Inspiration
Pulse Predict was born from a simple but harsh reality: cardiovascular diseases and strokes are not killing people because they are untreatable, but because they are detected too late.
In Côte d’Ivoire and many African countries, patients often arrive at hospitals when the damage is already done. Not because doctors are incompetent. Not because treatments don’t exist. But because there is no continuous monitoring, no early warning system, and no real-time detection of physiological anomalies.
We were inspired by this gap between medical knowledge and real-world access to prevention tools. Our goal was clear: build a system that moves healthcare from reaction to anticipation.
Not a gadget. Not a smartwatch clone. But a real preventive system designed for institutions, not just individuals.
What it does
Pulse Predict is a connected biometric t-shirt system designed to help institutes and healthcare structures detect early cardiovascular risks. It combines: A smart wearable t-shirt that captures vital physiological signals A cloud infrastructure that stores and organizes medical data An AI model that analyzes variations and detects abnormal trends A platform for healthcare institutes to monitor patients in real time The system focuses on: Heart rate Oxygen saturation (SpO₂) Body temperature Movement and posture Physiological trend analysis Predictive risk detection The objective is not to replace doctors, but to give them early signals before critical events happen.
How we built it
We designed Pulse Predict as a multi-layer system: Hardware layer Biometric sensors integrated into a textile structure A wearable electronic core for data acquisition Modular design for comfort, washing, and maintenance Communication layer Secure wireless transmission of data Real-time synchronization with cloud infrastructure Cloud layer Centralized storage of biometric data Real-time database management Secure access control AI layer A machine learning model (Random Forest architecture) Trend detection on physiological signals Risk classification based on data patterns Predictive logic instead of simple monitoring Platform layer Web interface for institutes and professionals Educational interface for families (non-medical interpretation) Alert and notification system Everything was built with one philosophy: practical, scalable, and deployable in real conditions.
Challenges we ran into
We faced real engineering problems, not just presentation problems: Sensor noise and instability caused by body movement Thermal issues with embedded electronics in wearable systems Data reliability from low-cost physiological sensors Cloud integration and real-time synchronization AI training limitations due to lack of local medical datasets Ergonomics: comfort, skin contact, irritation, washing constraints System complexity: hardware + cloud + AI + platform integration Also, a big human challenge: Explaining a complex system simply, under pressure, in front of a jury.
Accomplishments that we're proud of
Building a real functional prototype Designing a complete architecture (hardware + cloud + AI + platform) Creating a coherent medical logic, not just tech stacking Winning the Hackathon CID (Licence Category) Being selected for post-hackathon incubation and development Transforming an idea into a structured, scalable innovation project Not a concept. Not a mockup. A real system vision.
What we learned
We learned that innovation is not about complexity. It’s about usefulness. We learned that: Prevention is more powerful than treatment Data without interpretation is useless Technology without context is irrelevant AI without real-world integration is just math Confidence matters as much as competence Systems matter more than components And most importantly: Healthcare innovation must adapt to reality, not copy foreign models.
What's next for Pulse Predict
ext steps include: Advanced AI training using larger datasets Deployment on cloud AI platforms for scalable training Improvement of prediction models Development of clinical dashboards Integration of additional physiological indicators Field testing with healthcare institutions Medical validation partnerships Industrial prototype design Certification and regulatory alignment Our vision is clear: Pulse Predict is not a hackathon project. It’s a healthcare system in construction. From prevention. From data. From intelligence. From Africa.
Log in or sign up for Devpost to join the conversation.