Polarimetric analysis of the Alzheimer’s pathology in excised mouse brain tissue

SignificanceGiven the rapidly ageing global population and the projected rise in dementia cases, research into Alzheimer’s disease (AD) has become an urgent scientific and medical priority. Continued investigation into the molecular mechanisms and early detection of AD is therefore essential to mitigate its growing personal and societal impact. Most current AD therapies in advanced phases of development target amyloid β-peptide (Aβ) production, aggregation, or accumulation.AimMueller Matrix Polarimetry (MMP) has evolved into a prominent research subject, with a focus on identifying microstructural changes in biotissues. This is done by investigating light properties, which is especially useful in the early detection of brain cell degradation, among others. We set up and employed experimental MMP at three illuminating laser wavelengths (i.e., 445, 532, and 632 nm).ApproachWe investigated the application of MMP to mouse brain tissue containing Aβ plaques. The investigated samples on glass slides consisted of paraffin-embedded brain and paraffin-embedded tissue slices. The tissues were taken at various stages of ageing, i.e., 75, 100, 125, 150, 175, 200, and 225 days. Paraffin tissue blocks were used as an additional sample set for comparison.ResultsWe performed a comparative analysis based on the Mueller matrix elements for each age category and highlighted the importance of certain elements, e.g., m44, for further analysis. We also compared the trends of decomposition parameters and could correlate them with the ageing. Contrary to previous studies, we also report on retardation, diattenuation, and polarizance changes for later AD changes.ConclusionsFrom the higher-order statistics, we concluded that the mean and standard deviation remained constant across the ages. Skewness values were positive and increased as the age progressed, whereas kurtosis decreased with age. The large available dataset opens the possibility of implementing machine learning methods to assist clinical diagnosis in the future.