Teaching plan for the course unit

 

 

Close imatge de maquetació

 

Print

 

General information

 

Course unit name: Biomedical Informatics

Course unit code: 571471

Academic year: 2019-2020

Coordinator: Alexandre Perera Lluna

Department: Faculty of Physics

Credits: 2,5

Single program: S

 

 

Estimated learning time

Total number of hours 62.5

 

Face-to-face learning activities

28

 

-  Lecture

 

12

 

-  Lecture with practical component

 

6

 

-  IT-based class

 

10

Supervised project

10

Independent learning

24.5

 

 

Recommendations

 

Attendance: 
The course will be divided into theoretical master classes (minimum 80% compulsory attendance), active discussions, individual and teamwork practical exercises (attendance for the presentations is 100% compulsory), and practical sessions in the laboratory(100% compulsory attendance).

 

 

Competences to be gained during study

 


Specific competences

  • Gain skills for the analysis of biomedical databases.
  • Analysis of large databases.
  • Python and R for the Biomedical informatics.
  • Gain capabilities to understand and employ computer information systems and communication in health systems.
  • Gain proficiency in health related measurements, calculus, evaluation and assessment , reports and audits.

General competences
  • Team work
  • Gain adaptation in rapid evolving environments
  • Work in multilingual environments, show proficiency in clear explanations of complex concepts. Work in a multidisciplinary team. 

 

 

 

 

Learning objectives

 

Referring to knowledge

— Understand the issues involved in large databases and their structure.

— Get to know programming languages for biomedical engineering.

— Get to know programming languages for biostatistics.

— Get to know specific methodologies of R+D in public and private research centres and companies.

 

Referring to abilities, skills

— Acquire the skills to gain programmatic access to biomedical databases through Python and R.

— Acquire the skills for the visualisation of large clinical databases.

— Acquire the abilities to employ computer equipment in health data management systems.

 

 

Teaching blocks

 

1. Introduction

*  

  • Information society and health
  • Databases in medical environments
  • What is BigData?
  • BidData in biomedical engineering
  • Tools in large datasets/BigData

2. Introduction to R

*  

  • Introduction and history
  • Types of data (including dataframe objects)
  • Manipulation of dataframe objects
  • Plotting
  • Introduction to statistical analysis through R
  • Introduction to panel data analysis in R

3. Introduction to Python

*  

  • Introduction and history
  • Types of data in Python, mutability and immutability
  • Control structures
  • Classes in Python

4. Introduction to scientific Python

*  

  • Introduction and history
  • Introduction to Numpy
  • Introduction to SciPy
  • Machine learning through Scipy
  • Code profiling

5. Database access

*  

  • Relational databases, structure and access
  • Multidimensional databases
  • Optimisation
  • Parallel access
  • Information aggregation
  • New database architectures, noSQL databases

6. Integration

*  

  • Accessing databases through Python
  • The DB-API
  • Accessing NoSQL databases in Python
  • R/Python integration

 

 

Teaching methods and general organization

 

The course is structured in two parts. During the first part, lecturers conduct broad introductory and advanced theory sessions on the use of Python and Scientific Python, R and interfacing of both languages in database systems. Students are asked to read scientific papers, write reviews and present selected papers in the area of biomedical data mining, and are expected to participate in and contribute to class discussions. Classes are mainly practical, with interactive sessions aimed to solve small tasks while incrementally approaching the data mining process. This first part takes half of the teaching time of the course.

The second part of the course focuses on the development of a research assignment. This assignment involves the design of the programmatic analysis of a very large (in terms of BigData) biomedical dataset in an open challenge format. The project is carried out in groups, where students make use of the acquired data mining techniques with the integration of Python, Scientific Python and R to extract relevant information from the database, including statistics and visualisation.

 

 

Official assessment of learning outcomes

 

Assessment focuses on the following activities:

Tutored and independent work: 66% of the grade

Oral presentations: 34% of the grade

Reassessment:
All students are eligible to take the reassessment exam, which consists of a single written examination. Students who want to be reassessed should renounce to their previous grade, if any, before the test. The new obtained grade in this reassessment will replace the previous one.

 

Examination-based assessment

Single assessment follows the criteria established in UB regulations. Students who wish to request single assessment must submit the corresponding application within the first two weeks of class.

Students who follow single assessment must submit all the assignments set during the course.

Students should also take a final examination on the same date as students who follow continuous assessment. The duration of the exam is longer for students who follow single assessment.