About me...
I am pursuing my Master of Science in Electrical Engineering from Indian Institute of Technology-Madras. As a Research Scholar I work at Healthcare Technologies Innovation Centre, @ IIT-M Research Park, under the guidance o Dr. Mohansankar Sivaprakasam and Keerthi Ram. My research lies at the intersection of Medical Image Processing, Computer Vision and Deep Learning. I work on deveoping Deep Learning based algorithms to speed-up MRI acquisition process. Apart from research in AI, I am a cinephile. I love going on long cycle rides.
Project Associate
Working in Collaboration with GE Healthcare to develop AI for detecting and eliminating artifacts in MRI scans.
Project Associate
Working in Collaboration with GE Healthcare to develop AI for detecting and eliminating artifacts in MRI scans.
M.S. , IIT-Madras
Electrical Engineering.
M.S. , IIT-Madras
Electrical Engineering.
Summer Student
Eastern European Machine Learning Summer School, Budapest organised by Google DeepMind
Summer Student
Eastern European Machine Learning Summer School, Budapest organised by Google DeepMind
B.Tech, VSSUT-Burla
Electrical Engineering.
B.Tech, VSSUT-Burla
Electrical Engineering.
Recent Updates
EEML Summer School
Eastern European Summer School on Machine Learning , Budapest, Hungary
- An amazing month of great lectures, fireside chats and presentations by engineers, researchers and top acamedicians.
- Presented poster: Less is more-A Self-Supervised Learning Approach to MR Image Reconstruction.
- Project: MediSinGAN-using Generative methods for medical dataset generation.
Featured Works
Automatic diabetic detection
Diabetic Retinopathy, is the leading cause of blindness among aged adult. Machine learning model to assist doctors in disease detection.
Fast MRI Reconstruction
By producing accurate images from under-sampled MR data, AI image reconstruction has the potential to make MRIs accessible for more people.
MRI Reconstruction Challenge
Comparing different DL-based MR reconstruction models. Assessing reconstruction model generalizability to various datasets acquired with a different number of channels.
MediSinGAN
Synthetic medical data generation across different imaging tasks, using Generative Methods.
Publications
Dual-Encoder-Unet For Fast MRI Reconstruction
- published in 2020 IEEE 17th International Symposium on Biomedical Imaging Workshops (ISBI Workshops), Iowa City, USA . [paper] [code]
Comparing Accelerated MR Reconstruction Models and Assessing Their Genereralizability to Datasets Collected with Different Coils
- Multi-channel MR Image Reconstruction Challenge, Medical Imaging with Deep Learning, Montréal,Canada.
- to be submitted in NeuroImage, Elsevier Journal. [paper] [our model] [our code] [MC-MRRec] [MC-MRRec code] [dataset]
