TAMI - High-performing XAI systems for application in medicine



TAMI (Transparent Artificial Medical Intelligence) is an international project aiming, specially, at building high-performing Explainable Artificial Intelligence (XAI) systems for application in medicine, whose workings and outputs can be understood and cross-examined by users through human-friendly explanations of the automated results.

Led by the company First Solutions, TAMI is a three-year project (2020-2023) being developed in partnership with Fraunhofer AICOS, INESC TEC, Administração Regional de Saúde do Norte and Carnegie Mellon University. The research developed within this project will focus mainly on four topics: Generate self-explanatory AI-based decisions that minimize bias, act ethically in their context and enhance trust of the end-users; Customizable multimodal and privacy-preserving explanations, that adapt to who will consume them; Quantitative methods to objectively evaluate and compare the suitability of different types of explanations for specific use cases; Foster new strategies for presenting human understandable explanation.

The TAMI project also aims to create a new XAI-based platform for healthcare actors in four main imagiological domains: Cervical Cytology, Cervical Colposcopy, Glaucoma and Chest X-Rays. The TAMI platform will be available for commercial, scientific and academic use, and will provide "consumers" access to self-explanatory AI decisions for target screening and diagnosis procedures, as well as access to filtered and anonymized datasets for research purposes.

Within the TAMI project, FhP-AICOS’ specialists are responsible for the development of AI algorithms that will support clinical decisions on the target medical domains and, above all, will make possible to explain those automated decisions, by crossing textual and image data. Additionally, FhP-AICOS will also lead the development of proof-of-concepts and demonstrators that integrate the researched explainable AI approaches into real clinical workflows, particularly for cervical cancer screening and glaucoma detection.

TAMI will benefit from ongoing efforts from the research team in the collection and annotation of cervical data and chest radiographs to ramp up algorithm research. Also, benefiting from the accumulated expertise of the research team in the topic, a special focus will be put on the development of explanation paradigms able to cope with the high diversity and complexity in medical data. The results of this project will be tested in national and international healthcare institutions, including the TAMI clinical partners, namely: Administração Regional de Saúde do Norte, Centro Hospitalar do Porto, Hospital Prof. Dr. Fernando Fonseca and Instituto Português de Oncologia do Porto.

The TAMI project was selected by the 'Go Portugal - Global Science and Tecnology Partnerships Portugal' initiative in the context of the CMU Portugal program and is financed at 1.79 million euros by the COMPETE 2020 program, Fundação para a Ciência and Technology (FCT) and Carnegie Mellon University.

More information about the TAMI project here


Co-funded by: