.png)
Treatment Options Side-by-Side Navigator
A structured comparison experience that improves comprehension and appointment readiness with sourcing transparency and governance.
Overview
Newly diagnosed patients struggled to compare treatment options across fragmented content. I designed a side-by-side comparison experience with AI-assisted retrieval/summarization constrained to medically reviewed sources, with explicit sourcing and escalation.
Problem / Opportunity
What was broken (or missing), and why it mattered.
Content existed but was spread across multiple articles, making comparisons difficult.
Users needed quick, understandable comparisons to prepare for appointments.
AI summaries must not generate new claims; every summary should trace back to reviewed sources.
My role
Explicit ownership and responsibilities.
- Product strategy
- UX architecture
- AI retrieval framework planning
- Guardrail development
- Content governance planning
- Product specification development
- Success metric definition
- Stakeholder alignment
Process / Approach
Where the strategy, structure, governance, and tradeoffs live.
- 1) Comparison model
Defined core dimensions (mechanism, administration, side effects, schedules, typical contexts) and a table interaction model.
- 2) Constrained retrieval + summarization
Planned summaries grounded strictly in reviewed content, with no novel medical claims and explicit linking back to sources.
- 3) Governance + escalation
Created review pathways, content versioning expectations, and fallback logic to avoid speculative outputs when content is incomplete.
- 4) Decision support surfaces
Added “Questions to Ask Your Doctor” support patterns to improve confidence without providing advice.
- 5) Measurement plan
Defined success metrics (exports, engagement depth, return usage, qualitative helpfulness) to validate value.
The safest product is one that cites and compares reviewed information — not one that “reasons” about patient-specific recommendations.
Visual systems diagrams
Placeholders for architecture maps, flows, governance models, and lifecycle diagrams.
How each summary ties back to medically reviewed sources, with review and escalation pathways.
Select options → generate comparisons → export + doctor-question builder.
Technology
Tools and systems involved.
- AI retrieval systems
- Structured summarization workflows
- UX architecture
- Content governance frameworks
- Figma
- PDF export workflows
- Analytics planning
- AI guardrail systems
Outcome / Impact
The organizational and product-level effect.
- Created a structured AI-assisted educational navigation framework
- Established responsible AI governance models for healthcare education
- Reduced cognitive friction in treatment comparison workflows
- Improved conceptual readiness for medical appointments
- Built foundations for future AI-assisted educational products
Start with an AI Ops audit to identify high-ROI workflows, governance needs, and a pragmatic 30/60/90-day plan.