Health Data BI Dashboard — Cardiovascular EDA

Dataset Snapshot

Patient Records 918 Anonymised observations Structured

Positive HD Cases 55% Heart disease present High prevalence

Avg. Patient Age 53.5 Years across dataset Middle-aged skew

Male Patients 79% of total sample Male-dominant

Avg. Resting BP 132 mmHg across cohort Elevated

Avg. Max HR 137 bpm during stress test Within range

Risk Factor Distributions

Heart Disease by Age Group

Proportion of positive cases across five-year age bands

Chest Pain Type vs Diagnosis

Distribution of angina types among positive and negative cases

Sex Distribution — Full Cohort

Male vs female patient split across the dataset

Fasting Blood Sugar Prevalence

Patients with fasting blood sugar above 120 mg/dL

Clinical Indicators

Resting BP Distribution

Frequency of resting blood pressure readings across the cohort

Cholesterol by Outcome

Average serum cholesterol (mg/dL) for positive vs negative cases

Max Heart Rate by Age Band

Average maximum HR achieved during exercise stress test

ST Depression (Oldpeak) Spread

Exercise-induced ST depression values — key ischaemia indicator

Key Analytical Findings

Age is the strongest demographic predictor

Positive heart disease cases concentrate in the 55–65 age band. Patients under 40 show materially lower prevalence, supporting age as a primary screening criterion.

Asymptomatic chest pain carries highest risk signal

Patients presenting with asymptomatic chest pain (Type 4) have a disproportionately high rate of positive diagnosis compared to typical or atypical angina presentations.

ST depression elevation flags ischaemia risk

Higher oldpeak values correlate strongly with positive outcomes. Patients with ST depression above 2.0 show notably elevated diagnosis rates across both sexes.

Reduced max HR aligns with positive cases

Patients diagnosed with heart disease tend to achieve lower maximum heart rates during exercise testing — consistent with reduced cardiac reserve capacity.

Male patients represent most positive cases

The male-dominant sample skews overall prevalence figures. Relative risk by sex should be interpreted with caution given the 4:1 male-to-female ratio in this dataset.

Cholesterol alone is a weak differentiator

Serum cholesterol shows limited separation between positive and negative groups. It becomes more informative when combined with resting BP and ST depression metrics.

Project Scope & Workflow

Stage	Activity	Tool	Output
Data Preparation	Cleaning, type standardisation, null handling	Power Query	Processed dataset
Data Modelling	Star schema, calculated columns, DAX measures	Power BI Desktop	Analytical data model
EDA	Distribution analysis, cross-tabulation, outlier review	Power BI + R Script visuals	Risk factor profiles
Visualisation	Dashboard design, filter panels, KPI cards	Power BI Desktop	Interactive report
Insight Layer	Trend interpretation, demographic breakdowns	Power BI + DAX	Stakeholder-ready views

Dashboard Screenshots

Dashboard introduction page — Overview — Introduction & project context

Data characteristics view — Data characteristics & variable summary

Risk factors dashboard page — Risk factor breakdown by demographic

Provincial distribution — Geographic distribution view

R Script visual integration — R Script visual — statistical distribution plot

Next Steps

Refined DAX measures

Introduce composite risk scoring measures that combine multiple clinical indicators into a single interpretable metric per patient group.

Enhanced narrative storytelling

Add guided tooltip explanations and callout annotations to help non-technical stakeholders interpret clinical terminology directly within the dashboard.

Expanded dashboard views

Build additional pages covering exercise tolerance metrics, ECG result patterns, and multi-variable risk profiles as v2 extensions.