Artificial intelligence

The mutimodal CPR tutor Di Mitri developed uses several sensors that register the execution of CPR. The system registers the trainee's body position using Microsoft Kinect and his arm movements and muscle activity (electromyogram) with the Myo armband. The compression rates data are generated by the manikin. Alle these sensors track the CPR execution and generate data. Of course all these data have to be combined and analysed in order to generate feedback. Di Mitri: 'Since my background is in Artificial Intelligence, I am very interesting in using multimodal data not simply for analytics but also to generate adaptive feedback through machine learning techniques.'

PhD research

This study is the main experiment of a Daniele’s PhD project focusing on multimodal data support for investigating practice-based learning scenarios, such as psychomotor skills training in the classroom or at the workplace. The research forms part of the European project SafePat, a project that develops excellence in patient safety in cross-border regions through standardised procedures, policies and innovative tools.

Read more

This summer Di Mitri presented a paper on his work during the 19th international conferenc Artificial intelligence in Education. His paper Multimodal Tutor for CPR is online available.

In September Di Mitri received the Martin Wolper’s award for most promising youg researcher, an international award instituted by the Research Institute for Innovation & Technology in Education. In a video he explains his work.