Publications

Selected publications:

Bayesian Non-parametric Models for Synchronous Brain-Computer Interfaces: In this work a method for modeling temporal dynamics of brain signals is presented. The Dirichlet Process HMM automatically selects the number of hidden states directly form data. This is traduced in an increased performance in the classification of imaginary motor tasks from EEG signals. [Arxiv.org, 2016 ]

Asynchronous decoding of finger movements from ECoG signals using long-range dependencies conditional random fields: In this work a method based on Conditional Random Fields (CRF) is presented for asynchronous classification of finger movements from Electro-corticographic (ECoG) recordings. The CRF models extrinsic dynamics of the motor task executed by the subject, improving effectively the classification performance. [Journal of Neural Engineering, 2016].

Word-level language modeling for P300 spellers based on discriminative graphical models: In this word a framework for a BCI based on the P300 Speller is presented. A graphical model that includes modeling of the language at the level of words is used to predict the intention of the subject while writing words. The system provides a highly significant improvement in the number of letters correctly decoded per minute compared to traditional methods. [Journal of Neural Engineering, 2015 ]

Discriminative Methods for Classification of Asynchronous Imaginary Motor Tasks from EEG Data: In this work a probabilistic graphical model able to model intrinsic and extrinsic dynamics of the EEG signals during motor imagination is presented. Extrinsic dynamics are model using training data and then used as prior information for the prediction of the execution of the movement task in an asynchronous BCI. Results show a significant increase in the performance of the system. [IEEE Transactions on Neural Systems & Rehabilitation Engineering, 2013]

A latent discriminative model-based approach for classification of imaginary motor tasks from EEG : In this work a method based on the Hidden Conditional Random Field probabilistic graphical model is proposed for modeling of the temporal dynamics of the brain signals during imagination of motor movements. The method define hidden states that describe the course of the brain signals in particular frequency bands. This states are learned for each condition and in the testing state the model with higher likelihood to have had generated the data define the type of imagination executed for the subject. [Journal of Neural Engineering, 2012]

other publications

Delgado Saa, Jaime F. Bouton, Sophie. Giraud Anne-Lise. Spatio-Temporal Brain Signal Dynamics During Incongruent Audio-Visual Integration. Alpine Brain Image Meeting. Switzerland. 2017.
Delgado Saa, Jaime F. Using low frequency components to predict speed and position of fingers during execution of a motor task. Neuroscience 2016. Sand Diego, USA.
Cristancho, Jorge. Delgado Saa, Jaime F. A Bootstrapping Method for Improving the Classifier Performance in the P300 Speller. Neuroscience 2016. San Diego, USA.
Olivares, Cristian. Delgado Saa, Jaime F. A Novel Method for Fast Detection of SSVEP from EEG Recordings. Neuroscience 2016. San Diego, USA.
Cancino, Sandra. Delgado Saa, Jaime F. Classification of visual stimuli from Electrocorticographic recordings using stacked auto-encoders. Neroscience 2016. San Diego, USA.
Delgado Saa, Jaime F. and de Pesters, Adriana and McFarland, Dennis and Çetin, Müjdat. A probabilistic graphical model for word-level language modeling in P300 spellers. In: 6th International Brain-Computer Interface Conference 2014, Graz, Austria
Delgado Saa, Jaime F. Müjdat Cetin, Hidden Conditional Random Fields for Classiﬁcation of Imaginary Motor Tasks from EEG Data. Submitted to 19th European Signal Processing Conference - EUSIPCO 2011.
Danny Plass-Oude Bos, Bram Van De Laar, Delgado Saa, Jaime F., et al. Wild Photo shoot: Applying Overt and Covert Attention. Biennial Conference of the Society of Applied Neuroscience 2011.
Danny Plass-Oude Bos, Bram Van De Laar, Delgado Saa, Jaime F., et al. Looking Around with Your Brain in a Virtual World. IEEE Symposium Series on Computational Intelligence, 2011.
Delgado Saa, Jaime F., Miguel Sotaquirá, EEG Signal Classification Using AR-Power Spectral Features and linear Discriminant Analysis: A Brain Computer Interface Application. Eighth LACCEI Latin American and Caribbean Conference for Engineering and Technology, 2010, Perú.
Delgado Saa Jaime F., Blanco Sandra, Zamudio Angelo, (telemetry system for foot pressure monitoring, in Spanish, IFMB Proceedings, Venezuela, Proceedings Of The International Federation For Medical And Biological Engineering Springer-Verlag Berlín Heidelberg. 2007.
Delgado Saa Jaime F., Eric Vallejo, Jaime Torres, Design and implementation of an Electromyography signal acquisition system, (In Spanish), The Fifth Latin-American and Caribbean Conference for Engineering and Technology. México. 2007.
Delgado Saa Jaime F., Carlos Jaimes, Ingrid Oliveros, Electrodos para la Rehabilitacion de Hemiplejias", Latin-American and Caribbean Conference for Engineering and Technology. Colombia 2005.

Google Sites

Report abuse