Measurement-oriented deep-learning workflow for improved segmentation of myelin and axons in high-resolution images of human cerebral white matter.

BACKGROUND: Standard segmentation of high-contrast electron micrographs (EM) identifies myelin accurately but does not translate easily into measurements of individual axons and their myelin, even in cross-sections of parallel fibers. We describe automated segmentation and measurement of each myelinated axon and its sheath in EMs of arbitrarily oriented human white matter from autopsies. NEW METHODS: Preliminary segmentation of myelin, axons and background by machine learning, using selected filters, precedes automated correction of systematic errors. Final segmentation is done by a deep neural network (DNN). Automated measurement of each putative fiber rejects measures encountering pre-defined artifacts and excludes fibers failing to satisfy pre-defined conditions. RESULTS: Improved segmentation of three sets of 30 annotated images each (two sets from human prefrontal white matter and one from human optic nerve) is achieved with a DNN trained only with a subset of the first set from prefrontal white matter. Total number of myelinated axons identified by the DNN differed from expert segmentation by 0.2%, 2.9%, and -5.1%, respectively. G-ratios differed by 2.96%, 0.74% and 2.83%. Intraclass correlation coefficients between DNN and annotated segmentation were mostly >0.9, indicating nearly interchangeable performance. COMPARISON WITH EXISTING METHOD(S): Measurement-oriented studies of arbitrarily oriented fibers from central white matter are rare. Published methods are typically applied to cross-sections of fascicles and measure aggregated areas of myelin sheaths and axons, allowing estimation only of average g-ratio. CONCLUSIONS: Automated segmentation and measurement of axons and myelin is complex. We report a feasible approach that has so far proven comparable to manual segmentation.

Metabolic abnormalities in fronto-striatal-thalamic white matter tracts in schizophrenia.

The anterior limb of the internal capsule (ALIC) is the major white matter tract providing reciprocal connections between the frontal cortex, striatum and thalamus. Mounting evidence suggests that this tract may be affected in schizophrenia, with brain imaging studies reporting reductions in white matter volume and density, changes in fractional anisotropy and reduced asymmetry. However, the molecular correlates of these deficits are currently unknown. The aim of this study was to identify alterations in protein and metabolite levels in the ALIC in schizophrenia. Samples were obtained post-mortem from individuals with schizophrenia (n=15) and non-psychiatric controls (n=13). Immunoreactivity for the myelin-associated protein myelin basic protein (MBP), and the axonal-associated proteins phosphorylated neurofilament and SNAP-25 was measured by enzyme-linked immunoadsorbent assay (ELISA). Metabolite concentrations were quantified by proton nuclear magnetic resonance ((1)H NMR) spectroscopy. Levels of myelin- or axonal-associated proteins did not differ between groups. Overall differences in metabolite concentrations were observed between the two groups (MANOVA F=2.685, p=0.036), with post-hoc tests revealing lower lactate (19%) and alanine (24%) levels in the schizophrenia group relative to controls. Observed changes in lactate and alanine levels indicate metabolic abnormalities within the ALIC in schizophrenia.