Creating an Adaptive and Interactive AI at the University of Bologna
The University of Bologna (Italy) was recently awarded a grant by the European Union to conduct a number of research projects as part of a consortium led by Professor Patrizia Fattori. As part of the equipment requirements to carry out this work, Brainbox had the honour of providing two transcranial magnetic stimulators (a DuoMAG XT and DuoMAG MP-Dual) to the university.
Following the installation of these systems, we had the chance to speak to Dr Rossella Breveglieri of the University of Bologna who will be on the frontline of these research projects to find out more about the work being carried out.
“Today,” Rossella explains, “millions of people worldwide suffer from deteriorated motor abilities due to stroke, brain tumour, surgery, or accident. This represents a serious society challenge with missing adequate technological response. Patients need assistive devices that are trustworthy, multifunctional, adaptive and interactive, i.e., intelligent, unlike current neuroprosthetics that replace single motor impairments.”
The project — ‘Multifunctional, Adaptive and Interactive AI system for Acting in Multiple Contexts, or ‘MAIA’ for short — proposes a paradigmatic shift in the field, where human-centric artificial intelligence (AI) will control prosthetic and assistive devices.
At the University of Bologna, the research team will seek to investigate and resolve critical steps towards the rapid development of such human-central control — including a radically novel intention decoder that acquires information from multiple sources and a novel concept for trustworthy human-AI interactions.
This newly-developed MAIA AI technology will then be able to decode human intentions and communicate the decoded targets both to assistive devices and to users, to ensure full compliance and develop a relationship of trust through natural interactions and mutual learning.
“The technological outcome”, Rossella continues, “will be a multifunctional human-centric AI controller at TRL4 with embedded trustworthy characteristics, suitable to be integrated in robotic arms, wheelchair and exoskeletons.”
“To reach this, MAIA will investigate the principles underlying natural, fast and lean communication and new forms of combinations of neural and behavioural data beyond current data processing. MAIA’s approach will be guided by real needs of end users (patients and caregivers) through their direct involvement in the research program, and by all current knowledge from neuro-, cognitive, and social science research. The application domains of MAIA’s new paradigm span from healthcare to industry, and space exploration.”
Rossella’s work with MAIA builds upon her impressive body of existing research exploring the visuomotor processes in non-human primates while reaching and grasping, and she explains that, “Many fMRI data and TMS studies support a strict homology between humans’ and monkeys’ posterior parietal cortex. I also expect to find the human counterparts of the results obtained in our lab in the monkey brain — so a direct translation of monkey results to the human brain will be possible.”
Despite the ongoing disruption that we have seen to research over the past two years, Rossella is relieved to tell us that her research was only minimally impacted by the very early days of the pandemic in spring 2020, and that with the easing of lockdowns in Italy she is now hard at work again in the lab and undertaking testing for the MAIA project.
We are very excited to learn more about MAIA as the project progresses, and wish Rossella and the other members of her team at the University of Bologna the very best.
Breveglieri R, Borgomaneri S, Filippini M, De Vitis M, Tessari A, Fattori P.Brain Sci. 2021 Oct 15;11(10):1357. doi: 10.3390/brainsci11101357.PMID: 34679421