Overview

This repository contains the code and documentation for Assignment 1 of SPH-6004, where we build a predictor that estimates the patients' risk of kidney failure in the Intensive Care Unit (ICU). The assignment focuses on developing a predictive model using clinical data to help identify patients at higher risk of kidney failure, enabling early intervention and improved patient outcomes.

Dataset

The dataset used for this assignment is sourced from MIMIC-IV. It contains de-identified health-related data of over forty thousand patients who stayed in critical care units at the Beth Israel Deaconess Medical Center, which are used to train and evaluate the predictive model.

Methodology

• Data Preprocessing: The dataset underwent preprocessing steps such as handling missing values, encoding categorical variables, and scaling numerical features.
• Model Selection: Several machine learning models were considered and evaluated for their performance in predicting kidney failure risk. Models included Logistic Regression, Decision Tree, Random Forest, and Support Vector Machine.
• Model Training and Evaluation: The selected model was trained on the preprocessed data and evaluated using appropriate metrics such as accuracy, precision, recall, and F1-score.

Repository Content

• Assignment1_code.ipynb: Contains data preprocessing and all model implementation.
• Experimental Results.png: Table that shows model performance.
• Experimental Setup.png: Flowchart for model architecture.

Contributors

• LIN KUNSHI