New Algorithm Predicts Alzheimer’s with 99% accuracy

Researchers from Kaunas universities in Lithuania developed a deep learning-based method that can predict the possible onset of Alzheimer’s disease from brain images with an accuracy of over 99 per cent. The method was developed while analysing functional MRI images obtained from 138 subjects and performed better in terms of accuracy, sensitivity and specificity than previously developed methods.

According to World Health Organisation, Alzheimer’s disease is the most frequent cause of dementia, contributing to up to 70 per cent of dementia cases. Worldwide, approximately 24 million people are affected, and this number is expected to double every 20 years. Owing to societal ageing, the disease will become a costly public health burden in the years to come.

Medical professionals all over the world attempt to raise awareness of an early Alzheimer’s diagnosis, which provides the affected with a better chance of benefiting from treatment. This was one of the most important issues for choosing a topic for Modupe Odusami, a PhD student from Nigeria”, says Rytis Maskeliūnas, a researcher at the Department of Multimedia Engineering, Faculty of Informatics, Kaunas University of Technology (KTU), Odusami’s PhD supervisor. One of the possible Alzheimer’s first signs is mild cognitive impairment (MCI), which is the stage between the expected cognitive decline of normal ageing and dementia. Based on the previous research, functional magnetic resonance imaging (fMRI) can be used to identify the regions in the brain which can be associated with the onset of Alzheimer’s disease, according to Maskeliūnas. The earliest stages of MCI often have almost no clear symptoms, but in quite a few cases can be detected by neuroimaging.

However, although theoretically possible, manual analysing of fMRI images attempting to identify the changes associated with Alzheimer’s not only requires specific knowledge but is also time-consuming – application of Deep learning and other AI methods can speed this up by a significant time margin. Finding MCI features does not necessarily mean the presence of illness, as it can also be a symptom of other related diseases, but it is more of an indicator and possible helper to steer toward an evaluation by a medical professional.

Modern signal processing allows delegating the image processing to the machine, which can complete it faster and accurately enough. Of course, we don’t dare to suggest that a medical professional should ever rely on any algorithm one-hundred-per cent. Think of a machine as a robot capable of doing the most tedious task of sorting the data and searching for features. In this scenario, after the computer algorithm selects potentially affected cases, the specialist can look into them more closely, and at the end, everybody benefits as the diagnosis and the treatment reaches the patient much faster”, says Maskeliūnas, who supervised the team working on the model.


Breakthrough In The Fight Against Alzheimer’s

Eisai Co.,  a company located in Tokyo, and Biogen Inc. in Cambridge, United States, announced positive topline results from the Phase II study with BAN2401, an anti-amyloid beta protofibril antibody, in 856 patients with early Alzheimer’s disease. The study achieved statistical significance on key predefined endpoints evaluating efficacy at 18 months on slowing progression in Alzheimer’s Disease Composite Score (ADCOMS) and on reduction of amyloid accumulated in the brain as measured using amyloid-PET (positron emission tomography).

Study 201  is a placebo-controlled, double-blind, parallel-group, randomized study in 856 patients with mild cognitive impairment (MCI) due to Alzheimer’s disease (AD) or mild Alzheimer’s dementia (collectively known as early Alzheimer’s disease) with confirmed amyloid pathology in the brain. Efficacy was evaluated at 18 months by predefined conventional statistics on ADCOMS, which combines items from the Alzheimer’s Disease Assessment Scalecognitive subscale (ADAS-Cog), the Clinical Dementia Rating Sum of Boxes (CDR-SB) scale and the Mini-Mental State Examination (MMSE) to enable sensitive detection of changes in early AD symptoms. Patients were randomized to five dose regimens, 2.5 mg/kg biweekly, 5 mg/kg monthly, 5 mg/kg biweekly, 10 mg/kg monthly and 10 mg/kg biweekly, or placebo.

Topline results of the final analysis of the study demonstrated a statistically significant slowing of disease progression on the key clinical endpoint (ADCOMS) after 18 months of treatment in patients receiving the highest treatment dose (10 mg/kg biweekly) as compared to placebo. Results of amyloid PET analyses at 18 months, including reduction in amyloid PET standardized uptake value ratio (SUVR) and amyloid PET image visual read of subjects converting from positive to negative for amyloid in the brain, were also statistically significant at this dose. Dose-dependent changes from baseline were observed across the PET results and the clinical endpoints. Further, the highest treatment dose of BAN2401 began to show statistically significant clinical benefit as measured by ADCOMS as early as 6 months including at 12 months.