SwasthAI

Inspiration

Healthcare information is often difficult for ordinary people to understand. Patients frequently receive blood reports, MRI reports, prescriptions, and discharge summaries filled with complex medical terminology. This creates confusion, anxiety, and poor treatment adherence because many patients do not fully understand their diagnosis, medications, or the next steps in their treatment journey. We were inspired to bridge the gap between doctors and patients by building an AI-powered platform that translates complex medical documents into simple, easy-to-understand explanations. Our goal was to empower patients with knowledge and make healthcare more accessible, understandable, and user-friendly.

What it does

SwasthAI is an AI-powered healthcare platform that helps users understand medical documents such as prescriptions, blood reports, MRI reports, discharge summaries, and other healthcare records.

Users can upload a medical document, and the platform:

• Extracts information using OCR and AI analysis

• Identifies diseases and medical conditions

• Explains findings in simple 5th-grade English

• Describes possible symptoms related to the condition

• Explains tests and procedures found in the report

• Explains medicines prescribed and their purpose

• Provides personalized next steps and healthcare guidance

• Generates report-specific FAQs for better understanding

Additionally, SwasthAI includes symptom checking, voice assistance, wellness tracking, community support, and gamified health engagement features.

How we built it

We built SwasthAI using a modern full-stack architecture.

Frontend: • Next.js • React • TypeScript • Tailwind CSS • Framer Motion

Backend: • FastAPI • Python

Database & Authentication: • Firebase Firestore • Firebase Authentication

AI Layer: • Google Gemini • OCR Processing • Prompt Engineering

The workflow starts when a user uploads a medical document. OCR extracts the text, which is then processed by Gemini AI. The AI identifies diseases, symptoms, medications, test purposes, doctor findings, and recommended actions before generating a simplified patient-friendly explanation.

Challenges we ran into

One of the biggest challenges was handling different types of healthcare documents because reports, prescriptions, discharge summaries, and test results all have different formats and structures.

Another challenge was ensuring that AI explanations remained medically meaningful while still being simple enough for non-medical users. We had to carefully design prompts to avoid overly technical language while preserving accuracy.

We also faced challenges in extracting meaningful information from scanned reports and prescriptions, especially when documents contained poor image quality, handwritten content, or complex medical abbreviations.

Finally, creating a structured and patient-friendly output that included diseases, symptoms, medications, and next-step recommendations required significant prompt engineering and iterative testing.

Accomplishments that we're proud of

We are proud that SwasthAI goes beyond basic report summarization and acts as a true healthcare document interpreter.

Key accomplishments include:

• Converting complex medical reports into easy-to-understand explanations

• Identifying diseases and conditions from uploaded healthcare documents

• Explaining medicines and their purpose

• Providing symptom awareness and healthcare education

• Building an accessible platform with voice assistance and wellness tracking

• Creating a scalable healthcare solution that can help improve treatment adherence and healthcare literacy

Most importantly, we created a solution that has the potential to reduce patient confusion and empower users to make informed healthcare decisions.

What we learned

Through this project, we learned how AI can be applied to solve real-world healthcare problems.

We gained experience in:

• OCR and document processing

• Prompt engineering for healthcare applications

• Building AI-powered user experiences

• Integrating frontend, backend, database, and AI systems

• Designing accessible healthcare technology

We also learned that healthcare applications require a careful balance between simplicity, accuracy, and responsible AI-generated guidance.

What's next for SwasthAI

Our vision is to evolve SwasthAI into a comprehensive AI healthcare assistant.

Future enhancements include:

• Retrieval-Augmented Generation (RAG) using trusted medical knowledge sources

• Multi-language and regional language support

• Telemedicine integration for doctor consultations

• Wearable device integration for real-time health monitoring

• Family health management dashboards

• Advanced medicine interaction and side-effect analysis

• Predictive healthcare insights based on user health trends

• Personalized wellness and recovery plans powered by AI

Our long-term goal is to make healthcare information understandable, accessible, and actionable for everyone.

Built With

• css
• fastapi
• tailwind