Poteligeo iVA (Interactive Visual Aid)
CASE STUDY
CASE STUDY
CASE STUDY
CASE STUDY
Adaptive Life Companion Platform (Spec)
NeuroSphere (MS) + FlareCast (Autoimmune/RA)
A voice-first, predictive companion that helps people plan daily life with chronic illness using wearable + phone + environmental signals—built as a modular platform, not a single-condition tracker.
Project type: Spec concept (not built)
My role: Product & Experience Design Leadership (vision, system model, experience concept, storytelling)
Scope: Platform architecture + two condition modules + key screens + lifecycle loop
This is a conceptual case study. No product was built and no clinical claims are made.
The opportunity
Most trackers help patients record symptoms for their next visit. But chronic illness is lived in the in-between moments: fatigue, pain variability, brain fog, stress, and uncertainty—when users need decisions and support now, not charts later.
Patients face daily challenges like:


Most tools today:
What I set out to explore:
How might we shift from tracking → to predicting, planning, and preventing?
Design thesis

If we combine passive signals with lightweight conversation, we can reduce logging burden and produce daily “first value” moments:
-
Predict risk (with confidence + explainability)
-
Plan the day (based on energy windows)
-
Offer micro-interventions (small, doable actions)
-
Learn over time (personal baselines + patterns)
​
Why conversation matters:
Conversation captures nuance (timing, context, quality) and adapts to the user’s state—especially on low-energy days.
System loop: Collect → Predict → Guide → Learn
Wearables

HRV, sleep, gait/movement, stress signals
Mobile

Screen time, movement, GPS, 
app use
Environment
Weather, pressure, humidity, UV, allergens
Conversation
Mood, symptoms, focus, priorities

Collect
Wearables, mobile context, environment signals, and short conversational check-ins establish a personal baseline with minimal manual logging.
​
Predict
The system produces daily forecasts (risk / energy / clarity) with confidence levels and explainability—so users understand when guidance is strong vs uncertain.
​
Guide
Recommendations focus on the next best action: planning support, pacing suggestions, micro-interventions, and trigger-aware precautions—delivered in a calm, voice-first tone.
​
Learn
Lightweight reflection updates the model: what happened, what helped, and what didn’t—improving personalization without creating “homework.”
​
Outcome: less burden, more daily usefulness, and better personalization over time.
The solution: a platform, not an app

This concept is a central adaptive engine with modular tools that can be assembled into condition-specific experiences.
​
Shared primitives across modules:
-
Forecasting (risk / energy / clarity)
-
Planning (what’s feasible today)
-
Resilience support (micro-interventions)
-
Trigger intelligence (environment + behavior patterns)
Experience principles


This concept is designed to feel:
-
Warm (supportive tone vs clinical checklists)
-
Low-friction (passive signals + short check-ins)
-
Transparent (“confidence” and “why this”)
-
Actionable (recommendations, not dashboards)
-
Respectful (no nagging; user is in control)
Module 1: NeuroSphere (Multiple Sclerosis)
NeuroSphere supports daily variability in fatigue, cognition, and mobility by forecasting energy windows and reducing decision load.
Core engines (concept)




NeuroSphere: a day in the life
NeuroSphere is built for moment-to-moment usefulness:
-
Forecast the day with confidence
-
Recommend the smallest next-best action
-
Adjust plans proactively
-
Summarize + reinforce recovery

A day designed around prediction + planning—not retrospective reporting.
Module 2: FlareCast (Autoimmune / Rheumatoid Arthritis)
FlareCast helps users predict flare risk, pace activity using an energy budget, and respond to triggers (including environmental patterns)




FlareCast: a day in the life
FlareCast is built for proactive living between flares:
-
Anticipate flare risk and daily energy capacity
-
Guide pacing with the smallest supportive intervention
-
Adapt plans before symptoms escalate
-
Close the loop with recovery insight and learning

A day shaped around anticipation and pacing—not reacting after a flare.”
Why this is different than tracking
Traditional trackers answer
“What happened?”
This platform answers
“What should I do next—today?”

Differentiators:
-
Predictive forecasting (with confidence)
-
Multi-modal data fusion (wearable + phone + environment + conversation)
-
Planning support (reduce daily decision load)
-
Micro-interventions (small, achievable steps)
-
Modular design (multi-condition ready)
Responsible by design
This concept is intentionally framed as guidance, not diagnosis:
-
Clear uncertainty (confidence levels)
-
Explainability (“why this”)
-
User control (pause, edit, opt-out)
-
Non-judgmental tone
-
Accessibility-first layouts (readable hierarchy, large tap targets)
What I would validate first
MVP goal: prove daily usefulness and retention for one module.
Success signals (conceptual):
-
Faster time-to-first-value
-
Higher day-7 / day-28 return rate
-
User-rated usefulness of forecasts and check-ins
-
Reduced logging fatigue (less manual input required)
​
Experiment plan (example):
-
Compare “forecast-only” vs “forecast + planning”
-
Test check-in frequency to avoid alert fatigue
-
A/B “confidence + why this” vs no explanation
My contribution
I led the concept from ambiguity to a coherent product story:
-
Defined the product thesis and experience principles
-
Designed the modular platform model and module differentiation
-
Created the key screens and narrative flow for a portfolio-ready case study
-
Framed responsible UX considerations for a health-adjacent companion