Microglial cell proliferation in neural tissue occurs during infections, neurological disease, neurotoxicity, and other conditions. In common clinical studies, quantification of microglial proliferation requires an extensive degree of manual cell counting by a trained expert, but this approach is subjective, error prone, time- and labor-intensive. Automatic methods are needed such as Machine Learning. Some previous work has been done on the automatic segmentation of these cells at high magnification, but they require a human-in-the-loop in one form or another.
An alternative method has been purposed that uses Deep Learning at low magnification. This method uses an ensemble of snapshots to automatically classify mouse brains as having high or low density of cells based on the classification of images at low magnification with minimal expert time requirement. It has been seen on a novel dataset of 14 mice that this method gives quick and accurate estimates of cell density at low magnification. This approach could potentially benefit a wide variety of studies across the diverse disciplines of neuroscience.
Results for n=14 mouse brain predictions