MediCASP

MediCASP is an AI-powered medical screening assistant that combines medical knowledge and advanced reasoning to provide accessible preliminary health assessments across multiple conditions.

Comment

MediCASP: Your AI-Powered Medical Screening Assistant

Inspiration

Healthcare accessibility remains a significant barrier for many seeking initial medical guidance. We developed MediCASP to bridge this gap, creating a supportive first step for individuals who might be hesitant to seek immediate professional care. Our goal is to provide a knowledgeable, empathetic starting point for understanding potential health concerns.

What it does

MediCASP is an AI-powered medical screening assistant that combines medical knowledge and advanced reasoning to provide preliminary health assessments. Using diagnostic criteria from established medical resources* it screens for multiple conditions including:

  • Arthritis
  • Autism Spectrum Disorder
  • COPD
  • Dementia
  • Hypertension/Hypotension
  • Hypoglycemia
  • Pneumonia

Through an interactive chat interface, it offers structured symptom assessment while maintaining clear boundaries about its role as a preliminary screening tool rather than a replacement for professional medical care. *Symptoms are based on medical resources such as DSM-5-TR (Diagnostic and Statistical Manual of Mental Disorders) for mental health conditions, Mayo Clinic diagnostic criteria, and an award winning research paper, Hear Me, See Me, Understand Me: Audio-Visual Autism Behavior Recognition, co-authored by one of our own members.

How we built it

Our technology stack integrates modern tools for a robust medical screening platform:

Frontend:

  • React and Next.js
  • JavaScript, JSX
  • HTML/CSS

Backend:

  • Node.js
  • Python

AI Integration and Logic Engine:

  • Google's Gemini AI API
  • sCASP/Prolog for medical reasoning

Security:

  • CLERK API for authentication and user account login

Challenges we ran into

Development presented several key challenges:

  • Harmonizing multiple technologies while maintaining code quality.
  • Implementing medically accurate assessments based on established diagnostic criteria.
  • Creating clear boundaries between AI-assisted screening and professional medical advice.
  • Ensuring ethical handling of sensitive medical information.

Accomplishments that we're proud of

Key achievements include:

  • Development of a comprehensive screening system based on established medical criteria.
  • Creation of an intuitive, empathetic chat interface that maintains professional boundaries.
  • Successful integration of complex medical knowledge with modern AI technology.
  • Implementation of ethical safeguards for medical screening.

What we learned

This project provided valuable insights into:

  • The complexities of medical knowledge integration in AI systems.
  • The importance of ethical considerations in healthcare technology.
  • The balance between accessibility and professional medical guidance.
  • The challenges of cross-discipline software development.

What's next for MediCASP

Future development plans include:

  • Expanding condition coverage while maintaining diagnostic accuracy.
  • Enhancing AI capabilities for more nuanced symptom assessment.
  • Developing secure healthcare provider integration options.
  • Implementing additional evidence-based screening protocols.
  • Adding more disabilities, disease, and disorders.

Webpage _The webpage backend is not yet operable due to time constraints on AWS servers and limited resources on Vercel basic tier. To use the s(CASP) language, we must send a query to a linux server, but this takes time. Unfortunately, Vercel times out before this completes.

Built With

Share this project:

Updates