patient-outcomes

This repository contains the code used for Temporal Pointwise Convolutional Networks for Length of Stay Prediction in the Intensive Care Unit (https://dl.acm.org/doi/10.1145/3450439.3451860).

This mighty little deck of cards will change your health habits.

openPRO: Your health. Your voice.

Python toolkit for Medicaid claims data analysis — preprocessing, cleaning, risk adjustment (Elixhauser, CDPS-Rx), quality measures (PQI, BETOS, low-value care), and patient-level analytic file construction for MAX and TAF CMS data. Built on Dask for scalable processing of large-scale healthcare claims datasets.

CASSANDRA Retro is a comprehensive retrospective dataset covering the complete perioperative course of visceral surgical patients. It comprises four components – preoperative, intraoperative, postoperative, and laboratory – spanning 13,225 patients across clinical settings, with a total of 51,967,074 data points.

Patient facing app for generating health data on iOS devices and reporting back to the practitioner/health system

COVID-19 Critical Event Prediction Tools

Production-grade ML system predicting ICU 30-day readmission risk (MIMIC-IV, 48K patients). Complete pipeline: data cleaning → feature engineering (181 features) → LightGBM (0.79 AUC) → deployed Streamlit app. Healthcare AI portfolio project.

Pearson's Chi-Square Test of Independence for NYHA and KCCQ

Identifying clinical drivers for hospital readmissions in diabetic patients to improve patient outcomes using Python and Tableau.

Kaplan-Meier for Congestive Heart Failure Analysis

This project analyses NHS Referral to Treatment waiting times across the 2025/26 financial year to identify where the 92% constitutional standard is being missed and by how much. It quantifies backlog reduction, long waiter trends, and performance variation across specialties, trusts and ICBs, via Python, SQL and an interactive Power BI dashboard.

Discovering healthcare insights through data-driven analysis of EHRs. This project comprises three subprojects: data cleaning & preparation, exploratory data analysis and lastly, predictive modeling for in-hospital mortality and clinical risk stratification.

Exploratory data analysis of hospital patient outcomes using Python (Pandas, Matplotlib) to identify trends in length of stay, diagnosis, and readmission risk.

Predicting primary biliary cirrhosis patient outcomes with different machine learning algorithms.

Real-time healthcare & IoT data platform — Synthea + wearables → Kafka → AWS Glue Streaming → Iceberg v3 → ML predictions. HIPAA-governed. Zero-AWS local dev.

SQL + Tableau analysis of 101,766 patient records from 130 U.S. hospitals identifying key drivers of 30-day readmission rates across age, medication count, and ICD-9 primary diagnosis.