Tanmay Nath

 

Tanmay Nath

Host Institution

Digital Cell Imaging Laboratories - DCILABS
Contact details
N.1.09, Vision Lab
University of Antwerp(CDE)
Universiteitsplein 1
Antwerp 2610, Belgium
+32 32652881
tanmay.nath@ua.ac.be

 

 

 

 

 

 

 

 

 

 

 

 

Background and previous research experience

I am a Biomedical Engineer from Indian Institute of Technology, Varanasi, India.  I worked at University of Iowa, USA on removal of interleaved artifacts in MRI .Later I joined National Brain Research Center, where I worked on fMRI study on “Recitation-related structural changes in the brains of Vedic Sanskrit Pandits”- Poster accepted in CNS -2013.

Main areas of interest

Developing algorithms and techniques to quantify the social networks of drosophila melanogaster. We have developed FlyTracker which is fully automated software for detecting and tracking multiple flies in the flyworld.  It incorporates the techniques of computer vision and machine learning to automatically segment the flies from the background. The videos are recorded in the flyworld, an imaging platform, which has a high speed camera with low resolution. The software segments the flies from the background and finds the position and orientation of all the flies.  It tracks all the flies and also maintains the identity of each of the fly thus allowing quantitative analysis of each fly behavior.  The software is limited to tracking 12 flies. The aim of this software is to automate the behavior analysis of the fly and quantify the obscured social networks. Thus it is also capable of computing the necessary parameters for finding the social networks in the group of flies.

FLiACT project

Computational Neuroscience, Computer vision, Machine learning, Social Network analysis. Dyadic relationships of the flies have been identified and quantified, however their social behaviors are not yet explored. Drosophila Melanogaster display transient social interactions in groups. These interactions are yet to be quantified. With the aid of Flytracker and Flyworld, we construct the social network of the interacting flies and quantify these interactions.