Abhijeet Dhupia

Abhijeet Dhupia

Research Assistant @ Spectrum Lab

Indian Institute of Science, Bangalore

About Me

Hi, I am currently working as a Research Assistant under the direction of Dr. Chandra Sekhar Seelamantula at the Spectrum Lab, IISc. My research work focuses on Biomedical Imaging using Deep Neural Networks and Image Processing techniques. My colleagues and I are developing a one-stop end to end AI abnormality detection platform for Healthcare, a project funded by the Bill & Melinda Gates Foundation. During my undergraduate years, I worked on developing an automatic segmentation algorithm to delineate the lumen intima layer of common carotid arteries under the supervision of Prof. Harish Kumar.

Apart from work, I relish cooking, snapshotting, and watching Formula 1.

Interests

  • Biomedical Imaging
  • Computer Vision
  • Signal Processing

Education

  • B.Tech in Electrical Engineering, 2015 - 2019

    Manipal Institute of Technology

Publications

Automatic Segmentation of Lumen Intima Layer in Longitudinal Mode Ultrasound Images - EMBC 2020

The paper is published at the 42nd Annual International Conferences of the IEEE Engineering in Medicine and Biology Society held in Montréal, Québec, Canada, July 20th – 24th, 2020. Watch the presentation video below:

Experience

 
 
 
 
 

Intern

QPIAI

Apr 2020 – Sep 2020 Bangalore
Developed an end to end cloud-based AI platform to analyse the trails generated by Wireless capsule endoscope; a typical WCE procedure generates about 50 to 100K images. My work entailed designing and setting up the pipeline, which pre-processes the video, followed by the classification and segmentation tasks. Ultimately, selecting about 50 clinically relevant representative images per patient indicating the type of abnormality, which is then referred to the gastroenterologist for further examination.
 
 
 
 
 

Research Assistant

Spectrum Lab, Indian Institute of Science

Aug 2019 – Present Bangalore
I briefly worked on an IMPRINT project funded by the Government of India. The goal was to segment Optic Disc and Optic Cup using the neural architecture search based techniques. Next, I worked on a collaborative project with QPIAI to develop a cloud-based AI Healthcare platform. At present, I am applying Few Shot Learning models to segment gastrointestinal disorders and abnormalities in the trails generated by the wireless capsule endoscope under a project supported by the Bill & Melinda Gates Foundation.
 
 
 
 
 

Research Assistant

Manipal Institute of Technology

Dec 2018 – Jun 2019 Manipal
I worked on a medical imaging project for my undergraduate thesis under the supervision of Prof. Harish Kumar. We proposed a fully automated method for the segmentation of the lumen intima layer of the common carotid artery by optimizing a locally defined contrast function of an active oblong followed by post-processing using various image processing techniques. [Paper] [Thesis]
 
 
 
 
 

Intern

Celebal India Solutions Pvt. Ltd.

Dec 2018 – Jan 2019 Jaipur
As a part of industrial training I got to work with Salesforce platform which specializes in customer relationship management (CRM).

Posts

How to Setup a Desktop Environment on a Remote Linux Machine?

Are you using a remote Linux server and reminiscing the days you could interact with your system using a GUI? Well, the truth is you are not alone. Using terminal has its own challenges, and it might take you some time to learn to work your way around it efficiently.

Achievements

Fundamentals of Deep Learning

See certificate

Deep Learning Specialization

See certificate

Data Science Specialization

See certificate

Operating Systems and You: Becoming a Power User

See certificate

German A2.1 Level

See certificate

German A1 Level

See certificate

Contact

  • abhijeetdhupia (at) gmail (dot) com
  • Spectrum Lab, Indian Institute of Science, Bangalore, IN-560012