Human-Centered Design of a Semantic Annotation Tool

Miguel Guerreiro Girão Bastos

Key information

Authors:

Miguel Guerreiro Girão Bastos (Miguel Guerreiro Girão Bastos)

Supervisors:

Jacinto Carlos Marques Peixoto do Nascimento (Jacinto Carlos Marques Peixoto do Nascimento); Francisco Maria Galamba Ferrari Calisto (Francisco Maria Calisto)

Published in

November 7, 2024

Abstract

Breast cancer is the most frequently diagnosed cancer among women and the leading cause of cancer-related mortality. Artificial intelligence models can play a crucial role in enhancing diagnosis and treatment. These models rely on detailed information, such as lesion characteristics, provided through semantic annotation of medical images. This work developed a web-based system for semantic annotation integrated into a breast cancer imaging framework. Using a human-centered design approach, iterative development was guided by clinician feedback to ensure seamless integration into clinical workflows. User testing showed the system to be intuitive and adaptable across varying levels of clinical experience, with positive feedback leading to further refinements. The system’s output will be used to train machine learning models to assist clinicians in decision-making. Successfully integrated into real-world clinical practice, this tool has the potential to enhance breast cancer diagnosis and treatment.