A new artificial intelligence algorithm would make it possible to diagnose Parkinson’s disease early thanks to the patient’s breathing.
- Parkinson’s disease develops in the brain gradually and well before the appearance of the first symptoms which can be motor, such as akinesia (slowness of movement), rigidity or tremors, but also cognitive, or trouble sleeping, balance, or smell.
- To date, more than 200,000 people are affected by Parkinson’s disease in France and 25,000 new cases occur each year, i.e. one new case every two and a half hours. It is the second most common neurodegenerative disease, behind Alzheimer’s disease.
For now, there is no treatment that can reverse or stop the progression of Parkinson’s disease. And this remains particularly difficult to diagnose because it is often necessary to wait for the appearance of motor symptoms (tremors, muscle stiffness, chronic fatigue, etc.) to notice it.
But that could soon change: researchers at the Massachusetts Institute of Technology (MIT), in the United States, have developed a new artificial intelligence tool allowing early detection of this disease which, in France, affects more than 200,000 people, i.e. one adult in 250, according to Inserm figures. Their work has been published in the journal Nature Medicine.
Analyze nocturnal breathing
Tested on more than 7,000 patients, including 757 with Parkinson’s, the tool in question is an artificial neural network – in other words a series of connected algorithms that imitate the functioning of a human brain – capable of diagnosing the pathology (but also to assess its level of severity and its progression) just by analyzing the breathing of the person during his sleep.
And that’s the whole point: “Some scientific studies have shown that respiratory symptoms appear years before motor symptoms”explains Dina Katabi, one of the authors of the study, in a communicated. The researchers hope that early detection of the disease will thus make it possible to develop other treatments and therapies, in particular thanks to clinical trials that are significantly faster and require fewer volunteers.
Diagnose yourself effortlessly
Because the other advantage is that the diagnosis could, in the future, be carried out directly by the patient himself, at home, without requiring nursing staff or invasive medical tests. So far, in fact, the disease can only be detected using cerebrospinal fluid analysis and brain imaging. Heavy and expensive exams that can discourage more than one, especially when you suffer from reduced mobility or cognitive impairment.
While, as MIT reminds us, there was no “no therapeutic breakthrough in this century” concerning this disease which, however, “is experiencing the strongest growth in the world”, this new discovery therefore appears to be a ray of hope for future patients with Parkinson’s.
