Quantitative analysis methods for studying fenestrations in liver sinusoidal endothelial cells. A comparative study

Liver Sinusoidal Endothelial Cells (LSEC) line the hepatic vasculature providing blood filtration via trans-membrane nanopores called fenestrations. These structures are 50-300 nm in diameter, which is below the resolution limit of a conventional light microscopy. To date, there is no standardized method of fenestration image analysis. With this study, we provide and compare three different approaches: manual measurements, a semi-automatic (threshold-based) method, and an automatic method based on user-friendly open source machine learning software. Images were obtained using three super resolution techniques - atomic force microscopy (AFM), scanning electron microscopy (SEM), and structured illumination microscopy (SIM). Parameters describing fenestrations such as diameter, area, roundness, frequency, and porosity were measured. Finally, we studied the user bias by comparison of the data obtained by five different users applying provided analysis methods.

Automated summary

This study utilized three super resolution techniques to obtain images of liver sinusoidal endothelial cell fenestrations and compared three different methods of fenestration image analysis. In addition to comparing analysis methods, user bias was also studied through comparison of data obtained by different users.