North American scientists, through the use of images obtained during medical check-ups, managed to create an AI model that can diagnose this neurodegenerative pathology with more than 90% accuracy. The details
The study will serve to advance diseases such as Alzheimer’s and obtain more accurate diagnoses (Getty)
The use of Artificial Intelligence (AI) to detect diseases from brain MRIs promises to automate, standardize and become a diagnostic process at scale. It is that these clinical studies are routinely collected and accumulated in large databases that can be used to train AI algorithms .
In this sense, deep learning , for its part, has been shown to be successful in detecting multiple diseases in high-quality brain MRI data, which were collected in a controlled research environment. Hand in hand with these advances, researchers have made progress in detecting signs of Alzheimer ‘s disease using high-quality brain imaging tests collected as part of research studies.
With all these aspects “on the table”, a team of specialists from the Massachusetts General Hospital (MGH) recently developed an accurate method for the detection of Alzheimer’s that is based on clinical brain samples collected in the form of images in routine examinations. This development could lead to more accurate diagnoses .
For the study, which was published in PLOS ONE , Matthew Leming, a researcher at the MGH Center for Systems Biology and a researcher at the Massachusetts Alzheimer’s Disease Research Center, along with his colleagues Sudeshna Das and Hyungsoon Im, used a tool which is called deep learning , a type of machine learning in which artificial intelligence uses large amounts of data and complex algorithms to train models and draw conclusions.
Artificial intelligence will allow specialists to recognize early characteristics of diseases such as Alzheimer’s (Getty Images)
In this case, the scientists developed a model for the detection of Alzheimer’s disease based on brain magnetic resonance imaging (MRI) data collected from patients with and without Alzheimer’s disease who were seen at MGH prior to 2019. Next, the The group tested the model on five data sets: on MGH post-2019, on Brigham and Women’s Hospital before and after 2019, and on external health systems before and after 2019.
The objective involved analyzing whether its development was capable of accurately detecting Alzheimer’s disease based on real data, global clinical data, regardless of the hospital and the time in which these data were collected.
Overall, the research involved 11,103 images of 2,348 patients at risk for Alzheimer’s disease and 26,892 images of 8,456 patients without the disease. Across all five data sets, the model detected Alzheimer’s disease risk with 90.2% accuracy .
Massachusetts General Hospital recently developed an accurate method for the detection of Alzheimer’s that is based on clinical brain samples collected in the form of images in routine examinations (Getty)
Modeling the look on Alzheimer’s
Among the main novelties of the work we can mention its ability to detect Alzheimer’s disease independently of other variables, such as age. “Alzheimer’s disease typically occurs in older adults , so deep learning models often have difficulty detecting the rarer early-onset cases. We address this point by making the deep learning model blind to brain features that are too closely associated with the indicated age of the patient,” Leming said.
The expert pointed out that another common challenge in disease detection, especially in real-world settings, is dealing with data that is very different among the information pool that is used to feed the AI system. For example, a deep learning model trained on MRIs from a scanner made by one company may not recognize MRIs collected on a scanner made by another.
Alzheimer’s disease is the most common form of dementia among older people; the use of AI can change your quality of life (photo: Saber Vivir)
The model used an uncertainty metric to determine if the patient’s data was too different from what the same model had been trained on to make a successful prediction.
“This is one of the only studies that used routinely collected brain MRIs to try to detect dementia. While a large number of deep learning studies have been conducted for the detection of Alzheimer’s disease from brain MRIs, this study took substantial steps toward doing this in real-world clinical settings rather than perfect laboratory settings. Our results, with the ability to generalize across places, times, and populations, make a strong case for the clinical use of this diagnostic technology,” Leming concluded.
How to treat dementia-related pain and what caregivers should be aware of
Why the music of youth brings back memories, even for people with dementia
Alzheimer’s: Artificial intelligence can detect early signs of the disease