Multiple sclerosis and leukodystrophies, the diagnosis is made with artificial intelligence
[ad_1]
MONTREAL – “I had to stay in Canada for only one year, with a post-doctoral fellowship. Then, one thing led to another, and I stayed here”. Roberta La Piana is a professor in the Department of Neurology and Neurosurgery at McGill University in Montréal. But her training is Italian: first as a doctor in child neuropsychiatry at the University of Pavia, then at the Niguarda Hospital in Milan. “I was doing the clinic, but I missed research a lot. When I got here, I began to regret the relationship with the patient,” she recalls. Now however, also with the help of Artificial Intelligence, you have found the right balance, developing a line of clinical research oriented towards rare patients.
Doctor La Piana, what do you do here at The Neuro?
“As a pediatric neurologist I have always been interested in leukodystrophies, genetic diseases that affect the white matter, myelin, the substance that covers the axons and allows the conduction of the electrical stimulus along the central nervous system. In children, these diseases are relatively easy to diagnose because they are the most common type in that age group. However, these are very rare diseases, even if one in particular – adrenoleukodystrophy – became better known thanks to the film Lorenzo’s oil. The data show that in the countries of the European Union there are 1.1 cases out of 100,000 live births for each single genetic form, and according to the European Leukodystrophy Association, about 20-40 children with leukodystrophy are born every week”.
Rare diseases, Elena’s story and Lorenzo’s oil disease
by IRMA D’ARIA
When do they appear and what is the life expectancy of the patients?
“These are serious conditions that often lead to death before the age of two. However, there are also late-onset forms, with milder symptoms, although these are still patients with an average age of around 40 and with an often very short life expectancy. For one of the forms under study, life expectancy is currently 7 years from diagnosis and we know about 300 people affected worldwide. Interestingly, many of these adult-onset forms have characteristics very similar to those of multiple sclerosis (MS), with motor and cognitive symptoms. Thus many adult patients with these genetic diseases receive an incorrect diagnosis, and consequently also an unsuitable therapy. Since the Neurological Institute of Montréal – The Neuro – specializes in MS research, here I have the opportunity to study patients who have been diagnosed with an atypical form of MS and understand if they are not affected by leukodystrophy instead”.
Clinical Research: From Patient Digital Twins to Test Tube Organs
by Sara Carmignani
What type of therapy is provided for these rare forms?
“The treatments are still few and risky: one of the most common is bone marrow transplant. But the reason why this area is very interesting and full of potential is precisely this: since these late forms of leukodystrophy are so similar to multiple sclerosis, my research group and I are using these rare diseases as a model to better understand the SM. And vice versa, we can use the advanced knowledge we have about MS to apply it to these rare diseases. For example, considering the possibility of using some drugs already available for other conditions for this patient population. This is a topic of great interest to pharmaceutical companies working on MS: when they understand that some patients may have been misdiagnosed as having MS, they want to be sure it is leukodystrophy, to avoid treating patients with the wrong medicines. In this sense, knowing how to discriminate between the two diseases is essential. A better diagnosis is the first step towards a therapy. Fortunately, the speed with which we are approaching a treatment is much faster than we could have predicted a few years ago.”
540 thousand euros for clinical research projects and multiple sclerosis
by Tiziana Moriconi
And how do you know if a patient has the rare form of leukodystrophy or multiple sclerosis?
“It is a long and complex process. The diagnosis takes place thanks to the combination of three parameters, therefore with a multidisciplinary approach. The first suspicion is verified with magnetic resonance, thanks also to Artificial Intelligence algorithms capable of recognizing some patterns. In this university hospital, revolutionary research is being carried out on imaging in multiple sclerosis, thanks to a seven Tesla magnetic resonance device, a very advanced machine that allows very high resolution, much higher than that of the normal machines used in diagnostic imaging. The pattern learning system therefore helps to direct the diagnosis to the next step, i.e. perform genetic analyses, which however are able to reach a definitive diagnosis only in 50% of cases. For the remaining 50% we have to rely on parallel sequencing techniques, using Next Generation Sequencing, which gives us the possibility of sequencing the entire protein-coding region or the entire genome of a patient. For this we also use “reverse phenotyping”, which consists in starting from a certain diagnosis of leukodystrophy obtained with genetics to understand if the magnetic resonance gives us useful information to characterize it”.
Against multiple sclerosis put the immune system on a diet
by Anna Lisa Bonfranceschi
How else do AI systems help you?
“In my laboratory, also thanks to Artificial Intelligence, we are, for example, analyzing data to identify genes that cause diseases or variations of previously known disorders. Unfortunately we work with very small numbers, given that we are talking about rare diseases: our patient cohort includes more than 70 families, and we receive about 20 new requests for advice every year. In this sense, the first problem is the limited availability of data, and therefore we must be very careful to interpret them correctly. Machine learning helps us, but the ultimate validation must come from a human specialist”.
Rare diseases, avatars speed up research
by Anna Lisa Bonfranceschi
Could we imagine a form of prenatal genetic screening for families at risk?
“Of course, this is a fundamental point. But if we don’t improve diagnosis, if we continue to mistake rare leukodystrophies for multiple sclerosis, not only will we have patients treated inappropriately, but we will not be able to do genetic counseling of families or do prenatal diagnosis. And this will weigh not only on individual patients, but also on the national health service, which here in Canada, as in Italy, is public and universal”.
[ad_2]
Source link
