ArguMammoNet aims to empower clinicians with cutting-edge explainable and trustworthy AI, reduce mortality rates by facilitating earlier and more accurate breast cancer diagnoses, and improve resource efficiency by developing robust AI tools to serve as “second readers” or “tie breakers”, in resource-constrained settings.
Breast cancer continues to be a major public health concern globally, with early and accurate diagnosis playing a vital role in improving survival rates. While recent advances in Computer-Aided Diagnosis (CAD) have shown promise, current systems do not compare sequential mammograms and lack transparency in their decision-making processes. These algorithms utilize medical images as input and return a decision which, while accurate, does not provide any rationale.
ArguMammoNet is set to revolutionize breast cancer diagnosis by developing an AI-driven framework that utilizes sequential mammograms along with explainable Argumentation Convolutional Neural Networks (Ar-CNNs). This framework aims to provide accurate detection, classification, and prediction of breast cancer.
The ArguMammoNet project is expected to deliver several impactful outcomes that advance the state of breast cancer diagnosis, starting with the development of an algorithm for subtraction of temporally sequential mammograms. Building on this, the project will create a novel AI framework for the detection, classification, and prediction of breast cancer, combining the sequential mammograms and Convolutional Neural Networks. A key innovation lies in the integration of argumentation reasoning frameworks, which will enhance the interpretability and clinical transparency of the deep learning models. Ultimately, these technologies will be embedded into user-friendly software tools, allowing clinicians to seamlessly incorporate ArguMammoNet’s capabilities into diagnostic workflows.
The project is coordinated by KIOS Research and Innovation Center of Excellence at the University of Cyprus and the key partners are State Health Services Organization (SHSO), Ygia Polyclinic Limassol, Bank of Cyprus Oncology Centre, and 3aehealth LTD.
The project POST-DOC/0524/0095 is implemented under the programme of social cohesion “THALIA 2021-2027” co-funded by the European Union, through Research and Innovation Foundation.
