Zonepulse
We brought fragmented hospital data into one real-time dashboard, using AI to surface insights that help leaders plan ahead instead of reacting too late.
Role
UX Designer
timeline
October 2024 – February 2025
Research & Deliverables
The deliverables delve into user research, information architecture, interaction design, and the integration of AI features. I worked closely with the product manager and developers to turn clinical insights into practical, easy-to-use designs. Our main focus was making complex healthcare data more accessible and understandable. One key improvement was adding predictive analytics, which helped users manage issues before they happened.
Interviews & Personas
Hospital Director: “We scramble when occupancy spikes, but data arrives too late.
ICU Head: “I need real-time alerts before we hit critical load.”
Compliance Officer: “Reporting is reactive—by then, it’s a crisis.” The primary goal of SwiftLogistics is to enhance operational efficiency, reduce costs, and improve decision-making in warehouse logistics.
Competitive Analysis
Benchmarked Epic Systems, Tableau Healthcare
Key Findings
Centralised dashboards with predictive alerts are rare in Swedish systems.
Unique statements by users
"When occupancy is trending upward, I want an alert so I can open surge capacity.” “Before shift change, I want to simulate tomorrow’s census to adjust staffing.” “Each quarter, I want a one-click compliance report to submit to regulators.”
Core Features
- Real-time bed and ICU usage with trend sparklines
- Project tomorrow’s capacity based on current trends
- Compare unit performance and staffing ratios side by side
- Critical Alert clarifications
The outcome
The key UI principles of the dashboard focused on deterministic, real-time visibility.
Key capabilities included:
Live staffing levels by unit and shift
Equipment availability and utilization status
Filters by department, role, and time window
Clear indicators for shortages and capacity limits
AI-assisted features focused on:
Trend detection: Identifying emerging staffing or equipment gaps before they became critical
Scenario highlighting: Flagging units at higher risk based on historical and current patterns
Planning support: Suggesting where attention may be needed, without automating decisions
AI outputs were always:
Clearly labeled
Explainable in plain language
Optional rather than mandatory
This ensured trust and avoided over-reliance on automated recommendations.
