Heart Failure Survival is a clinical visualization dashboard created in Tableau to explore patient survival outcomes related to congestive heart failure. The dataset is sourced from Kaggle, consisting of 299 patient records with clinical attributes like age, sex, serum sodium, blood pressure, creatinine levels, and comorbidities.All 299 patients had left ventricular systolic dysfunction and had previous heart failures that put them in classes III or IV of the New York Heart Association (NYHA) classification of the stages of heart failure.

This dashboard visualizes both categorical and continuous risk factors, helping understand their correlation with survival and mortality. It is intended purely as a technical showcase and does not make or suggest any medical decisions or interpretations.

• Tableau – For building dashboards and visual analytics
• Interactive Filters – For sex and outcome-based comparisons
• Charts Used – Histograms, bar charts, boxplots, stacked bar charts

• Analyze survival vs. death rates across different patient profiles
• Visualize the distribution of clinical parameters such as ejection fraction, CPK levels, sodium, and creatinine
• Compare risk factors like diabetes, hypertension, anemia, and smoking with survival outcomes
• Understand the effect of age and gender on heart failure mortality

1. Top-Level Metrics

   • Total Admissions
   • % of Deceased
   • Median Age (Deceased vs. Survived) by Gender

2. Comorbidities Overview

   • Diabetes, High Blood Pressure, Anemia, Smoking

3. Vital Signs and Labs

   • Serum Sodium Classification
   • CPK Enzyme Level Alerts
   • Serum Creatinine Analysis
   • Platelet Count Distributions

4. Cardiac Function

   • Ejection Fraction Categorization (Normal, Mild, HFrEF, Severe Dysfunction)

5. Demographics

   • Age Distribution by Outcome and Gender

1. Clinical Dashboards – Demonstrates the ability to build comprehensive dashboards for patient risk profiling.

2. Decision Support – Shows how Tableau can be used in medical environments for identifying mortality-related trends.

3. Health Insights – Highlights early warning parameters like abnormal CPK or low ejection fraction that could be integrated into predictive models.

4. Public Health Demonstration – Provides a non-invasive example of real-world health monitoring with anonymized data.

Dataset from Davide Chicco, Giuseppe Jurman: “Machine learning can predict survival of patients with heart failure from serum creatinine and ejection fraction alone. BMC Medical Informatics and Decision Making 20, 16 (2020) Dataset:Heart Failure Clinical Records Dataset – Kaggle

🔗 Click to View on Tableau Public