A comparison of machine learning approaches for the quantification of microglial cells in the brain of mice, rats and non-human primates.

Microglial cells are brain-specific macrophages that swiftly react to disruptive events in the brain. Microglial activation leads to specific modifications, including proliferation, morphological changes, migration to the site of insult, and changes in gene expression profiles. A change in inflammatory status has been linked to many neurodegenerative diseases such as Parkinson's disease and Alzheimer's disease. For this reason, the investigation and quantification of microglial cells is essential for better understanding their role in disease progression as well as for evaluating the cytocompatibility of novel therapeutic approaches for such conditions. In the following study we implemented a machine learning-based approach for the fast and automatized quantification of microglial cells; this tool was compared with manual quantification (ground truth), and with alternative free-ware such as the threshold-based ImageJ and the machine learning-based Ilastik. We first trained the algorithms on brain tissue obtained from rats and non-human primate immunohistochemically labelled for microglia. Subsequently we validated the accuracy of the trained algorithms in a preclinical rodent model of Parkinson's disease and demonstrated the robustness of the algorithms on tissue obtained from mice, as well as from images provided by three collaborating laboratories. Our results indicate that machine learning algorithms can detect and quantify microglial cells in all the three mammalian species in a precise manner, equipotent to the one observed following manual counting. Using this tool, we were able to detect and quantify small changes between the hemispheres, suggesting the power and reliability of the algorithm. Such a tool will be very useful for investigation of microglial response in disease development, as well as in the investigation of compatible novel therapeutics targeting the brain. As all network weights and labelled training data are made available, together with our step-by-step user guide, we anticipate that many laboratories will implement machine learning-based quantification of microglial cells in their research.

Automated quantification of neuronal swellings in a preclinical rodent model of Parkinson's disease detects region-specific changes in pathology.

BACKGROUND: The development of axonal pathology is a key characteristic of many neurodegenerative disease such as Parkinson's disease and Alzheimer's disease. With advanced disease progression, affected axons do display several signs of pathology such as swelling and fragmentation. In the AAV vector-mediated alpha-synuclein overexpression model of Parkinson's disease, large (> 20 µm2) pathological swellings are prominent characteristics in cortical and subcortical structures. NEW METHOD: This report describes a novel, macro-based workflow to quantify axonal pathology in the form of axonal swellings in the AAV vector-based alpha-synuclein overexpression model. Specifically, the approach is using background correction and thresholding before quantification of structures in 3D throughout a tissue stack. RESULTS: The method was used to quantify TH and aSYN axonal swellings in the prefrontal cortex, striatum, and hippocampus. Regional differences in volume and number of axonal swellings were observed for both in TH and aSYN, with the striatum displaying the greatest signs of pathology. COMPARISON WITH EXISTING METHODS: Existing methods for the quantification of axonal pathology do either rely on proprietary software or are based on manual quantification. The ImageJ workflow described here provides a method to objectively quantify axonal swellings both in volume and number. CONCLUSION: The method described can readily assess axonal pathology in preclinical rodent models of Parkinson's disease and can be easily adapted to other model systems and/or markers.

The role of sex and emotion on emotion perception in artificial faces: An ERP study.

Sex has a significant impact on the perception of emotional expressions. However, it remains unclear whether sex influences the perception of emotions in artificial faces, which are becoming popular in emotion research. We used an emotion recognition task with FaceGen faces portraying six basic emotions aiming to investigate the effect of sex and emotion on behavioural and electrophysiological parameters. 71 participants performed the task while EEG was recorded. The recognition of sadness was the poorest, however, females recognized sadness better than males. ERP results indicated that fear, disgust, and anger evoked higher amplitudes of late positive potential over the left parietal region compared to neutral expression. Females demonstrated higher values of global field power as compared to males. The interaction between sex and emotion on ERPs was not significant. The results of our study may be valuable for future therapies and research, as it emphasizes possibly distinct processing of emotions and potential sex differences in the recognition of emotional expressions in FaceGen faces.