From imaging tests to clinical records, and from physical activity sensors to hospital systems, the information that underpins artificial intelligence (AI) in healthcare originates from material, clinical and human contexts. The way this data is collected, protected, shared and transformed into decision support was the focus of the session “Artificial Intelligence in Healthcare: Diagnosis, Treatment and Personalised Care”, held on 23 January 2026 at the Técnico Innovation Center powered by Fidelidade.
The event marked the return, in 2026, of the series of talks titled “Artificial intelligence in interaction with the physical world”, organised by Institute for Systems and Robotics (ISR-Lisboa/LARSyS) in collaboration with Instituto Superior Técnico, Universidade de Lisboa. The session was moderated by Patrícia Figueiredo, a professor at Técnico, and brought together different perspectives of AI in clinical practice, health data management and the associated ethical and legal challenges.
Catarina Barata, a professor at Técnico and a researcher at ISR-Lisboa/LARSyS, highlighted the clinical origin of the information used by AI systems, emphasising that “the data comes from a very physical process”. In the context of medical imaging and clinical data analysis, she highlighted the relationship between the physical and digital worlds and the “need to preserve and organise patient information, ensuring equitable access and the protection of their data”. Additionally, Catarina Barata mentioned the existence of public repositories and the importance of creating and making data available for research.
The structural challenges facing healthcare systems were addressed by Francisca Leite, a Técnico alumna and a representative of Hospital da Luz Learning Health, who described an “ageing population with an increasing burden of disease in a context of scarce resources”. In this scenario, she explained that personalised medicine – tailored to the characteristics of each patient – depends on greater digitisation and automation of processes that are still affected by delays. According to Francisca Leite, the integration of AI requires “a re-engineering of healthcare”, with a review of processes, a focus on prevention and a strengthening of confidence in the models. She also highlighted the role of data spaces, higher-quality data sets for researching new algorithms, in a sector that “faces both significant challenges and opportunities for transformation”.
The need for organisational and technological adaptation was reinforced by Marco Manso, from PARTICLE-Summary, a company providing advanced data solutions and cyber-civic systems, who framed the current moment as “two major transformations: digital and digital reinforced by the emergence of artificial intelligence”. In light of new AI models, he stressed the “importance of adaptation by professionals and organisations, advocating for system automation to be accompanied by human validation”. Continuous monitoring of habits and physical activity was cited as an example of data support platforms, and he called for the involvement of various stakeholders in innovation processes.
Henrique Martins, a professor at ISCTE –University Institute of Lisbon, also highlighted the importance of creating systems that are “well-suited to the reality and adaptability of Portugal”, emphasizing the connection between technology, regulation and the national context. He also defended the urgency of specific legislation on health data and its availability as a basis for the sustained development of AI solutions in this field.
The practical and legal implications of using AI in healthcare were discussed during the Q&A session, including liability for misdiagnosis. Catarina Barata reiterated the importance of integrating ethical processes into research and clinical application, mentioning the “essential contribution of these technologies to the discovery of new biomarkers and to supporting hospital occupancy management”.
Throughout the session, artificial intelligence in healthcare was presented as a field in which clinical decisions, data infrastructures, service organisation and legal frameworks evolve interdependently, requiring coordination between research, clinical practice, technological development and public policy. The series of conferences “Artificial intelligence in interaction with the physical world” ends on 13 February 2026, with a session dedicated to the theme “Ethics and Social Responsibility of AI”.
