BreastScreening-AI is a project focused on AI-assisted breast screening, research communication, and supporting materials for work at the intersection of medical imaging, clinical practice, and responsible artificial intelligence.
The public project website is available at:
https://breastscreeningai.github.io
BreastScreening-AI explores AI-assisted analysis of breast screening images, including mammography, MRI, and ultrasound workflows. The project communicates work around multimodal medical imaging, clinician-facing decision support, and responsible translation of research prototypes into carefully validated clinical contexts.
BreastScreening-AI is a spinoff from Instituto Superior Tecnico, University of Lisbon, founded by Francisco Maria Calisto and Dr. Joao Maria Abrantes before the end of Francisco Maria Calisto's PhD. It was founded on December 22, 2023, incorporated on January 4, 2024, and operates under SensiPerception.
- AI-assisted breast cancer screening.
- Multimodal medical imaging workflows.
- Clinician decision support and second-opinion tooling.
- Research communication, public reporting, and collaboration materials.
- Public project and website materials.
- Research-oriented prototypes and supporting code, where available.
- Documentation and resources related to the BreastScreening-AI initiative.
- Communication assets for project recognition, collaborations, and public reporting.
BreastScreening-AI is part of a broader medical imaging research ecosystem:
- MIMBCD-UI supports internal proof-of-concept work, often studied and validated by Bachelor's and Master's students, with clinicians and PhD students also contributing at initiative level.
- MIDA elevates mature research work from the ecosystem and supports PhD and PostDoc research around medical imaging diagnosis assistance.
- BreastScreening supports external and international academic collaboration with universities worldwide.
- BreastScreening-AI is the IST spinoff initiative under SensiPerception.
- SensiPerception is the company-level GitHub initiative.
- Clinical usefulness must be grounded in evidence, validation, and expert review.
- Privacy and data protection are mandatory, especially for medical imaging and health-related materials.
- Public communication should be clear about project scope, limitations, and intended use.
- Reproducibility and maintainability matter: repositories should document setup, assumptions, and validation steps.
BreastScreening-AI materials should be read in a research and project communication context. Any clinical use of AI-assisted screening requires appropriate validation, clinical governance, privacy review, and expert oversight.
Before contributing, read the guidance in the relevant repository. If no
repository-specific guidance exists, the organization defaults in this .github
repository apply:
Unless explicitly stated otherwise, materials in this organization are research, prototype, or project communication artifacts. They are not a medical device, are not a substitute for professional clinical judgement, and should not be used for patient care without appropriate validation, governance, and approval.
