ICLR 2023 paper - ManyDG - Dataset processing and mode codes

An advanced data mining model to predict hospital readmission in dataset of diabetes patients.

Repo for the final project of UIUC's CS598 Deep Learning for Healthcare to reproduce the DeepNote-GNN model

Importance of HBA1c in predictive Modeling of probability of Hospital Re-admission (CAPSTONE PROJECT)

Tatva-AI assist healthcare professionals in identifying high-risk patients and implementing interventions to reduce readmission rates, ultimately enhancing patient outcomes

This is an Machine Learning Course Academic Project where we worked extensively on Understanding the Health Care Data & Developing Machine Learning, Neural Network Models

[ACL Findings 2026] Official Implementation of "RePrompT: Recurrent Prompt Tuning for Integrating Structured EHR Encoders with Large Language Models"

Analysis on Diabetic Patients’ Hospital Admission & Classification of Readmission

XGBoost pipeline predicting hospital Excess Readmission Ratios on FY2024 CMS HRRP data. SHAP explainability, 5-fold CV (R² 0.938), and a live Streamlit dashboard for per-hospital risk assessment.

Code for the paper "A Novel Hyperparameter Search Approach for Accuracy and Simplicity in Disease Prediction Risk Scoring".

To predict whether a patient will readmit withon 30 days

Our objectives are to predict the readmission of Diabetic Patients and uncover the underlying reasons behind those readmissions

Production-style diabetes hospital readmission prediction pipeline with leakage-aware preprocessing, XGBoost modeling, FastAPI serving, Streamlit demo, MLflow tracking, monitoring, Docker, and CI.

Machine learning project that predicts hospital readmission risk using patient data to improve healthcare outcomes and reduce costs.

Repo for applications of BERT in clinical settings

Predictive analytics on diabetic patient readmissions using dbt, DuckDB and Python – with explainability and clustering.

This project builds a machine learning pipeline to predict hospital readmissions within 30 days using electronic health record (EHR) data from diabetic patients

This end-to-end project predicts 30-day hospital readmission risk for diabetic patients using 100k+ encounters. I built a Scikit-Learn pipeline for data cleaning , deploying a Random Forest model with 0.94 recall.

Project on predicting whether and when will patient with diabetes be readmitted in hospital after the treatment.

Machine learning system to predict 30-day hospital readmissions using Logistic Regression, SMOTE, GridSearchCV, and threshold optimization.