jooyoun kang !
Info & Contact
→ Samsung Galaxy Unpacked 2023
→ The Invasion of Galaxy
→ Ogeul
→ Mediator
→ Journaling 101
→ Plant Your Feelings!
→ That’s How It Goes
→ Rhino Experiments
Mediator
: UX research & design solution for patients with limited English proficiency.
Collaborated with 2 students in RISD Industrial Design.
US federal law requires all healthcare organizations that receive federal funds to provide services in a language that patients with Limited English Proficiency understand. Despite this, health care systems have not adapted to best serve this population. Language and cultural barriers contribute to extensive health disparities experienced by immigrants and non-English speaking communities. Patients with limited English proficiency experience high rates of medical errors than English-proficient patients, with access to far fewer services and credible resources.
How might we provide a more effective and personalized communication experience for LEP patients, while addressing not just their linguistic, but also emotional and cultural needs?
Natural Language Processing (NLP) utilized Artificial Intelligence and Machine Learning to understand and respond to text or voice data in the same way humans do.
Mediator incorporates the NLP technology to convert the voice data to text, and to categorically summarize information.
Onboarding
Translator UI adapts to user’s proficiency level
Scenario 1. User and doctor speak in the same language
Based on the user’s proficiency determined from the onboarding process, the system UI adapts to the user.
For more English-proficient patients who would speak in doctor’s language, the system adapts the side-by-side UI where the patient can occasionally refer to the screen to get brief information on complex medical terms.
For less English-proficient patients who would speak in their preferred language, the system adapts the face-to-face UI that both the doctor and the patient can read and hear from, with every detail of the conversation translated.
Scenario 1. Medical terms
Scenario 2. Cultural contexts
Scenario 3. Medication information
Summary generation
Summary history
During translation, NLP detects words from the glossary to display realtime infographics.
After translation, NLP generates a visit summary from the dialog text collected from the translation’s speech recognition process.
With Neural Machine Translation (NMT), the summary can be cross-checked with documents given by the health care provider.
This design solution can be applied to more environments where complex jargon is used, and where cultural/past context is important.
next project
Jooyoun Kang is a designer/art director offering visual and experience-driven creative solutions. She is currently at Samsung Cheil Worldwide