Alzheimer’s disease is characterized by progressive cognitive impairment and brain atrophy (the loss of neurons in the brain). It is also one of the most common causes of dementia worldwide, with common symptoms including difficulties with language, problem-solving, and thinking, along with memory loss.
Unfortunately, there is no cure, however, early diagnosis greatly helps patients, as it provides access to help and support and helps them form a treatment plan. Now, a single brain scan can help diagnose Alzheimer’s disease, a study published in Communications Medicine suggests.
Currently, for doctors to diagnose this disease, they need to use an array of tests which include brain scans (to check for shrinkage of the hippocampus and protein deposits in the brain) and cognitive and memory tests. All these tests can take weeks to arrange and process, which can cause delays in any treatment plan.
The new study looked at machine learning technology and used it to peek at the structural features in the brain (including areas that were not previously associated with Alzheimer’s).
Using magnetic resonance imaging (MRI) on a machine commonly found in most hospitals, the researchers applied an algorithm to the brain. This algorithm was originally used in cancer tumor classification.
The brain image was then sliced into 115 regions and allocated 660 different features, including shape, size, and texture. The algorithm was then trained to recognize changes in these features and determine and predict the existence of Alzheimer’s disease.
The team tested their algorithm on scans of more than 400 patients with early and later stage Alzheimer’s, patients with other neurological conditions, and healthy controls.
The MRI-based machine learning system could accurately predict whether someone had Alzheimer’s disease or not in 98 percent of cases. The system was also able to distinguish in 79 percent of patients whether they had the early or late-stage disease.
“Currently no other simple and widely available methods can predict Alzheimer’s disease with this level of accuracy, so our research is an important step forward. Many patients who present with Alzheimer’s at memory clinics do also have other neurological conditions, but even within this group our system could pick out those patients who had Alzheimer’s from those who did not,” said Professor Eric Aboagye, who led the research.
“Waiting for a diagnosis can be a horrible experience for patients and their families. If we could cut down the amount of time they have to wait, make diagnosis a simpler process, and reduce some of the uncertainty, that would help a great deal. Our new approach could also identify early-stage patients for clinical trials of new drug treatments or lifestyle changes, which is currently very hard to do.”
Interestingly, the system also was able to spot changes that were not known to be linked to Alzheimer’s disease. This information could lead to new avenues of research.
“Although neuroradiologists already interpret MRI scans to help diagnose Alzheimer’s, there are likely to be features of the scans that aren’t visible, even to specialists. Using an algorithm able to select texture and subtle structural features in the brain that are affected by Alzheimer’s could really enhance the information we can gain from standard imaging techniques,” study author Dr Paresh Malhotra added.