Leveraging AI in CNS and Alzheimer's clinical trial recruitment.
TrialClinIQ is building the clinical trial recruitment data infrastructure of the future, using real-world health records and AI to accelerate patient enrollment, starting with CNS and Alzheimer’s trials. I designed the end-to-end product experience, from trial discovery through to recruitment decisioning.
.png)
The Problem
Clinical trial recruitment fails quietly. Sponsors lose months — sometimes years — not because patients don't exist, but because the systems for finding and qualifying them are broken.
Three things make it worse:
Health records live in silos.
Recruiters piece together patient eligibility manually, across disconnected systems, under time pressure.
Trial criteria are written for regulators
Interpreting complex eligibility protocols without clinical training is slow and error-prone.
Nobody has visibility.
Sponsors can't see what's happening across sites until it's too late to course-correct.
.png)
I owned product design end-to-end across two user types with very different needs:
Volunteers - patients or caregivers exploring trial options, often with no clinical background, navigating a process that feels opaque and intimidating.
Recruiters - clinical staff managing candidate pipelines, eligibility decisions, and enrollment targets across multiple sites simultaneously.
My mandate: make a complex AI-driven system feel clear, trustworthy, and fast for both.
The Process
Discover - My MSc in Pharmacology gave me a genuine clinical foundation to work from. Rather than relying solely on user interviews, I could identify real constraints the product needed to respect from the start.
Define - I mapped the core workflow as a clear linear journey: Trial Discovery → Patient Matching → Recruitment Decisioning. Each stage had distinct user needs and decision points that shaped everything downstream.
Design - I focused on making AI outputs explainable and trustworthy. Clinicians can review, interrogate, and override every match. Progressive disclosure keeps the interface from overwhelming users, moving them from summary → detail → full patient context at their own pace.
Deliver - In clinical settings, interaction errors have real consequences. Every state change, confirmation, and error message was designed with that weight in mind — fast and reliable, not just visually polished.
.gif)
For CNS and Alzheimer's trials where eligibility windows are narrow and patient populations are small, these failures are especially costly.
TrialClinIQ is currently in pre-launch.
Impact is measured through design scope and stakeholder validation:
- 80+ screens across volunteer and recruiter experiences
- 16 core flows from trial discovery to enrollment tracking
- 2 dashboard systems designed in parallel, consistent across both user types
The core design challenge was finding a common ground to surface AI matching results alongside medical records, giving clinicians everything they needed to confidently determine patient qualification in one view, without switching contexts or second-guessing the data.
-p-2000.png)
Impact
Design - I focused on making AI outputs explainable and trustworthy. Clinicians can review, interrogate, and override every match. Progressive disclosure keeps the interface from overwhelming users, moving them from summary → detail → full patient context at their own pace.
Deliver - In clinical settings, interaction errors have real consequences. Every state change, confirmation, and error message was designed with that weight in mind — fast and reliable, not just visually polished.
%20(2).gif)
This project sits at the intersection of healthcare systems, AI, and human-centered design. My design work focused on making complex systems usable under real-world constraints.
You can also see...
Apperr Website, 2024-2025
View ProjectGoSendeet, 2023-2025
View Project