ForeStall
🌲 STANFORD TREEHACKS
🩺 An app to predict medication errors -
before their devastating consequences.
Timeline:
36 Hours (02/17 - 02/19/23)
Role:
UI/UX designer
Team:
Tools:
Figma, Adobe Photoshop
🏆 1st Place: Intersystems IntegratedML Challenge
$2,000 Cash Prize
A PRESSING PROBLEM:
On average, every single patient in the United States will experience 1 significant diagnostic error in their lifetime.¹
Every year:
👤 400,000
patients affected²
💸 $20 billion
in costs³
🪦 100,000
deaths⁴
In a whirlwind 36 hours, our team designed, built, and integrated machine learning (ML) into an app to solve this problem.
⚡️ (Fueled by 30 minutes of sleep—but mainly adrenaline &, of course, lots of caffeine ☕️.)
OUR SOLUTION:
💡 ForeStall works to predict potential diagnosis errors through analyzing a patient’s risk level to a certain medication at both a macro and micro scale.
Searchable macro dashboard identifies high risk patients across entire hospitals
Practitioner micro dashboard identifies high risk patients for individual practitioners
Output patient risk level analysis by finding mappings of medication incompatibility using IntegratedML
Detailed patient profiles with thread of patient notes so that different practitioners can remain on the same page
Anonymous error reporting feature, addressing the stigma of not reporting medical errors due to fear of backlash
THE 36-HOUR PROCESS:
(abridged & illustrated)
Fri 2/17: 10 P.M.
Problem exploration
Research
Literature study
Observation
DISCOVER
Synthesizing research
Concept organization
Product requirements
DEFINE
Sat 2/18: 9 A.M.
User Flows
Wireframes
Prototyping
Style guide
DEVELOP
High fidelity design
Presentation
Pitching & storytelling
DELIVER
Sun 2/19: 10 A.M.
💡 We wanted our app to feel:
Clean
Modern
Intuitive
BRANDING/IDENTITY:
Life in the hospital can be chaotic, so I used cool, unobtrusive tones and minimal, Sans-Serif fonts to make using the app a calm, easy experience. (and not add to the chaos of a doctor’s already too-busy day)
Color & theme explorations:
Version 1:
Version 2: (the winner!)
Logo explorations:
Iconography:
PROTOTYPES!
Wireframed & designed using Figma
💡 Faces were generated with an AI program (but may or may not be named after my teammates, lol)
Hospital Dashboard:
Searchable patient database for at-a-glance risk analysis across an entire hospital
Easily identify patient demographics and dangerous medications
Patient Profile:
Search bar for incompatible medication powered by IntegratedML
Track detailed notes to ensure consistent care for patient across different practitioners
Practitioner Dashboard:
Detailed board view for each practitioner to check in on patients
Identify at a glance which patients have higher medication risk and need more diagnosis time
Hospital Analytics:
Track medication misdiagnosis statistics across entire hospital
Identify improvements over time as well as unresolved error areas
Anonymous Error Reporting:
Eliminate medical stigma around errors by allowing practitioners to report misdiagnosis anonymously
Allow for detailed description of medication errors
Error Database:
Allow hospital administrator to view entire database of all reported errors
Easily identify which medications were misdiagnosed, as well as important details of patient symptoms
PULLING EVERYTHING TOGETHER…
🌙 I put everything into a digestible, slide-deck format for the judges (at the very reasonable hour of 4 a.m. on Sun. Feb 19th)
Using the presentation deck below, as well as a live front-end implementation coded by Allen and Kevin’s IntegratedML work, we presented ForeStall to the judges and communicated our vision for the elimination of careless medical mistakes in hospitals.
Click through for our full presentation! ➡️
Although I had to take a massive nap afterward, the whole process of ideating, researching, prototyping, refining (basically the entire design thinking process fitted into 36 hours) was an amazing & fulfilling experience.
🏆 The most mind-blowing part is – we somehow took home a prize too! Shoutout to the entire InterSystems team for being some of the most dedicated and helpful company sponsors at the event.
FINAL THOUGHTS
Glad that I did not forestall coming to this hackathon… ✨
💤 Ignore the dark circles under our eyes haha … focus on our miraculous win and surprised smiles instead :)
I’m glad I ran into teammates who were absolutely incredible at programming & backend engineering so that I could focus mainly on the design and UX. However, this experience has also made me even more determined to expand my (currently somewhat limited) knowledge of technical frontend implementation so that I can play a bigger role in bringing my designs to life through code.
Citations
Committee on Diagnostic Error in Health Care, Board on Health Care Services, Institute of Medicine, The National Academies of Sciences, Engineering, and Medicine. Improving diagnosis in health care. In: Balogh, EP, Miller, BT, Ball, JR, editors. Improving diagnosis in health care [Internet]. Washington, D.C.: National Academies Press; 2015. http://www.nap.edu/catalog/21794.
James, J. T. A new, evidence-based estimate of patient harms associated with hospital care. J Patient Saf. 2013 Sep;9(3):122-8.
Henriksen, K., Battles, J. B., Keyes, M. A., & Grady, M. L. (Eds.). (2008). Chapter 2: The need for a national focus on patient safety. In Advances in patient safety: New directions and alternative approaches (Vol. 1: Assessment). Agency for Healthcare Research and Quality (US). https://www.ncbi.nlm.nih.gov/books/NBK499956/.
Singh H, Schiff GD, Graber ML, Onakpoya I, Thompson MJ. The global burden of diagnostic errors in primary care. BMJ Qual Saf. 2017 Jun;26(6):484-494.
Also check out …
Internship: Designing for User Communication
Product Strategy - Customer Acquisition