Pixcell: Novel Automated Software for Adipocyte and Lipid Tracing in Immunofluorescent Images

Adipose tissue is a highly dynamic endocrine organ that serves as the body’s primary energy reservoir through the storage and mobilization of lipids. Adipocyte cellular size can be indicative of cellular status; for example, hypertrophic adipocytes are more prone to insulin resistance and the secretion of pro-inflammatory cytokines. Thus, accurate analysis of lipid size and distribution is vital, especially in biopsy samples, disease modeling, and regenerative tissue constructs. Presently, available software solutions are optimized for hematoxylin-eosin images where lipids are removed during the processing and appear as voids. Unfortunately, these softwares have difficulty with confocal derived z-stack images, struggling to capture cells both in the foreground and background of an image due to the interference of shadows and blemishes. This limitation has made manual analysis with ImageJ the most commonly utilized method for three-dimensional images. However as ImageJ is cumbersome, time consuming, and adds a degree of user-dependent subjectivity to analysis, we developed PixCell, an automated Matlab-based software, to reduce the subjectivity and time involved in adipocyte size analysis. PixCell sequentially thresholds the images and creates masking layers to capture the increased dimensionality of these images. The algorithm then cross-references cells captured across thresholding steps to prevent double counting. With its stepwise thresholding and masking steps, PixCell retains an average of >80% accuracy when tested on a variety of complex samples, including excised human adipose tissue, adipocyte laden collagen gels, and lipoaspirate seeded silk scaffold while reducing analysis time from hours to seconds. Additionally, PixCell has the added feature of tracking much smaller extracellular lipids, which are an item of interest in intercellular communication and disease states. This makes PixCell a powerful, user-friendly tool for use in both the clinical and preclinical setting, while greatly enhancing the efficiency and accuracy of lipid size analysis. While this tool was initially developed for adipose tissue analysis, it has broad applicability and future use in tracking cell proliferation and development, analyzing hepatocyte organoids, as well as investigating colloids.