Opal
Custom desktop web application, Opal was designed to transcribe high-fidelity transcripts from ambient recordings from patient encounters and to extract and label key patient health data enriching the gov. agencies Knowledge Management System.
Role: Lead designer
Squad Composition:
Lead Designer
Jr. Designer
Business Technical Leader
Solution Architect
AI Engineers
Front End Developer
Platform Engineer
Timeline:
January - March 2024
Impact: Reducing time spent on documenting patient encounters by 95%.
Lean UX Design & Research, Collaborative Design, Visual Design
Background
This pilot was developed as part of an AI Tech Sprint which is a challenge hosted by a large government agency to foster innovation tailored to addressing healthcare challenges. These sprints bring together industry, academia and the government agency. This agency faced critical challenges in delivering efficient and high-quality healthcare to their patients and the need for precision in capturing and processing patient health data across care settings.
The goal was to alleviate these pain points to enhance to patient care, reduce administrative burden and prevent clinician burnout in turn improving healthcare outcomes for their patients.
Discovery
I started off with some lean user research from collecting information from a physician who works at the client agency to map their day to day processes and understand the problems from their perspective.
Discovery Phase
Learn about client problem and interview physicians to hear about their first hand pain points and day to day experience around how they are currently taking notes and inputting them into their systems.
Process
Prepare Phase
Start crafting experience prototype. (Wireframing, prototyping, and visual design). Iterate with peers to ensure feasiblity and stakeholder acceptance.
Build Phase
Collaborate with engineers to bring our technologies value to life.
Prepare Stage
As lead designer on this project two of the biggest challenges were time and ensuring we met all requirements given by the government agency. Since it was an iterative challenge we built the application in chunks separated by different gates dictated by the customer.
I kept things as collaborative and low fidelity as possible until we had things more solidified in order to save time. I worked collaboratively with a solution architect and a junior designer on these wireframes in Mural.
Screenshot from our lo fidelity wireframing sessions with the cross-functional team in Mural. We kept things highly collaborative and low-fidelity to flesh out wireframes to include user flows and all the requirements needed.
Artifact from Persona Map activity to further understand my target user’s goals, needs, tasks and pain points.
As-Is Scenario map to understand and create a detailed picture of our target user’s day to day current situation. Including Steps, different roles, what actions are being taken, any data and systems being used and any associated pain points.
High Fidelity Protoype
After making progress and iterating on the low fidelity wireframes we then started bringing things into Figma to work more high fidelity.
Screen recording of the protoype
Click each thumbnail for larger view.
Login Screen
New Encounter Flow-Multi Step:
Confirm Identity
New Encounter Flow-Multi Step:
Playback Recording & Submit
Homepage View
New Encounter Flow - Multi Step:
Select Provider Type & Start Recording
New Encounter Flow - Multi Step:
Confirmation of submission
New Encounter Flow - Multi Step:
Recording
New Encounter Flow - Multi Step:
View recording
Impact
This project resulted in greater efficiency in Clinical documentation, reducing time spent on documenting patient encounters by 95%. Improvement in patient care and clinician satisfaction, customer satisfaction increased 50%.