Screening for Cognitive Impairment by Model-Assisted Cerebral Blood Flow Estimation

Area of Science:

• Neuroscience
• Medical Imaging
• Physiological Modeling

Background:

• Alzheimer's disease (AD) is a progressive neurodegenerative disorder affecting millions globally.
• Early diagnosis of AD is critical for timely intervention and treatment.
• Cerebral hypoperfusion and blood-brain barrier dysfunction are implicated in early AD stages.

Purpose of the Study:

• To investigate the utility of ambulatory measurements for predicting 24-hour cerebral blood flow (CBF) profiles.
• To assess the effectiveness of model-assisted CBF prediction in diagnosing mild cognitive impairment (MCI).
• To enhance early AD diagnosis by integrating physiological modeling with clinical data.

Main Methods:

• Combined physiological modeling with Holter blood pressure monitoring and carotid ultrasound imaging.
• Predicted 24-hour CBF profiles in 103 participants (53 MCI, 50 healthy controls).
• Utilized lasso regression with model-predicted CBF and neuropsychological test scores for MCI diagnosis.

Main Results:

• A CBF-enhanced classifier achieved an AUC of 0.889, outperforming a classifier based solely on neuropsychological tests (AUC = 0.818).
• Model-predicted CBF was validated against arterial spin-labeling MRI in an independent cohort.
• Established a correlation between measured CBF and model-predicted CBF.

Conclusions:

• Ultrasound imaging and ambulatory blood pressure monitoring, enhanced by physiological modeling, significantly improve MCI diagnostic accuracy.
• This approach offers a promising non-invasive method for early detection of neurodegenerative changes.
• The findings support the screening of chronic cerebral hypoperfusion for early AD diagnosis.