Automatic tracking of in vivo dendritic spines

A nueron's dendrites have tiny protrusions called dendritic spines. These spines have the important job of receiving input from a neighboring neuron, and thus are the subject of intense study. Most of these studies are done in vitro, but in vivo imaging of dendritic spines is possible with specialized equipment.

In vivo imaging allows spines to be watched over time in a natural environment and allows observations during disease progression. However, counting and comparing dendrites from image to image taken with time-lapse microscopy is quite labor intensive.

Researchers at the Methodist Hospital Research Institute & the Methodist Hospital, Weill Cornell Medical College in Houston, TX recently published a paper detailing a technique that automates spine detection and tracking in a live animal model. They tested the technique on multiphoton microscopy images of an anesthetized mouse model of Alzheimer’s disease. Their method was able to map the dendritic backbone and its associated spines, quantify spine length and area, track the growth or loss of spines, and deal with poor image quality.

As we learn more about neurodegenerative disease such as Alzheimer’s disease, in vivo imaging of animal models will be key for studying how the brain changes during disease progression or reacts to therapies. Automating key parts of image analysis will greatly aid these studies.

Free full text .pdf of research paper: Jing Fan, Xiaobo Zhou, Jennifer G. Dy, Yong Zhang and Stephen T. C. Wong, An Automated Pipeline for Dendrite Spine Detection and Tracking of 3D Optical Microscopy Neuron Images of In Vivo Mouse Models, Neuroinformatics.