Computing, Robotics and Bionics 1
Main language of instruction: English
Other languages of instruction: Catalan, Spanish
|
Head instructor
The course will focus on the field of the neural engineering, a discipline at the frontier between neuroscience and engineering. First, fundamental concepts of neuroengineering will be studied, including: interfaces with the nervous system, neuroprosthesis, brain-computer interfaces (BCI) and electrophysiology records. Seminary classes are developed for theoretical training, computer practices. Biological neuronal systems and monitoring in the development of work and studies by students. On the other side, measurement and analysis techniques of neural data will be introduced applying artificial intelligence and signal processing. Seminar-type classes for theoretical training are combined with computer practices. Biological neuronal systems and monitoring in the development of work and studies by students.
To access the course it is required to have completed the following subjects:
Course 1 subjects
Calculation
Course 2 subjects
Computing
Fundamentals and Electronic Systems
Signal and Systems Theory
Biostatistics
Course 3 subjects
Neuroscience Applied to Orthoprosthesis
Block 1. Computation. Classical Machine Learning
1. Unsupervised learning (Clustering and dimensionality reduction/factorization)
1.1 Clustering algorithms
1.1.1 k-means clustering
1.1.2 Hierarchical clustering
1.1.3 Spectral clustering
1.1.4 Gaussian Mixture Model clustering (GMM)
1.1.5 Hiden Markov model clustering
1.2 Dimensionality reduction and factorization
1.2.1 Principal components analysis (PCA)
1.2.2 Non-Negative Matrix Factorization (NNMF)
1.2.3 Factor Analysis (FA)
2. Supervised learning (classification regression)
2.1 Regression algorithms
2.1.1 Linear regression
2.1.2 Non-liner regression
2.1.3 Logistic regression
2.1.4 Generalized Linear Model (GLM)
2.1.5 Regression trees
2.1.6 Gaussian Process Regression Model
2.2 Classification algorithms
2.1.1 Linear discriminant Analysis (LDA)
2.1.2 Support Vector Machines (SVM)
2.1.3 k-Nearest Neighbor (kNN)
2.1.4 Decision trees
2.1.5 Naïve Bayes
2.1.6 Neural networks
2.1.7 Bagged/Boosted decision trees
3. Blind source Separation
3.1 Independent Component Analysis (ICA)
3.2 Fast Independent Component Analysis (Fast-ICA)
Block 2. Bionics. Brain function and motor neuroprosthetics
1. Origin of the brain signal
2. Recording brain activity
2.1 Electrical brain measures
2.1.1 Extracellular recordings
2.1.1.1 Electroencephalogram (EEG)
2.1.1.2 Electrocorticography (ECoG) (Epidural, Subdural)
2.1.2 Intracellular recordings
2.1.2.1 Single Unit Activity (SUA)
2.1.2.2 Local Field Potentials (LFP) and Multi Unit Activity (MUA)
2.2. Magnetic brain measures
2.2.1 Magnetic Resonance Imaging (MRI)
2.3. Metabolic brain measures
2.3.1 Functional Magnetic Resonance Imaging (fMRI)
2.3.2 Functional near-infrared spectroscopy (fNIRS)
3. Analysis of brain activity
3.1 Spike Sorting
3.2 Raster plot decoding
3.3 Peri-Stimulus Time Histogram (PSTH)
3.4. Spike density function
3.5 Tuning curves and population vector
3.6 EMG signal processing
3.7 BOLD Signal (fMRI) processing
4. Origin of the muscular signal
5. Recording muscular activity
5.1 Electromyography (EMG)
5.2 PNS neural interfaces
6. Analysis of muscular activity
6.1 Muscular synergies from EMG signal
6.2 EMG processing and prosthetics arm control
6.3 Cuff ENG cumulative signal processing
TRAINING ACTIVITY Project-oriented learning is a method based on experiential and reflective learning in which the research process on a particular subject is of great importance. The aim is to resolve complex problems based on open solutions or tackle difficult issues that allow new knowledge to be generated and new skills to be developed by students. Lectures are the setting for: learning and managing the terminology and language structures related to each scientific field. Practicing and developing oral and written communication skills. And learning how to analyse the bibliography and literature on Bioengineering. Using guidelines to identify and understand the main ideas during lectures. This academic activity has been an essential tool in education since it first began and should have a significant presence within the framework of this degree programme. Reading texts with the aim of engaging critical thinking plays a fundamental role in learning for citizens who are both aware and responsible. An activity for outside the classroom. This activity means students can allow their knowledge to settle and rest, which is always necessary before beginning a new task. The professor sets out exercises and problems, helps students to progress in terms of the engineering process the design involves, and guides the student, thus partial goals are achieved that facilitate the incorporation of the theoretical knowledge acquired. |
METHODOLOGY Practical classes allow students to interact at first hand with the tools they will need to use in their work. In small groups or individually practical demonstrations will be carried out based on the theoretical knowledge acquired during the theory classes. In theory classes the fundamental and scientific knowledge that forms the basis of the knowledge and rigour that engineering studies require must be established. This teaching method is based on reflection, it can provide students with useful knowledge and skills to tackle problems efficiently in a shorter period of time. Individual work, involving study, the search for information, data processing and the internalisation of knowledge will allow students to consolidate their learning. |
COMPETENCES CB1 CB2 CB3 CB4 CB5 CE12 CE13 CE15 CE17 CE20 CE8 CG10 CG2 CG3 CG4 CG6 CG7 CT2 CT3 CT4 CT5 CT6 CT7 |
The final grade of the subject will be obtained as
Nota=0,4·Nef +0,3·Nlab+0,3·Ntreb
where
Nef : Final exam grade
Nlab : Lab grade
Ntreb : Course work grade
No partial exam.
To apply for the apt, it is essential to carry out the laboratory practices of the subject.
Important considerations:
[1] Dornhege, G. Millán, J.d.R., Hinterberger, T., McFarland, D.J., and Müller, K.-R. (eds.) (2007). Towards Brain-Computing Interfacing. Cambridge, MA: MIT Press.
[2] Wolpaw, J. and Wolpaw E.W. (eds.) (2012). Brain-Computer Interfaces: Principles and Practice. Oxford University Press.
[3] Farina et al. Introduction to Neural Engineering for Motor Rehabilitation. IEEE Press Series on Biomedical Engineering Book.
E: exam date | R: revision date | 1: first session | 2: second session: