Artificial intelligence, 18 year old creates algorithm for early diagnosis of Parkinson’s

An artificial intelligence algorithm that favors the early diagnosis of Parkinson’s: it was created by an 18-year-old boy, Tommaso Caligari, and several doctors are already experimenting with it – with interesting results – in collaboration with the association of Parkinson’s patients. Tommaso’s is probably the most fascinating story among those presented today at the Novara Castle as part of “Innova”, an event promoted by the Cna National Crafts Confederation to offer businesses, above all, a window on Blockchain, Metaverse and artificial intelligence. Because, as underlined by Marco Vicentini, national vice-president of Cna “we must welcome change, we have no alternatives” And Caligari perfectly represents the “natural” openness to change of the young generations.

The study of grandfather’s illness

It all started from a personal and family experience: “My grandfather – the boy tells AGI – was hospitalized last year for Parkinson’s disease. In this way I began to get to know this universe that is often unknown to most people. “. Tommaso, who studies robotics at ITS “Omar” in Novara, began to wonder about the possibility of using technology to better treat Parkinson’s patients.

How the algorithm works

“In the initial stages of the disease – explains the boy – disturbances such as tremor and difficulty walking are not yet visible but begin to appear imperceptibly. I have developed this algorithm which is able to recognize the human figure and trace its movements and in particular alterations in shoulder and elbow swing which are one of the first symptoms in the early stages of Parkinson’s disease”. “These are oscillations – continues the young inventor – so small that the human eye is not able to recognize them. This system uses two cameras which are positioned one in front of the other about five meters away and analyze the path of the patient under examination. The program is able to detect the movements of the walking patient and compare them with a machine learning algorithm “trained” on a group of healthy people and on a group of patients with full-blown Parkinson’s. The comparison is able to put into highlights early indicators of the disease.” Tommaso Caligari’s invention is being tested with surprising results. The aim is to apply it in hospitals where Parkinson’s is diagnosed as the first screening for the disease.