Brain imaging reveals hierarchical topology changes and stage-dependent impairments in autoimmune encephalitis

Comprehensive Summary

Lin et al. studied how brain connections changed across different stages of autoimmune encephalitis (AE). The researchers used rs-fMRI and diffusion tensor imaging to compare 52 AE patients with 32 healthy controls, and 30 patients were followed over time from the acute to the recovery phase. They also examined the efficiency with which different brain regions communicated with each other and used machine learning models to test whether these patterns could distinguish between patients and controls and between the acute and recovery stages. Despite minor changes in global network structures across the world, the patients demonstrated significant changes in local network efficiency, especially in the medial occipital and inferior temporal lobes. Patients during the acute phase showed higher functional connectivity and lower structural connectivity, showing that there was a temporary discrepancy between the functional status and structural status of the brain. The classification models were effective in distinguishing AE from controls. Network efficiency measures also predicted clinical severity and length of hospitalization.

Outcomes and Implications

The study is important because it is often difficult to diagnose autoimmune encephalitis at an early stage, especially when the MRI scan results are normal. Identifying the stage-related changes in the efficiency of the brain network could be helpful in distinguishing active disease from recovery. The results of functional connectivity showed the best performance in the diagnosis and could potentially aid in the decision-making process and the evaluation of treatment response. Although the results of the model were limited due to the small sample size and lack of external validation, the results demonstrated the potential use of multimodal brain imaging in the diagnosis and prognosis of AE.