Posted in Former Members, Members

PhD in Biomedical Engineering at The City College of New York (2008-2012)

Currently Enrolled in Computational Biology PhD. and specialized in the more specific field of Neuroscience and Signal Processing. I have studied the visual processing and object detection mechanism during free eye movement paradigms measuring synchronously Electroencephalography and Eye tracking.

Education

2006-2007: Internship in Plant Molecular Biology Group under supervision of Paula Duque, in Instituto Gulbenkian de Ciências, Portugal (January to May)
: Enrolment in PhD Program of Computational Biology in IGC, Portugal (September 2007)

2004-2005: Research in Mathematical Modeling, in Biophysics Lab in Faculdade de Ciência e Tecnologia, UNL, Portugal under the supervision of Prof. Teresa Moura

1998-2003: B.Sc and M.Sc in Applied Chemistry, Universidade Nova de Lisboa – UNL, Portugal

Research

 

My current interests are the areas of Machine Learning and Neuroscience. The research project is under the supervision of Professor Lucas C. Parra and is concerned with information processing in the visual systems. In the process we learn and develop methods on how to access the neural signal originated by the visual system. I am concerned also with the integration and usage of this informtation to develop Brain Computer Interfaces for search tasks.

There is evidence that our peripherial visual system is able to detect targets before looking directly at them. And we know that it is possible to correlate EEG signal with target detection in RSVP tasks where eye movements are unnecessary for target detection, can we use the EEG measurements to uncover the neural signatures that correlate with target detection with simultaneous eye movements?

The work currently involves devoloping Eye Tracking experiments with EEG simultaneous recording, using the following softwares: Experiment BuilderBiosemi ActiveViewPython and Matlab.

I am a ‘freak’ of automating data processing, anything data processing that can be automated has to be implemented like that! With a push from data to results!

Eye Movements and EEG experiments:

Ongoing experiments:

The experiment tries to acess the EEG neural correlates in free viewing conditions, associated with target detection. Subject are asked to search for a specific object among distractors for 240 trials.

We design three pradigm of increasing dificulty to study the neural correlates in different levels. The hypothesis s that for easy level it is possible to identify the neural signatures for target discrimination even before subject look at the target.

More over we study the neural correlates involved in the widely experienced behavior of inattentional blindness. That happens when we are looking at something but fail to see it. Are there any particular signatures in the EEG that can identify these particular moments?

We are trying to answer to these questions by recording simultaneous EEG and Eye movements.

 

Older projects :

1)

With the objective to study if the neural correlates present in the RSVP paradigm like the P300 are still present if the target to be detected is not present in the fovea. We developed an experiment to test detection of targets (helipads in areal images) with increasing distance from the center of the image with target. BY simultaneously recording EEG we wanted to titrate with eccentricity the neural correlates involved in this visual periphery detection.

In the task the subject was asked to look and detect the presence of a single helipad in a RSVP of 100 images at 10Hz. Performance of detection was measured against eccentricity. And EEG signal was used to see if it was possible to train a classifier for target detection, classification was measured at different eccentricities.

The images used for the stimuli were prepared using Python. The paradigm was implement with E-prime 2.0 Professional software, EEG was recorded using BioSemi equipment and the data was analyzed using Matlab.

2)

Multiple target detection, here the objective was to determine what was the threshold of multiple target detection. We prepared images that contained multiple targets.

In this task subjects would be asked to push a button each time they saw a target. The amount of time given to the subject would vary from 1s to 2s. Performance was measured against the time given. This was to determine how much time subject need to detect multiple targets in an image.

The images used for the stimuli were prepared using Python. The paradigm was implement with Experiment Builder software, eye movements were recorded using EyeLink II and the data was analyzed using Matlab.

 

If you have an interest in EEG or Eye movements, and would like to learn more about these topics, do not hesitate, email me I will be happy to give you a tour of the work we do in the Lab. We are always looking for new interested subjects.

Publications:

1. Dias JC, Sajda P, Parra LC.”When seeing and looking do not coincide:Scalp EEG correlates of found and missed targets during freeviewing”. to be Submitted.

2. Dmochowski JP, Sajda P, Dias J, Parra LC. “Correlated components of ongoing EEG point to emotionally laden attention – possible marker of engagement?”. Front. Hum. Neurosci. 2012; 6:112. abstract pdf

3. Dias JC, Parra LC. “No EEG evidence for subconscious detection during Rapid Serial Visual Presentation”.  IEEE Signal Processing in Medicine and Biology Symposium, Brookly, NY, December 10, 2011. pdf

4. Dias JC, Ferreira FC, Ferreira HG, Moura TF. “A mathematical model of the diluting power of the cortical thick ascending limb of the loop of Henle”. J Membr Biol. 2006; 214 (2): 59-73. pdf