A Multi-Modal Approach to Reliable Syringe Exchange

Syringe exchange programs, such as Kentucky Outreach Service Kiosk (KyOSK) at the University of Kentucky’s College of Public Health, aim to reduce the spread of disease among people who use intravenous drugs by facilitating one-for-one syringe exchanges through a vending machine interface. The original system relied solely on weight measurements, which could be manipulated by adding foreign objects. Additionally, the recommended practice of placing used syringes in containers (e.g., plastic bottles) often distorted weight-based readings.

To address these limitations, we developed a computer vision–based system using DINOv2 embeddings to detect and classify foreign objects and accurately count syringes. When a container is present, it is identified (e.g., plastic bottle). Then, the container’s predetermined weight is subtracted. If both syringes and foreign objects are present, the system visually counts the syringes, enhancing accuracy beyond weight-based estimation alone. 

This project is currently ongoing.