Preoperative brain connectome predicts postoperative changes in processing speed in moyamoya disease

Area of Science:

• Neuroscience
• Medical Imaging
• Machine Learning

Background:

• Moyamoya disease is a rare cerebrovascular disorder often leading to cognitive dysfunction.
• Surgical revascularization is a common treatment, but its impact on neurocognitive outcomes remains debated.
• Early detection of postoperative cognitive changes is crucial for patient management.

Purpose of the Study:

• To develop machine learning models predicting postoperative processing speed changes in moyamoya disease patients.
• To utilize preoperative resting-state functional connectivity data for these predictions.
• To investigate the relationship between brain connectivity and cognitive recovery after surgery.

Main Methods:

• Applied connectome-based predictive modeling using preoperative resting-state functional connectivity.
• Developed machine learning models to predict changes in processing speed at 1 and 6 months post-surgery.
• Recruited 12 adult moyamoya disease patients and 20 healthy controls for comparison.

Main Results:

• Preoperative functional connectivity significantly predicted postoperative processing speed changes at both 1 and 6 months.
• Specific cerebro-cerebellar and cortico-subcortical connectivities were associated with processing speed.
• No significant behavioral changes in processing speed were observed at 1 or 6 months post-surgery compared to baseline.

Conclusions:

• Preoperative resting-state functional connectivity shows potential for predicting post-surgical cognitive changes in moyamoya disease.
• Findings highlight the role of specific brain networks in processing speed recovery.
• This predictive approach offers valuable insights for clinical management and patient care.