This is the repository for Sepsis Prediction Project in CS6440 (Gatech).

Calculate qSOFA (Quick Sepsis-Related Organ Failure Assessment) Following Sepsis Guideline 3

Reinforcement Learning for optimal sepsis treatment policies

E6893 Final Project for egm2143, djf2150, and kr2741

This is the repository for paper "Improving Sepsis Treatment Strategies using Deep Reinforcement Learning and Mixture-of-Experts"

React-based web application for sepsis detection

Deep learning model for sepsis prediction using high-frequency data

Deep learning model for sepsis prediction using high-frequency data

Web application for sepsis clinical assessment scores using batch ICU patient data

To Detect Sepsis Disease using six Classifiers on clinical data

One of the top solutions for The 2019 DII National Data Science Challenge: https://sbmi.uth.edu/dii-challenge/. More details in the paper "An interpretable deep-learning model for early prediction of sepsis in the emergency department", published on Patterns 2021.

Code and Datasets for the paper "An Interpretable Risk Prediction Model for Healthcare with Pattern Attention", published on BMC Medical Informatics and Decision Making.

This repository contains the code for the project: Transcriptomic meta-analysis reveals up-regulation of gene expression functional in osteoclast differentiation in human septic shock

Code and Datasets for the paper "Estimating Individual Treatment Effects with Time-Varying Confounders", published on ICDM 2020.

Prediction of sepsis disease using LSTM network

Mobile and Web application for real time early prediction of sepsis(6 hours before the onset). I did pre processing, data visualisation , feature engineering and all the data science and machine learning work, I also handled Google Cloud and real time analysis. A flask app is built and deployed on heroku platform

A Python program for automated calculation and selection of fluids to manage conditions, e.g. hypernatremia, severe sepsis, severe dehydration, some dehydration, along with deducing MAP and drop rates easily.

An interactive CLI tool to select and calculate components to manage medical emergencies

Prediction of Sepsis

The early prediction of sepsis is potentially life-saving, and we challenge participants to predict sepsis 6 hours before the clinical prediction of sepsis.