Diptyajit Das

Email  /  CV  /  Google Scholar  /  Github  /  LinkedIn

Short Bio

I'm currently working as a scientific software engineer at Max Planck Institute for Empirical Aesthetics, where I work on the reproducibility of neuroimaging research and support the principles of open science data. For more details, please follow Cogitate.

In addition, I am finishing my PhD at Ruprecht-Karls-Universität Heidelberg, Germany. During my PhD, I worked as a Research Engineer at University Hospital Heidelberg and specialized in advanced multi-modal neuroimaging techniques, including EEG, MEG, and MRI (fMRI), under the supervision of Prof. Dr. med. Alexander Gutschalk. My research focuses on investigating the origin and function of the P300 brain marker.

Previously, I worked as a Research Assistant at Medical Imaging Physics (INM-4), Forschungszentrum Jülich GmbH, Germany.

My research interests include scientific computing, experimental design, signal and image processing, brain functional imaging (MEG, EEG, MRI/fMRI), clinical research, open-source software and hardware development, multi-dimensional data analysis, and simulation modeling.

COGITATE is an innovative Open Science project where researchers are teaming up in a registered collaboration to compare two top theories of consciousness: Integrated Information Theory (IIT) and Global Neuronal Workspace theory (GNW). For more details please follow the COGITATE website.

This project aims to verify the role of the retrosplenial cortex as a primary source of the P3 response to auditory stimuli, previously identified in our other study. We used combined magneto- and electroencephalography during a visual oddball task and analyzed responses to rare target and non-target stimuli. This work is currently under review.

What are the neural generators of P300 (P3) brainwave? Despite its ubiquitous presence, the generators of the P300 are controversial and not well identified. In this work, we compared source analysis of combined magneto-and electroencephalography (M/EEG) data with functional magnetic resonance imaging (fMRI) and simulation studies to better understand the sources of the P3 in an auditory oddball paradigm.

The primary goal of this project is to classify EEG data from P300 speller into target and non-target events based on the power spectral density (PSD) features extracted from EEG epochs. The classification model used is a Random Forest classifier, aiming for high accuracy in differentiating between these two classes.

Previous attempts to use parametric approaches to perform statistical analysis are mainly restricted because of unknown distribution of the MEG data. To avoid such difficulties in this work, cluster based non-parametric permutation techniques have been utilized. The main concept of these approaches is so generic that this provides the examiner the complete freedom to choose any test statistic that ultimately helps to quantify the experimental effects (i.e., changes in brain responses) in multiple dimensions (space-spectral-temporal).

A major problem in all MEG systems is the lack of information about the head positions during the MEG experiment. Here we have developed an Arduino microcontroller based cross-platform to track and control the head motion procedue during MEG recordings.

In this work, we have designed and developed a GSM (Global System for Mobile) based wearable smart system with 3-axis accelerometer and three lead ECG recording system. The whole system is wearable for patients and can be used as a real-time monitoring, self-diagnosis, and remote diagnosis tool as it will detect whenever there is an abnormal heart condition and a sudden fall situation.