Smart Data Acquisition and Analysis
Code: 44023 ECTS Credits: 6| Degree | Type | Year | Semester |
|---|---|---|---|
| 4316624 Internet of Things for e-Health | OB | 0 | 1 |
Contact
- Name:
- Aura Hernández Sabaté
- Email:
- Aura.Hernandez@uab.cat
Use of Languages
- Principal working language:
- english (eng)
Teachers
- Aura Hernández Sabaté
- Debora Gil Resina
External teachers
- david geronimo
Prerequisites
Knowledge of programming languages (prefearably C++, Python or Matlab) and good mathematical background is highly recommended
Objectives and Contextualisation
This module will provide students with the techniques and algorithms necessary to extract and analyse patient data that have relevance in the field of EHealth. On the one hand, you will be provided with image and video processing algorithms to obtain information on the anatomy and physiology of the relevant patient from the point of view of the health application. We will explain the methods of artificial intelligence necessary for the analysis of patterns and decision making in the field of EHealth. Finally, an introduction to statistical methods of comparison of populations necessary for the validation of algorithms and methodologies will be made.
Competences
- Analyse and model phenomena with data, graphics and complex images in the context of IoT in the area of health using techniques of probability, statistics and artificial intelligence.
- Apply the ethical rules applicable in the health sector.
- Communicate and justify conclusions clearly and unambiguously to both specialist and non-specialist audiences.
- Continue the learning process, to a large extent autonomously.
- Solve problems in new or little-known situations within broader (or multidisciplinary) contexts related to the field of study.
- Solves problems of health and health services which require elements of the value chain of the IoT using specific concepts and technologies.
- Understand, analyse and evaluate theories, results and developments in the language of reference (English) as well as the mother tongue (Catalan, Spanish) in the area of IoT in health.
Learning Outcomes
- Apply anonymisation techniques to sensitive patient data in order to protect their privacy.
- Communicate and justify conclusions clearly and unambiguously to both specialist and non-specialist audiences.
- Continue the learning process, to a large extent autonomously.
- Correctly interpret the result of a test or statistical model for the population analysis of experimental data or algorithm validation.
- Define the best mixed regression model for the statistical analysis of data and algorithm validation in the area of health.
- Implement and evaluate IoT data-processing techniques: filtering and pre-processing of sensor signals, image processing and medical and video scanners.
- Solve problems in new or little-known situations within broader (or multidisciplinary) contexts related to the field of study.
- Understand, analyse and evaluate theories, results and developments in the language of reference (English) as well as the mother tongue (Catalan, Spanish) in the area of IoT in health.
Content
- Filtering and pre-processing of sensor signals
- Medical image processing and scanner techniques
- Video processing techniques
- Estimation of parameters and confidence intervals
- Simple and multiple hypothesis Test
- Supervised learning and unsupervised learning
- Neural networks and deep learning
Methodology
We will follow a problem based methodology, so learning will we based on the solution of usage cases related to real applications in the field of Iot. Students will be provided with the basic materials and tools required to solve each usage case. Teachers will also give some explanations at some lectures in order that students can understand usage cases and the provided tools. The remainining lectures will focus on helping students to solve the proposed usage cases and extending explanations related to techniques.
Activities
| Title | Hours | ECTS | Learning Outcomes |
|---|---|---|---|
| Type: Directed | |||
| Lecture Sessions | 50 | 2 | 1, 5, 6, 4 |
| Type: Supervised | |||
| Tutorized classroom activities (resolution of usage cases) | 92 | 3.68 | 1, 8, 5, 6, 4, 7, 3 |
Assessment
Resolution of Usage Cases. Following a PBL methodology, students will solve some usage cases in groups and with the help of the teacher (who will take the role of expert) during the course.
Individual Tests. Students' capability to apply the techniques will be also evaluated individually.
Assessment Activities
| Title | Weighting | Hours | ECTS | Learning Outcomes |
|---|---|---|---|---|
| Individual Tests | up to 60% | 2 | 0.08 | 1, 5, 6, 4 |
| Resolution of Usage Cases (Project) | up to 60% | 6 | 0.24 | 1, 8, 5, 6, 4, 7, 2, 3 |
Bibliography
Richard O. Duda, Peter E. Hart, David G. Stork, Pattern classification, Wiley, 2001
Steel, R. and Torrie, J. H. (1976), Introduction to Statistics -McGraw-Hill
Fisher, R.A. (1925), Statistical Methods for Research Workers - Edinburgh: Oliver & Boyd.
Curs online (MOOC Coursera): Image and video processing: From Mars to Hollywood with a stop at the hospital. (https://www.coursera.org/course/images)
Curs online (MOOC Coursera): Machine Learning. (https://es.coursera.org/learn/machine-learning)
Bruce Eckel, Thinking in PYTHON (on line at http://www.bruceeckel.com).
Paul Suetens, Fundamentals of medical imaging
David A. Forsyth and Jean Ponce, Computer Vision: A Modern Approach (2nd Edition), Prentice Hall 2011