A method of correcting for linear drift in computed tomography brain scans

Area of Science:

• Radiology and Medical Imaging
• Neuroscience
• Quantitative Analysis

Background:

• Quantitative comparisons of serial computed tomography (CT) brain scans require correction for linear drift in X-ray attenuation coefficients.
• This drift can impede accurate assessment of neurological conditions over time.

Observation:

• Linear drift in X-ray attenuation coefficients is a common issue in serial CT brain imaging.
• The low end of attenuation coefficient frequency histograms, specifically cerebrospinal fluid values, can be used to detect this drift.

Findings:

• A percentile-percentile plot method effectively identifies and corrects linear drift in X-ray attenuation coefficients.
• This drift correction method allows for reliable quantitative comparisons between scans performed at different times.

Implications:

• Enables accurate serial quantitative assessment of white matter hypodensity progression or regression.
• Facilitates monitoring of conditions like drug-induced leukoencephalopathy.
• Improves the reliability of longitudinal neuroimaging studies.