AUTHOR=Yu Qingyun , Wang Rong , Sun Chong , Hu Bin , Liu Xueling , Yang Liqin , Lin Jie , Geng Daoying , Li Yuxin 

TITLE=Dynamic reconfiguration and transition of whole-brain networks in patients with MELAS revealed by a hidden Markov model

JOURNAL=Frontiers in Neurology

VOLUME=Volume 16 - 2025

YEAR=2025

URL=https://www.frontiersin.org/journals/neurology/articles/10.3389/fneur.2025.1625888

DOI=10.3389/fneur.2025.1625888

ISSN=1664-2295

ABSTRACT=ObjectivesMitochondrial encephalomyopathy with lactic acidosis and stroke-like episodes (MELAS) is a rare maternally inherited disease. The neuropathologic mechanisms and neural network alterations underlying stroke-like episodes (SLEs), a recurrent paroxysmal clinical event, remain unclear. The hidden Markov model (HMM) can detect profound alterations in neural activities across the whole-brain network.Materials and methodsWe initially collected data from a prospective cohort from 2019 to 2024. The confirmed diagnosis of MELAS was conducted through genetic testing or a muscle biopsy. Healthy control volunteers were recruited from the local community. By utilizing the HMM, we evaluated the temporal characteristics and transitions of HMM states and the specific community pattern of transitions and activation maps of the whole brain for subjects.ResultsThirty-six MELAS patients at the acute stage (MELAS-acute group) and 30 healthy controls (HC group) were included in this study. Based on HMM, fractional occupancies in states 5 and 6 for MELAS were significantly decreased (p < 0.001), but fractional occupancies in states 2, 3, 4, 7, 8, 9, 10, and 11 were significantly increased (p < 0.05), compared to HCs. The lifetimes of HMM states showed a similar decrease as fractional occupancies. The switching frequency of HMM states was significantly increased in MELAS (p < 0.001). Combined with the special community patterns of transitions, MELAS displayed differential activity patterns in crucial areas of the default mode network (DMN) and visual network (VN).ConclusionThis study suggests dynamic reconfiguration of HMM states, special transition modules, and multiple transition pathways in MELAS, providing novel insights into the neural network mechanisms underlying MELAS.