Amir Hosein Asadi – Computational Neuroscience Researcher

About

I'm researcher at Sharif Brain Center and currently studying decision making. I received my Master's under the supervision of Dr. Reza Ebrahimpour and Dr.S. Hamid Amiri.

I used GLMs to better understand the brain activities in high and low cognitive load while continuous multimedia is being played. My main interest is using computational methods to do model-driven research on neural data. In addition, I'm interested in developing and improving methods of analyzing neuroimaging data.

Outside of research, I enjoy stargazing, hiking, and lifelong learning.

Email: asaadi.amir [at] gmail [dot] com
Twitter

News

My first article "Effects and prediction of cognitive load..." was published in Scientific Reports (April 2024).
Successfully defended Master's thesis on "The relation between brain activities and educational multimedia properties" (Feb 2023).

Education and Teaching

Degrees

MSc in AI, Shahid Rajaee University, 2019–2022
Courses: Machine Learning, Pattern Recognition, Stochastic Process, Evolutionary Computation, Neural Networks, etc.
BSc in Computer Engineering, University of Sistan and Baluchestan, 2013–2017

Teaching Experience

NeuromatchAcademy, a computational neuroscience crash course (teaching assistant) (15 days)
Online course, 2022
[content] Topics covered: Modeling practice, Model fitting, GLMs, Dimensionality reduction, Bayesian statistics, Linear dynamical systems, Markov process, Decision making, Hidden Markov models, Kalman filter, Optimal control, Markov decision process, Reinforcement learning, Modeling real neuron, Modeling neuronal networks using Wilson-Cowan model, Causality, Deep learning (with Pytorch), Autoencoders
[course page]
[certificate]
Machine Learning
Advanced Computer Programming, two times
Computer Fundamentals and Programming, three times

Workshops

CNSP 2021–2022
SCAN 2021
CuttingEEG 2021
EEGLAB workshop (Ali Motie Nasrabadi)

Skills

Analysis of continuous EEG data

I have great experience working with EEG data while studying the dynamics and topography of neural components. First I use EEGLAB and its plugins such as Clean Rawdata ASR, ICLabel, and Adjust to preprocess the data then I use the mTRF toolbox to run cross-validation and find the best temporal response function to predict real data. I have learned these by EEGLAB and CNSP workshops.

Working with fMRI and ECOG data

In Neuromatch Academy summer school I had the chance to do projects with International teams and TAs, We have had team work together, to learn a little about literature and special concesrs about each kind of data, then load them in Google Colab and try to look and model them with tools such as MNE-Python and scikit-learn.

Publication

Manuscript titled:

Effects and prediction of cognitive load on encoding model of brain response to auditory and linguistic stimuli in educational multimedia

Help in the editing of published paper titled:

Assessment of Instantaneous Cognitive Load Imposed by Educational Multimedia using EEG Signals

Projects

Title	Supervisors	Stage	Methods
Neural Response to Daylight in Virtual Offices	Dr. Reza Ebrahimpour, Dr. Pegah Payedar-Ardakani	Data Analysis	EEG
Real vs Imaginary Hand Movements (ECOG)	Dr. Litvak, Dr. Zibman, Mohammad Rabiei	Presented (NMA)	MNE-Python
V4 Prediction from Natural Images (fMRI)	Dr. Ehsan Rezayat, Raul Hernandez	Presented (NMA)	scikit-learn
Persian Digit Recognition	Dr. Reza Ebrahimpour	Passed	Python, Deep Learning
Python Educational Multimedia	Dr. Samira Noferesti	Published	LaTeX, Video Editing
Library Management System	maktabkhooneh.org	Passed	Django

Amir Hosein Asaadi