Alzheimer’s disease is the most common cause of dementia all over the world. Although there is no treatment, early discovery is essential for developing effective therapies before the disease irreversibly progresses.
Alzheimer’s disease affects more than 50 million people worldwide. Image Credit: Jack Finnigan / unsplash.com.
Mild cognitive impairment is a stage that precedes Alzheimer’s disease, but not everyone who has it develops the condition. A study headed by researchers at the Universitat Oberta de Catalunya (UOC) and reported in the IEEE Journal of Biomedical and Health Informatics has pinpointed the difference between individuals whose degeneration is steady and those who will develop the disease.
The new technology, which compares magnetic resonance images using particular artificial intelligence algorithms, is more successful than existing methods already in use.
Fine-Tuning the Diagnosis
Alzheimer’s disease can affect more than 50 million people globally, and the population’s aging implies that many more people will develop the condition in the future decades. Although it normally develops without symptoms over a long period of time, it is frequently preceded by moderate cognitive impairment, which is significantly milder than the damage seen in Alzheimer’s patients but more severe than would be anticipated for someone their age.
These patients may progress and worsen or remain in the same condition as time passes. That is why it is important to distinguish between progressive and stable cognitive impairment in order to prevent the rapid progression of the disease.
Mona Ashtari-Majlan, Study Lead Author and Researcher, Universitat Oberta de Catalunya
Ashtari-Mailan is also a researcher in AI for Human Wellbeing group (AIWELL) which is affiliated with eHealth Center, Faculty of Computer Science, Multimedia and Telecommunications, and a Ph.D. student in Network and Information Technologies, supervised by David Masip and Mohammad Mahdi Dehshibi.
Correctly identifying these cases might help to enhance the quality of clinical studies used to evaluate therapies, which are increasingly aimed at the disease’s early stages. To accomplish this, the researchers employed a multi-stream convolutional neural network, which is a deep learning and artificial intelligence technology that is particularly beneficial for image identification and classification.
“We first compared MRIs from patients with Alzheimer’s disease and healthy people to find distinct landmarks,” stated Ashtari-Majlan. After training the system, scientists used resonance pictures from individuals who had previously been identified with stable or progressive cognitive impairment with few variations to fine-tune the proposed architecture. Nearly 700 images from publicly available databases were utilized in total.
Overcomes the complexity of learning caused by the subtle structural changes that occur between the two forms of mild cognitive impairment, which are much smaller than those between a normal brain and a brain affected by the disease. Furthermore, the proposed method could address the small sample size problem, where the number of MRIs for mild cognitive impairment cases is lower than for Alzheimer’s.
Mona Ashtari-Majlan, Study Lead Author and Researcher, Universitat Oberta de Catalunya
The novel approach can differentiate and classify the two types of mild cognitive impairment with an accuracy rate of about 85%. Ashtari-Majlan includes traditional and other deep learning-based techniques in research analysis, even when they are paired with indicators like age and cognitive testing.
The evaluation criteria show that our proposed method outperforms existing ones. We can share our implementation with anyone wishing to reproduce the results and compare their methods with ours. We believe that this method can help professionals to expand the research.
Mona Ashtari-Majlan, Study Lead Author and Researcher, Universitat Oberta de Catalunya
Journal Reference:
Ashtari-Majlan, M., et al. (2022) A multi-stream convolutional neural network for classification of progressive MCI in Alzheimer’s disease using structural MRI images. IEEE Journal of Biomedical and Health Informatics. doi.org/10.1109/JBHI.2022.3155705.