recommended Additional Bibliography

Pigott, T. Advances in meta-analysis. – Springer Science & Business Media, 2012.

Course Syllabus