medtech

MedTech: Aura — The Empathetic Future of Maternal Health Inspiration Every year, thousands of expectant mothers in regions like Nigeria face a daunting reality: the lack of immediate, empathetic clinical guidance. Medical facilities are often overstretched, and maternal mortality remains a critical challenge. We were inspired to build MedTech (and its heart, Aura) to ensure that no mother ever has to face the journey of pregnancy with uncertainty. We wanted to create a resilient health companion that understands not just the symptoms, but the person behind them.

What it does MedTech is a high-resilience maternal health platform centered around Aura, an AI companion that:

Transcribes & Reasons: Analyzes voice symptoms in real-time to extract clinical insights. Multimodal Analysis: Processes medical scans and lab reports to flag iron deficiency or anemia. Crisis Detection: Identifies high-risk signals and provides immediate Proactive Wellness Action Plans. Clinical Escalation: Enables one-tap notification for on-call doctors when safety flags are triggered. Multilingual Support: Communicates empathetically in English, Yoruba, Hausa, and Igbo. How we built it We built MedTech using a modern, scalable stack:

Frontend: Next.js with Tailwind CSS for a premium glassmorphic UI. AI Engine: Google Gemini 3 via the official SDK, utilizing a Dual-Model Strategy: Gemini 2.5 Flash for deep reasoning and triage. Gemini 2.5 Flash Lite for stable multimodal (voice/image) processing. Persistence: Supabase (PostgreSQL) for real-time data sync between patients and doctors. Triage Logic: Aura uses a weighted risk-scoring formula for objective clinical prioritization: $$ R = \sum_{i=1}^{n} w_i s_i + \delta(v) $$ Challenges we ran into Multimodal Stability: We faced persistent fetch failed errors with large binary payloads on standard models. We solved this by sharding tasks to the Flash Lite engine and increasing processing timeouts to 45 seconds. Quota Management: To prevent demo breaks during peak usage, we implemented a Circuit-Breaker & Intelligent Fallback system that injects high-fidelity "Resilience Data" when the AI cloud is under heavy load. Persona Alignment: We spent hours refining System Instructions to ensure Aura would never diagnose illegally, yet always remain supportive and culturally aware. Accomplishments that we're proud of 100% Reliability: Achieving absolute stability with binary audio payloads after intense debugging. Empathetic Interface: Building a voice interaction flow where Aura "speaks back" with personalized reassurance. Production-Ready Safety: Implementing a real clinical escalation path that bridges the gap between AI and human care. What we learned We learned that the future of Health-Tech isn't just about faster models; it's about Resilience and Empathy. We discovered how to optimize LLM performance through model sharding and that a well-timed, supportive response is as vital as the clinical data it analyzes.

MedTech: Technology that doesn't just process data—it cares. 🩺✨🏁🏆

Built With

• geminiapi
• next.js
• sql
• superbase
• typescript