MEDICINE AND HEALTH

Brain network research lays the foundation for accurate diagnosis and treatment of Alzheimer’s disease


Alzheimer’s disease (AD) is a complex neurodegenerative disease, but also the most common cause of dementia in old age, with the advent of an aging society, the incidence of AD has increased year by year, becoming a health problem that cannot be ignored.

Recently, the team of Liu Yong, a professor at the School of Artificial Intelligence of Beijing University of Posts and Telecommunications, and Zhou Bo, deputy chief physician of the Second Medical Center of the General Hospital of the People’s Liberation Army of Chinese, revealed four AD subtypes with different patterns of brain network damage based on multi-center functional magnetic resonance imaging data, and systematically evaluated the differences in functional connectivity, brain atrophy and cognitive ability of these subtypes, and the relevant research has been published online in the journal Biopsychiatry.

The framework for brain functional network research .png

Brain Functional Networks Research Framework Courtesy of Respondents

“AD has great heterogeneity (individualized differences) in terms of pathological distribution, brain atrophy, and clinical manifestations, which cause a high rate of early misdiagnosis of the disease and hinder clinical drug development.” Liu Yong told China Science Daily, “Patients with AD usually have abnormal brain activity involving multiple brain functional networks, resulting in continuous decline in cognitive ability. However, the heterogeneity of AD brain network abnormalities and their corresponding structural changes and cognitive decline are unclear. ”

The team used the MCAD database jointly established by many hospitals in China and the ADNI public database in the United States to collect 1100 cases of data, based on the AD abnormal brain network found in the previous period, using the non-negative matrix decomposition (NMF) method to cluster AD patients into different subtypes, and systematically evaluated the functional abnormalities, structural atrophy and longitudinal changes of each subtype. The results showed that there were four stable and reproducible functional subtypes in AD patients, each corresponding to a representative brain network (RFN), namely the prefrontal lobe network, the default network, the cingulate gingival-related network, and the basal ganglia-related network.

Further studies found that there were significant differences in network impairment patterns between the four AD functional subtypes. Subtype 1 has mild but diffuse abnormalities of the functional network of the whole brain relative to normal older adults; Functional network abnormalities of subtype 2 are mainly concentrated in the default network and accompanied by abnormal elevation of the prefrontal lobe network; Subtype 3 has a significant decrease in the anterior cingulate gyrus-related network and an abnormal increase in the prefrontal lobe network; Decreased functional networks of subtype 4 are concentrated in bilateral basal ganglia and are accompanied by abnormal elevation of the prefrontal network.

AD functional subtype features .png

AD functional subtype features Courtesy of the interviewee

More importantly, with the exception of subtype 1 of diffuse damage, specific damage patterns of other subtypes are highly reproducible on MCAD and ADNI datasets. In addition, brain network damage was most pronounced in each subtype of RFN, while damage was milder in RFNs of other subtypes.

“Demographically, the four subtypes are distributed across all sites in both databases and do not differ significantly in sex ratios.” Chen Pindong, the first author of the article and a doctoral candidate at the Institute of Automation of the Chinese Academy of Sciences, said, “In terms of age distribution, subtype 3 is older than subtype 1 and subtype 2; In terms of cognitive function, subtype 1 has the best cognitive ability, subtype 3 and subtype 4 are second, and subtype 2 has the worst cognitive ability. ”

The team further used follow-up information from the ADNI database to find that although subtype 1 has better cognitive ability, its disease progression rate is still very fast, and subtype 3, which has relatively poor cognitive ability, has the slowest decline rate. In addition, each subtype has different trajectories of change in different cognitive domains. This result suggests that identifying heterogeneity in brain networks can help uncover different patterns of cognitive impairment, and that precise interventions for these different subtypes are expected to be given in the future to effectively delay the disease process.

The functional connection impairment mode of the AD subtype .png

Functional connectivity impairment pattern of the AD subtype Courtesy of respondents

“Different subtypes of brain networks correspond to different patterns of brain atrophy and disease processes, which reflect potentially different pathological mechanisms and provide new perspectives for studying the heterogeneity of AD.” Zhou Bo said, “A stable and reproducible subtypes of AD based on multi-center, large-sample definitions are expected to lay a theoretical foundation for accurate diagnosis and treatment of AD, and provide technical support for optimizing clinical decision-making.” (Source: China Science Daily Zhang Shuanghu)

Related paper information:https://doi.org/10.1016/j.biopsych.2022.06.019



Source link

Related Articles

Leave a Reply

Your email address will not be published. Required fields are marked *

Back to top button