Mapped the full journey from entry → scan → AI result → history

• Counted the number of interactions required to complete core tasks

• Identified friction points where users felt confused, overloaded, or blocked

This helped uncover major UX gaps:

• Inconsistent UI patterns across screens

• Poor visual hierarchy in AI-generated reports

• Cognitive overload due to dense medical text

Design Approach & Process

I followed an iterative UX process, refining the product step by step:

• Flowcharting the experience- Rebuilt the entire end-to-end flow to reduce unnecessary steps and make the journey more intuitive.

• Wireframing & iteration- Created wireframes to restructure layouts, improve hierarchy, and simplify interactions before moving to visuals.

• Visual redesign & consistency- Established a consistent UI system- typography, colors, spacing, and components to ensure clarity and trust throughout the app.

• Usability & accessibility testing- Tested readability, contrast, tap targets, and scan-ability to ensure users could understand reports quickly without strain.

Key UX Decisions & Features Implemented

1. Guest Scan Experience

One of the most impactful UX decisions was introducing a Guest Login flow.

• Users in a hurry can scan reports instantly without creating an account

• AI results are still saved in history

• Users can review past reports without logging in

• Data is deleted only when the user explicitly chooses to remove history

This respected urgency, privacy, and user control critical in medical contexts.

2. AI Report Readability

The AI report screens were redesigned to be easy on the eyes and easy to scan:

• Clear visual hierarchy for important findings

• Color usage to highlight critical insights without causing anxiety

• Structured layouts that allow users to grasp meaning without reading every line

The goal was to make AI explanations feel supportive, not overwhelming.

3. Consistency Across the App

The earlier version lacked UI consistency, which reduced trust. I unified:

• Navigation patterns

• Button behaviors

• Typography scales

• Spacing and layout logic

This resulted in a smoother, more predictable experience across all screens.

4. Testing & Validation

Testing focused on real-world usage scenarios:

• Multiple login/logout flows

• Location tracking validation

• AI accuracy checks across different report scans

• Edge cases like partial scans or unclear reports

These tests helped ensure the experience was reliable, not just visually clean.

Outcome

• Successfully launched LumyScan v1

• Improved usability, readability, and overall user confidence

• The app is now live on the Playstore

• Users can understand medical reports faster without immediately needing a doctor visit

  

Looking Ahead (V2 Vision)

For the next version, I proposed expanding beyond radiology reports:

• Blood reports

• Urine reports

• ECG and text-heavy medical reports etc

  And more medical summaries, while also enhancing the experience with anxiety-first design, simplified explanations, trend tracking over time, doctor-ready summaries, visual risk indicators, accessibility improvements, and preventive insights evolving the app from a scan tool into a calm, intelligent health companion that empowers users to understand their reports confidently and make informed decisions.

The idea is to create a single place where users can understand any medical report through AI, reducing anxiety and empowering informed decisions.

Conclusion

By restructuring the flows, improving visual hierarchy, and introducing thoughtful features like guest access and readable AI summaries, I helped transform the app into a more intuitive and user-friendly experience. This project strengthened my ability to design meaningful AI-driven experiences that feel human, trustworthy, and effortless to use.

“Creating clear direction with minimal interactions, and building around real user needs not just for users is what truly defines meaningful UX.”

Wanna know more about the project…