AIDERS: AI-powered Detection and Emergency Response System

Recognition of choking

Our journey as a diverse group of students with backgrounds from Morocco, Ghana, Afghanistan, and China brought us to HackHarvard with a shared desire to make an impact on the world of healthcare. Our experiences and those of our parents, who come from various parts of the world, made us acutely aware of the gaps in healthcare systems globally. Whether it’s a lack of access to emergency care or inadequate monitoring systems, these experiences deeply influenced our decision to work on something meaningful.

We were inspired by the Patient Safety 101 workshop, which highlighted the importance of addressing patient harm and the urgent need for innovative solutions to prevent medical errors. Hearing that nearly 70% of Americans have never even heard of patient safety made it clear that this is an area where technology can—and should—make a difference. The Pittsburgh Regional Health Initiative’s commitment to patient safety technology and its emphasis on reducing preventable harm pushed us further to take up this challenge and do our part.

Thus, we decided to create AIDERS—a real-time patient monitoring system utilizing a vision model to detect emergencies like falls or choking incidents in hospitals. By promptly alerting healthcare professionals and designated contacts, we aim to enhance patient safety and bring peace of mind to patients and families alike. Through countless hours of work on the technical stack—leveraging YOLOv8 for detection, combining front-end technologies like Go, Handlebars, and JavaScript, and building our backend in Python—we built a system that not only recognizes critical incidents but also ensures rapid response.

As a team, we have learned that creating a tech-driven solution for healthcare is not just about coding or models—it’s about empathy, safety, and giving back to the people who need these services the most. Our work on AIDERS is a step toward contributing to a safer healthcare environment and engaging our tech-savvy generation in building solutions that truly matter.

We hope to expand AIDERS to support additional emergency scenarios and integrate it into more healthcare facilities. We will continue refining the model for improved accuracy and broader usability.