Perinatal Companion

We empower underserved pregnant people with a private, multilingual AI companion that helps them understand symptoms, track their wellbeing, and confidently communicate with NHS services.

Comment

Inspiration

During our placements in maternity wards and mother-baby units, we repeatedly saw mothers struggle to communicate symptoms, understand medical information, and access support when they needed it most. Language barriers, cultural differences, and fear of being misunderstood often left women feeling isolated and unheard. Seeing these challenges firsthand inspired us to create a simple, multilingual AI companion that helps mothers understand their health, track concerns, and feel confident seeking help before small issues become serious ones.

What it does

Perinatal Companion is a privacy-first, multilingual AI assistant designed to support mothers throughout pregnancy. It provides daily wellbeing check-ins, explains medical terminology in plain language, helps users track symptoms and mood, and automatically creates appointment summaries that can be shared with healthcare professionals. The platform supports multiple languages commonly spoken across UK maternity services, helping bridge communication gaps between mothers and care providers while encouraging timely engagement with NHS services.

How we built it

We built Perinatal Companion as a lightweight Progressive Web App (PWA) to maximize accessibility across devices without requiring app store downloads. The frontend was designed around a chatbot-first experience with multilingual support and local-first data storage for privacy. We integrated large language models through OpenRouter to power translation, symptom explanation, conversation summarization, and automated journaling features. The system is designed to provide education, guidance, and NHS signposting while avoiding clinical diagnosis or triage.

Challenges we ran into

One of our biggest challenges was balancing helpful AI support with patient safety. We needed to ensure the assistant could provide useful information without crossing into regulated clinical advice or diagnosis. Another challenge was designing for diverse language and cultural needs while maintaining a simple user experience. We also had to carefully consider privacy, leading us to adopt a local-first architecture that minimises unnecessary data collection and storage.

Accomplishments that we're proud of

We are proud of creating a solution inspired by real experiences from our placements in maternity wards and mother-baby units. We successfully built a multilingual, privacy-focused platform that addresses genuine communication barriers faced by vulnerable mothers.

What we learned

This project taught us that many of the biggest challenges in maternity care are not purely medical—they are communication challenges. Through research and conversations informed by our placements, we learned how language barriers, health literacy, stigma, and digital accessibility can significantly affect maternal outcomes. We also gained valuable experience in responsible AI design, healthcare safety considerations, and user-centered product development.

What's next for Perinatal Companion

We would like to pilot the platform with healthcare professionals and service users to gather feedback and validate its impact. Future plans include deeper NHS integration, enhanced accessibility features, culturally tailored content, and evidence-based wellbeing resources that can support mothers throughout the entire perinatal journey.

Built With

Share this project:

Updates