A standardized Method to Automatically Segment Amyloid Plaques in Congo Red Stained Sections from Alzheimer Transgenic Mice

Abstract

Automated detection of amyloid plaques (AP) in post mortem brain sections of patients with Alzheimer disease (AD) or in mouse models of the disease is a major issue to improve quantitative, standardized and accurate assessment of neuropathological lesions as well as of their modulation by treatment. We propose a new segmentation method to automatically detect amyloid plaques in Congo Red stained sections based on adaptive thresholds and a dedicated amyloid plaque/tissue modelling. A set of histological sections focusing on anatomical structures was used to validate the method in comparison to expert segmentation.

Automatic segmentation of amyloid plaques in MR images using unsupervised support vector machines

“…This study describes an automatic segmentation algorithm that can be used for amyloid PL quantification in MR images. The high correlation between PLs quantified from MR imaging using the proposed algorithm and the gold standard histological measurements is comparable to that reported in literature for validating an automatic method of plaque segmentation from Congo Red‐stained brain sections (37, 38). Although Teboul et al (38) compared their method based on adaptive thresholds to expert segmentation of 2D histology sections, our validation scheme is more complex as our data include volumetric ROIs from multiple brain regions in images acquired by two different modalities.…”

“…The high correlation between PLs quantified from MR imaging using the proposed algorithm and the gold standard histological measurements is comparable to that reported in literature for validating an automatic method of plaque segmentation from Congo Red‐stained brain sections (37, 38). Although Teboul et al (38) compared their method based on adaptive thresholds to expert segmentation of 2D histology sections, our validation scheme is more complex as our data include volumetric ROIs from multiple brain regions in images acquired by two different modalities. Furthermore, the validation of the proposed algorithm by histology shows that the OCSVM ν estimation method is suitable for segmenting plaques in MR data and that it can be used to compute a ν range where the PL values are stable.…”

High-throughput quantification of Alzheimer’s disease pathological markers in the post-mortem human brain

“…A variety of (semi-) automated methods have been developed with a vision to standardise quantitative analysis of the neuropathological markers of Alzheimer's disease (Chubb et al, 2006;Cummings et al, 2002;Defigueiredo et al, 1995;Hibbard et al, 1994;Hibbard and McKeel, 1997;McKenzie et al, 1996;Teboul et al, 2007). However, some of these involve complex segmentation or algorithmic processing and can require high-level mathematical knowledge in order to replicate the protocol.…”

High content analysis in amyotrophic lateral sclerosis

“…Both sporadic and familial cases of ALS present accumulation of SOD1 and/or of TDP43 159 proteins; it is yet to be determined whether they are a source of toxicity for the cells or a defense mechanism adopted to segregate toxic material. HCA protocols for quantifying cellular aggregates have already been successfully validated for Huntington’s 160 and Alzheimer’s 161,162 disease. As demonstrated by some of the studies reviewed in this article, HCA systems can perform automated, unbiased quantification of aggregates in ALS settings at low throughput.…”