AI Clinical: elderly, disabled, low-income patients get care

Missed appointments cost the healthcare system $150 billion/year. People without reliable transport are 30% more likely to delay or forgo medical care. ~3.6 million Americans/yr no ride2 medical appts

Comment

Applying AI to bridge transportation and healthcare access for underserved populations ties directly into SDG 3 (Good Health and Well-being) and SDG 10 (Reduced Inequalities). Here’s a polished and complete version of your hackathon submission write-up using your given details:

Inspiration

Millions of patients—especially those who are elderly, disabled, or from low-income communities—miss critical medical appointments simply because they can’t get to the clinic. Missed appointments delay treatment, increase healthcare costs, and worsen health outcomes. Inspired by the UN Sustainable Development Goals (SDGs), especially Goal 3: Good Health and Well-being and Goal 10: Reduced Inequalities, we set out to solve one simple but powerful problem: How can AI help people get to the care they need—on time?

What it does

Our app integrates with a healthcare clinic’s appointment calendar and automatically: Sends a text reminder to patients about 30 minutes before their appointment. Asks for confirmation and verifies location. Uses an AI-powered bot to schedule a ride (Uber or Lyft) for patients who need transportation to their appointment. Tracks ride status and notifies the clinic when the patient is en route or has arrived. The result: fewer missed appointments, improved patient outcomes, and more efficient clinic operations.

How we built it

Frontend: Streamlit web app for clinic staff to view appointments and manage notifications. Backend: Python Flask API integrated with Twilio for SMS communication and Appium to avoid having to obtain Uber/Lyft APIs for ride scheduling. AI/Automation: Natural Language Processing (NLP) bot built using Snowflake's Cortex AI to handle patient responses and decision-making. Database: SecureConfig (custom class) connected to Firebase / PostgreSQL to store clinic, patient, and ride data. Deployment: Hosted on AWS / Streamlit Cloud for real-time clinic access.

Challenges we ran into

Handling asynchronous communication between text responses, ride scheduling, and appointment data. Ensuring data privacy and HIPAA compliance when dealing with patient information. Integrating multiple APIs (Twilio, Uber/Lyft, calendar systems) smoothly. Managing edge cases — e.g., patients replying late or changing addresses at the last minute.

Accomplishments that we're proud of

Built a working AI transport scheduling prototype in under 12 hours. Successfully automated ride requests through conversational AI. Created a real-world solution that directly supports the UN SDGs by improving healthcare accessibility.

What we learned

The power of combining AI + automation + empathy to solve real social problems. Practical lessons on secure data handling, API integration, and AI-driven workflows. How small-scale technical innovations can have massive community impact.

What's next for AI Clinical: elderly, disabled, low-income patients get care

Support elderly and disabled patients with mobility assistance options. Integrate insurance and telehealth data for coordinated care. Develop a mobile app version for both patients and clinics. Partner with local health departments and nonprofits to scale the solution nationwide.

Built With

Share this project:

Updates