Reconstruct high-dimensional spectral representations of tinnitus using reverse correlation

MRI reconstruction (e.g., QSM) using deep learning methods

This is a repository for CS4ML. It is a general framework for active learning in regression problems. It approximates a target function arising from general types of data, rather than pointwise samples.

This is a repository associated with the chapter book "Towards optimal sampling for learning sparse approximations in high dimensions" by Ben Adcock, Juan M. Cardenas, Nick Dexter and Sebastian Moraga to be published by Springer in late 2021, available at https://arxiv.org/abs/2202.02360

Research homepage

BART: Toolbox for Computational Magnetic Resonance Imaging

Assignment solutions for the course CS754 Advanced Image Processing, Spring 2024 at IIT Bombay

Single pixel camera emulation with pytorch

Course Project for CS754 Advanced Image Processing, Spring 2024

Deep Physics-Guided Unrolling Generalization for Compressed Sensing (IJCV 2023) [PyTorch]

Data Consistency Toolbox for Magnetic Resonance Imaging

Reverse correlation using linear regression and compressed sensing for uncovering the psychoacoustic tinnitus spectrum

Context-dependent Probabilistic Prior Information for Improved Compressed Sensing MRI Reconstruction

A recursive framework to enhance the efficiency of deep unfolding networks.

Python package for signal processing, with emphasis on iterative methods

Compiled sequence and reconstruction of the MP2RAGE + CS sequence for Bruker scanner

Code to reproduce the results from the thesis: "Compressed Sensing - Theoretical Foundations & Application in Magnetic Resonance Imaging"

Efficient Algorithms for L0 Regularized Learning

Privacy-Preserving Statistical Analysis of Genomic Data using Compressive Mechanism with Haar Wavelet Transform