rohit-pardasani

DeblurCNN

Convolutional Neural Network to deblur the images. A 20 layer DnCNN trained to remove blur introduced by averaging filter.

PhysionetSepsisChallengeSub-3

First submission to Physionet Sepsis Challenge 2019. The method uses segmentation of multivariate time series using CNN

ECG1DCNNV2

CinC Challenge 2020 (Team AI Strollers, Classification of ECG Signal using 1D CNN, Version 2)

RDQuantization

Quantiative Assessment of Respiratory Distress Severity

RDSegmentation_v1

Quantitative Assessment of RD using segmentation of multi-variate time series

rohit-pardasani.github.io

TimeSeriesSegmentation

Convolutional Nerual Network to segment time series. Comprising of only Conv,BatchNorm,Activ layers, so once trained can be used for variable length time series.

DAV-Assignment

DeepSino

Super Resolution of Undersampled Noisy 2D Sinogram using Deep Learning. It uses replica of DnCNN (Denoising Convolutional Neural Network) for denoising and super-resolution of sinogram. The project also contains a MATLAB GUI that has trained kereas network at backend to allow comparison between various reconstruction methods.

ds200

Repository for DS 200 plot assignment

MedicalAppointmentsNoShow

Predicting No Show for Medical Appointments

SuRDCNN

Super-Resolution Denoising Convolutional Neural Network: Image Denoising and Super-Resolution using Residual Learning of Deep Convolutional Network. A replica of DnCNN (Denoising Convolutional Neural Network) that does super-resolution and dnoising in single pass.

Tankaar

UnetModels

This is repository of Unet Models.