neuroscience · slides · talks & lectures

Soft Introduction to Brain-Computer Interfaces

Slide 1 of 80: BRAIN-COMPUTER INTERFACE Ilya Kuzovkin 7 June 2014Slide 2 of 80Slide 3 of 80Slide 4 of 80Slide 5 of 80Slide 6 of 80: Now I know how your brain signal looks like when you think “LEFT” and “RIGHT”Slide 7 of 80: Now I know how your brain signal looks like when you think “LEFT” and “RIGHT” Try me — think orSlide 8 of 80: Now I know how your brain signal looks like when you think “LEFT” and “RIGHT” Try me — think or It was ! ! wasn’t it?Slide 9 of 80: Now I know how your brain signal looks like when you think “LEFT” and “RIGHT” Try me — think or How would you use such technology? It was ! ! wasn’t it?Slide 10 of 80: BRAIN-COMPUTER INTERFACE Mental intentionSlide 11 of 80: Mental intention Neuroimaging BRAIN-COMPUTER INTERFACESlide 12 of 80: Mental intention Neuroimaging Name some neuroimaging techniques BRAIN-COMPUTER INTERFACESlide 13 of 80: Mental intention Signal Neuroimaging BRAIN-COMPUTER INTERFACESlide 14 of 80: Mental intention Signal Data Neuroimaging BRAIN-COMPUTER INTERFACESlide 15 of 80: Mental intention Signal Data Algorithm Neuroimaging BRAIN-COMPUTER INTERFACESlide 16 of 80: BRAIN-COMPUTER INTERFACE Mental intention Signal Data Algorithm With 87% certainty I can say that you are thinking “LEFT” right now NeuroimagingSlide 17 of 80: NEURONS http://biomedicalengineering.yolasite.comSlide 18 of 80: NEURONS http://biomedicalengineering.yolasite.comSlide 19 of 80: NEURONS http://www.conncad.com/gallery/single_cells.htmlSlide 20 of 80: NEURONS http://en.wikipedia.org/wiki/Neural_oscillationSlide 21 of 80: NEURONS http://en.wikipedia.org/wiki/Neural_oscillationSlide 22 of 80: NEURONS http://en.wikipedia.org/wiki/Neural_oscillationSlide 23 of 80: NEURONS http://en.wikipedia.org/wiki/Neural_oscillation What is the frequency in this example?Slide 24 of 80: BRAINWAVESSlide 25 of 80: BRAINWAVES Delta 0-4 Hz Theta 4-7 Hz Alpha 7-14 Hz Mu 8-13 Hz Beta 15-30 Hz Gamma 30-100 HzSlide 26 of 80: BRAINWAVES Delta 0-4 Hz Theta 4-7 Hz Alpha 7-14 Hz Mu 8-13 Hz Beta 15-30 Hz Gamma 30-100 Hz slow wave sleep, babies, lesions children, drowsiness, meditation, relaxed closed eyes, relaxed motor neuron in rest, mirror neurons motor activity, anxious thinking, concentration…Slide 27 of 80: EEGSlide 28 of 80: EEGSlide 29 of 80: EEG TIME CHANNELSSlide 30 of 80: EEG TIME CHANNELSSlide 31 of 80: EEG Alpha 7-14 Hz Beta 15-30 Hz Gamma 30-100 Hz ? TIME CHANNELSSlide 32 of 80: EEG Alpha 7-14 Hz Beta 15-30 Hz Gamma 30-100 Hz ? TIME CHANNELS Jean Baptiste Joseph Fourier 1768 — 1830Slide 33 of 80: FOURIER TRANSFORM*Slide 34 of 80: FOURIER TRANSFORM* *discreteSlide 35 of 80: FOURIER TRANSFORM* *discrete =Slide 36 of 80: FOURIER TRANSFORM* *discrete signal at time t frequency complex number =Slide 37 of 80: FOURIER TRANSFORM* *discrete signal at time t frequency complex number = Amplitude of the component with frequency kSlide 38 of 80: FOURIER TRANSFORM* *discrete signal at time t frequency complex number = Amplitude of the component with frequency kSlide 39 of 80: DATASlide 40 of 80: DATASlide 41 of 80: DATASlide 42 of 80: DATA Why not like this?Slide 43 of 80: DATASlide 44 of 80: TIME-FREQUENCY DOMAINSlide 45 of 80: DATA Are we done?Slide 46 of 80: DATA Are we done? Hint:Slide 47 of 80: DATA 300 MSSlide 48 of 80: DATA 300 MSSlide 49 of 80: DATA 300 MSSlide 50 of 80: DATA 300 MSSlide 51 of 80: DATA 300 MS 11 channels 50 frequencies on each 3 seconds of data 300 ms window • How many numbers to describe 1 reading of 300 ms?Slide 52 of 80: DATA 300 MS 11 channels 50 frequencies on each 3 seconds of data 300 ms window • How many numbers to describe 1 reading of 300 ms? • How many numbers to describe all 3 seconds of data?Slide 53 of 80: DATASET INSTANCES FEATURES CLASSES . ! . ! .Slide 54 of 80: MACHINE LEARNINGSlide 55 of 80: MACHINE LEARNING ?Slide 56 of 80: set of sample objects ( samples ) is called training set Machine Learning algorithm learns from examples, MACHINE LEARNINGSlide 57 of 80: Each object ! ! ! ! ! ! ! ! ! ! ! ! can be described with a set of parameters called features MACHINE LEARNINGSlide 58 of 80: Tail length : ... f 1 MACHINE LEARNINGSlide 59 of 80: Furriness : ... Tail length : ... f 1 f 2 MACHINE LEARNINGSlide 60 of 80: Furriness : ... Tail length : ... f 1 f 2 f = ( f 1 , f 2 ) Form a feature vector MACHINE LEARNINGSlide 61 of 80: Instance Feature 1 Feature 2 Class Cat 1 8 cm 546 h/cm M Cat 2 7.5 cm 363 h/cm M ... ... ... Cat N 11 cm 614 h/cm F Together feature vectors and corresponding classes form a dataset MACHINE LEARNINGSlide 62 of 80: Feature vectors live in a feature space MACHINE LEARNINGSlide 63 of 80: ? MACHINE LEARNING Feature vectors live in a feature spaceSlide 64 of 80: MACHINE LEARNING Feature vectors live in a feature spaceSlide 65 of 80: MACHINE LEARNING Feature vectors live in a feature space K-Nearest NeighborsSlide 66 of 80: ! • Decision trees • C4.5 • Random forests • Bayesian networks • Hidden Markov models • Artificial neural network • Data clustering • Expectation-­‐maximization algorithm • Self-­‐organizing map • Radial basis function network • Vector Quantization • Generative topographic map •…Slide 67 of 80: BRAIN-COMPUTER INTERFACESlide 68 of 80: BRAIN-COMPUTER INTERFACESlide 69 of 80: BRAIN-COMPUTER INTERFACESlide 70 of 80: BRAIN-COMPUTER INTERFACESlide 71 of 80: BRAIN-COMPUTER INTERFACESlide 72 of 80: BRAIN-COMPUTER INTERFACESlide 73 of 80: BRAIN-COMPUTER INTERFACESlide 74 of 80: BRAIN-COMPUTER INTERFACESlide 75 of 80: BRAIN-COMPUTER INTERFACESlide 76 of 80: BRAIN-COMPUTER INTERFACESlide 77 of 80: BRAIN-COMPUTER INTERFACE ?Slide 78 of 80: BRAIN-COMPUTER INTERFACE ?Slide 79 of 80Slide 80 of 80: THE END
1 / 80

Soft Introduction to Brain-Computer Interfaces and Machine Learning

Set of slides presented at the Spring School organized by the Estonian Association of Psychology Students.

Download PDF