2. Teaching
  3. International Teaching
  4. Single courses (exams) taught in english language

HEALTH DATA ANALYTICS

International Teaching HEALTH DATA ANALYTICS

Back

0623200012
DEPARTMENT OF INFORMATION AND ELECTRICAL ENGINEERING AND APPLIED MATHEMATICS
EQF7
INFORMATION ENGINEERING FOR DIGITAL MEDICINE
2023/2024



YEAR OF COURSE 2
YEAR OF DIDACTIC SYSTEM 2022
SPRING SEMESTER
CFUHOURSACTIVITY
1HEALTH DATA ANALYTICS
216LESSONS
18EXERCISES
2HEALTH DATA ANALYTICS
216LESSONS
18EXERCISES
FABIO POSTIGLIONE21 Curriculum
Objectives
THE COURSE PROVIDES BASIC METHODOLOGICAL AND TECHNOLOGICAL TOOLS FOR DATA ANALYSIS, STATISTICAL INFERENCE AND SIGNIFICANCE ASSESSMENT OF CLINICAL/BIOMEDICAL INFORMATION.

KNOWLEDGE AND UNDERSTANDING
- METHODOLOGIES FOR THE ANALYSIS OF CLINICAL/BIOMEDICAL DATA (REGRESSION AND STATISTICAL INFERENCE TECHNIQUES; HYPOTHESIS TESTS AND DECISION STRATEGIES; CLUSTERING ALGORITHMS).
- ARCHITECTURES OF DATA-ANALYSIS AND DECISION-SUPPORT SYSTEMS, E.G., OLAP (ONLINE ANALYTICS PROCESSING), COMMONLY ADOPTED IN CLINICAL/BIOMEDICAL APPLICATIONS.

APPLYING KNOWLEDGE AND UNDERSTANDING
- EXTRACTING USEFUL INFORMATION FROM CLINICAL/BIOMEDICAL DATASETS BY APPLYING THE DATA ANALYSIS TECHNIQUES ILLUSTRATED DURING THE COURSE.
- MASTERING SOFTWARE FRAMEWORKS AND TOOLS FOR THE ANALYSIS OF CLINICAL/BIOMEDICAL DATASETS.
- DESIGNING OLAP SOLUTIONS TO SUPPORT DECISIONS IN CLINICAL/BIOMEDICAL APPLICATIONS.
Prerequisites
FUNDAMENTALS OF PROBABILITY AND PROGRAMMING.
Contents
Didactic unit 1: Regression
(LECTURE/PRACTICE/LABORATORY HOURS 14/0/6)
- 1 (2 Hours Lecture): Introduction to the analysis of clinical/biomedical data. Parametric and minimum-mean-squared-error estimators.
- 2 (2 Hours Lecture): Regression function. Simple linear regression.
- 3 (2 Hours Lecture): Multiple linear regression.
- 4 (2 Hours Lecture): Statistical inference. Hypothesis tests and p-value.
- 5 (2 Hours Lecture): Variable selection. Stepwise procedures.
- 6 (2 Hours Lecture): Regularization/shrinkage strategies.
- 7 (2 Hours Laboratory): Computer-aided implementation of simple and multiple linear regression algorithms.
- 8 (2 Hours Laboratory): Computer-aided implementation of inference and regularization/shrinkage strategies.
- 9 (2 Hours Lecture): Supervised non-parametric methods. K-NN method.
- 10 (2 Hours Laboratory): Computer-aided implementation of the K-NN method.

KNOWLEDGE AND UNDERSTANDING:
Regression models and statistical inference. Regularization/shrinkage strategies in high-dimensional problems.
APPLYING KNOWLEDGE AND UNDERSTANDING:
Extracting information from clinical/biomedical datasets exploiting regression algorithms. Using software tools to implement regression and inference strategies applied to clinical/biomedical datasets.


Didactic unit 2: Decision strategies
(LECTURE/PRACTICE/LABORATORY HOURS 6/0/4)
- 11 (2 Hours Lecture): Decision problems in clinical/biomedical applications. Hypothesis tests.
- 12 (2 Hours Lecture): Supervised decision methods. Naïve-Bayes.
- 13 (2 Hours Lecture): Supervised decision methods. Logistic regression.
- 14 (2 Hours Laboratory): Computer-aided implementation of hypothesis tests.
- 15 (2 Hours Laboratory): Computer-aided implementation of naïve-Bayes and logistic-regression classifiers.

KNOWLEDGE AND UNDERSTANDING:
Hypothesis tests and decision strategies.
APPLYING KNOWLEDGE AND UNDERSTANDING:
Designing and implementing decision algorithms for clinical/biomedical datasets. Using software frameworks to develop decision algorithms for clinical/biomedical datasets.


Didactic unit 3: Unsupervised strategies
(LECTURE/PRACTICE/LABORATORY HOURS 6/0/4)
- 16 (2 Hours Lecture): Principal Component analysis (PCA).
- 17 (2 Hours Laboratory): Computer-aided implementation of PCA.
- 18 (2 Hours Lecture): CLUSTERING. K-MEANS CLUSTERING. HIERARCHICAL CLUSTERING.
- 19 (2 Hours Lecture): Expectation-maximization algorithm. DBscan algorithm.
- 20 (2 Hours Laboratory): Computer-aided implementation of clustering algorithms.

KNOWLEDGE AND UNDERSTANDING:
Unsupervised strategies (PCA and clustering) for data analysis.
APPLYING KNOWLEDGE AND UNDERSTANDING:
Extracting information from clinical/biomedical datasets by implementing unsupervised algorithms.
Using software frameworks to implement PCA and clustering algorithms for clinical/biomedical datasets.


Didactic unit 4: Health data analytics systems
(LECTURE/PRACTICE/LABORATORY HOURS 4/0/4)
- 21 (2 Hours Lecture): Architectures and systems for data analysis and decision-making in clinical/biomedical applications.
- 22 (2 Hours Lecture): Overview of standards for health-data-analytics systems.
- 23 (2 Hours Laboratory): Framework for big data analytics relevant to clinical/biomedical applications.
- 24 (2 Hours Laboratory): Framework for big data analytics relevant to clinical/biomedical applications.

KNOWLEDGE AND UNDERSTANDING:
Architectures and systems for data analysis and decision-making in clinical/biomedical applications.
APPLYING KNOWLEDGE AND UNDERSTANDING:
Using frameworks for the statistical analysis of clinical/biomedical datasets.
Designing solutions for data analysis and decision-making in clinical/biomedical applications.



TOTAL LECTURE/PRACTICE/LABORATORY HOURS 30/0/18
Teaching Methods
THE COURSE INCLUDES THEORETICAL LECTURES, CLASSROOM EXERCISES, AND THE USAGE OF SOFTWARE TOOLS FOR DATA ANALYSIS.
Verification of learning
SUCCESSFUL ACHIEVEMENT OF THE LEARNING OUTCOMES WILL BE ASSESSED THROUGH A PROJECT WORK DEALING WITH THE ANALYSIS OF CLINICAL/BIOMEDICAL DATA.
Texts
AN INTRODUCTION TO STATISTICAL LEARNING, G. JAMES, D. WITTEN, T. HASTIE, R. TIBSHIRANI, SPRINGER, 2013.

STATISTICS FOR HEALTH DATA SCIENCE, R. ETZIONI, M. MANDEL, AND R. GULATI, SPRINGER, 2021.

SUPPLEMENTARY TEACHING MATERIAL WILL BE AVAILABLE ON THE UNIVERSITY E-LEARNING PLATFORM (HTTP://ELEARNING.UNISA.IT) ACCESSIBLE TO STUDENTS USING THEIR OWN UNIVERSITY CREDENTIALS.
More Information
THE COURSE IS HELD IN ENGLISH.
  BETA VERSION Data source ESSE3