Artificial Intelligence Can Predict Parkinson's Disease Evolution, Italian Study Says

Predicting the progression of Parkinson's disease by 'reading' a patient's brain activity , allowing for early and personalized intervention. This scenario has become a reality thanks to artificial intelligence at the Parkinson's Center of the Gaetano Pini-CTO Hospital in Milan . In a study published in the journal 'NPJ Parkinson's Disease' (Nature group), the company reports, a new algorithm was presented that can predict the progression of brain signals in patients undergoing deep brain stimulation (DBS) who are continuously monitored at home, thus gradually adjusting treatment based on the AI's indications.

"Reading the Future" Thanks to AI

"We can now 'read the future' of Parkinson's disease, estimating clinical progression a week in advance and promptly intervening in neuromodulation therapy to make it more effective," says Ioannis U. Isaias, director of the Parkinson's and Parkinsonism Center at Asst Pini-Cto. "This result is very important for fully leveraging new adaptive deep brain stimulation strategies, of which our center is among the first in the world to experiment," emphasizes the specialist, who led the study together with Alberto Mazzoni of the Biorobotics Institute at the Sant'Anna School of Advanced Studies in Pisa. "I'm very pleased with this result," says Mazzoni, "as it translates our lines of research on developing new engineering methods for analyzing neural signals into immediate clinical use. I'm also particularly pleased because the lead authors of this work, Dr. Salvatore Falciglia and Dr. Laura Caffi, are young researchers pursuing a PhD in Biorobotics at the Scuola Sant'Anna, but within an international context. This research project is, in fact, co-supervised not only by Professor Isaias, but also by Dr. Chiara Palmisano of the University of Würzburg, Germany."

A key contribution, the Pini-Cto Association emphasizes, was also made by the Pezzoli Foundation for Parkinson's Disease , the promoter and partner of research at the Parkinson's Center at the Royal Botanic Gardens. "Our commitment to patients translates into concrete projects to improve care, from preventive therapies to the most technologically advanced therapies," says the Foundation's president, Gianni Pezzoli. The patients involved in the study received a new experimental stimulator during surgery performed by Marco Locatelli's team at the Policlinico Hospital in Milan, one of the Neurosurgery Units with which the Pini-Cto Parkinson's Center has developed a network of collaborations throughout Lombardy. The network also involves the Santi Paolo e Carlo Association in Milan and the San Gerardo Institute of Clinical Research in Monza, to meet the high demand for procedures. The Pini-Cto Parkinson's Center sees over 7,000 patients each year, with more than 1,500 new admissions.

"Deep brain stimulation," explains Salvatore Bonvegna, head of the Neuromodulation Outpatient Clinic at the Parkinson's Center at Asst Pini-Cto, "is a neurosurgical technique that involves implanting electrodes in well-defined regions of the brain, such as the subthalamic nucleus or the internal globus pallidus. Impeccably performed neurosurgical intervention, with precise electrode placement, is crucial to achieving a successful treatment outcome." The electrodes are then connected to a subcutaneous pacemaker (stimulator), usually implanted in the right subclavian region. This system regulates the electrical impulses sent to the brain. Although complex, the procedure is one of the leading treatment options for advanced Parkinson's disease, providing many patients with a significant improvement in their quality of life, experts note.

Thanks to the ability to adapt stimulation in real time based on the patient's brain activity, the devices used represent a breakthrough in disease treatment, making therapy even more targeted and effective. "Compared to conventional stimulation modalities, which maintain fixed stimulation parameters," explains Isaias, "in adaptive DBS, the current is modulated in real time based on brain signals, used as a marker of disease symptoms. This allows stimulation to be automatically adjusted based on the patient's actual needs. In our experience, adaptive stimulation, when tolerated, is preferred by over 80% of patients due to the greater efficacy and personalized nature of the therapy." The new study represents a further development, opening the door to "AI algorithms for automatic programming and adjustment that anticipate the patient's needs," the specialists predict. On the horizon, "'smart neuroprostheses' are hoped to lead to complete functional recovery for patients in the future."

"It is hoped that these technological innovations and improved treatments will spark greater interest in this new therapeutic strategy for Parkinson's disease, which is still underused in Italy," comments Paola Lattuada, general director of the Gaetano Pini-Cto Association of Milan. "Currently, of the approximately 2,000 Parkinson's patients estimated to benefit from deep brain stimulation, only about 300 undergo the procedure each year. A team of multidisciplinary specialists and a network of collaborations are crucial to the success of the treatment."