Michael Rogalski
UIUC/SDU Collaboration: Multimodal Image Coregistration, 3D Histology, and Smørrebrød
Thanks to the ISB travel grant, I spent four weeks collaborating with experts in histomorphometry at the University of Southern Denmark (SDU). Our goal was to combine our expertise in CT image processing, finite element analysis and high-resolution microscopy imaging in order to create a computational pipeline that will allow for a quantitative understanding of the mechanisms leading to dysfunctional bone remodeling processes. Additionally, we aimed to automate our collaborators’ laborious processing of histological slides using our image processing techniques as a baseline. Ultimately, the long-term goal of our collaboration is to advance us towards our mission of identifying targets for the diagnosis and treatment of musculoskeletal diseases such as osteoporosis.
First, (after overcoming jetlag) I aimed to understand the technical challenges associated with the process of sectioning bone biopsies, section imaging, and histomorphometry. We successfully automated the evaluation of histological sections by converting stained images with labelled remodeling events into binary images. Our deliverable was processing over 200 images and then aligning this 2D image stack into one coordinate system in 3D to begin understanding the relationship between remodeling events along the intracortical pore network. Semi-automated processing of aligned histological sections will help us answer questions such as whether remodeling events propagate throughout longitudinal or transverse canals and what causes them to ultimately diverge?
Second, I aimed to coregister microCT images and histological sections of PTH-treated femoral neck samples. We created a protocol wherein microCT scans before and after sectioning were used to align the image volumes into the same coordinate system in order to locate the 3D position of the histological cutting plane (Figure 1). Back at UIUC, we are now creating microFE models of whole femoral neck samples and cortical bone subsections. With this final piece, we will have CT, histological, and FE data (such as strain energy density) aligned in the same global coordinate system in order to correlate local mechanical strength and histological events.
Throughout our ongoing collaboration, I witnessed firsthand how our Danish colleagues share our values for interdisciplinary projects and the importance of working as a team towards a collective goal across research fields. Outside of the lab, I spent time exploring Danish cities and culture, easily travelling across Denmark thanks to their high-speed train and biking infrastructure, as well as the kindness of Danish strangers. Of course, you cannot visit Denmark without mentioning smørrebrød. Open face sandwiches with cold cuts, salmon, cheese, and/or garnishes on a rectangular piece of rye bread is the perfect meal for a sunny summer afternoon. Although I was only in Denmark for four weeks, it was easy to see why it is consistently ranked one of the happiest countries in the world. Thank you to ISB for the incredible opportunity for this collaboration; working with experts in histomorphometry and experiencing a new culture will forever be one of the highlights of my graduate school experience.
Sincerely,
Michael Rogalski
