Tablet Fractionation in Blisters Using Object Detection Algorithms: Promoting Process Automation at the USP University Hospital.

• Grant number: 25/05640-5
• Support Opportunities: Scholarships in Brazil - Scientific Initiation
• Start date: July 01, 2025
• End date: June 30, 2026
• Field of knowledge: Physical Sciences and Mathematics - Computer Science - Computing Methodologies and Techniques
• Principal Investigator: Leila Cristina Carneiro Bergamasco
• Grantee: Luana Bortko Rodrigues
• Host Institution: Centro Universitário FEI (UNIFEI). Campus de São Bernardo do Campo. São Bernardo do Campo , SP, Brazil

Abstract

The University Hospital of USP (HU-USP), with 134 beds and 1,305 staff, annually handles 76,000 emergency visits and performs 3,100 surgeries. This teaching hospital serves both the USP academic community and the population of the Butantã Health District. The pharmacy department is crucial in drug management, overseeing inventory, expiration dates, and distribution. In 2023, the HU-USP processed 40,000 prescriptions and dispensed 682,000 medications. Although operationally more complex, unit dose dispensing reduces costs and errors while improving patient safety. In this context, Computer Vision emerges as a promising tool for automation, including capsule detection and segmentation. This project aims to enhance detection techniques using the YOLOv11 network, semantic segmentation, and Voronoi Diagrams to promote precise drug cutting. The dataset will include public collections and HU-USP images, aiming to optimize fractionation and improve operational efficiency while contributing to pharmaceutical automation.