Microscale Tissue Models
for Alpha Particle Dosimetry
Principal Investigator: Dr. Wesley Bolch
Department: Biomedical Engineering
NCI funded study (R43 CA224643) with Rapid Dosimetry, LLC to establish alpha-particle cellular-level dosimetry within organs at risk for toxicity in radiopharmaceutical cancer therapy. The project is a collaboration with the University of Florida Department of Anatomy and Cell Biology, and utilizes a large collection of tissue histology slides for model construction.
Two organs that are most susceptible to toxicity from radiopharmaceuticals and are considered dose limiting are bone marrow and the kidney. While the cell population of bone marrow may decline after a high dose delivery, it will eventually regenerate itself. On the other hand, the kidney may experience a loss of function if enough damage is delivered to the organ. Additionally, bone marrow transplants are more readily accessible than those of kindeys.
This project involves the creation of polygon mesh models of the kidney. This includes imaging histology sections from the University of Florida histology library and then segmenting major structures from these images in order to construct a 3D model. This model will be used in radiation transport simulations for radiopharmaceutical therapy.
Project Responsibilities
Segmentation
Using the provided histology slides, after being sorted, aligned, and imaged, each region of interest (or ROI) is mapped out and each structure is labeled using 3D Slicer. In the case of the kidney, the structures of interest are the proximal and distal tubules, Bowman's capsule, and glomeruli. My task included identifying the center of each of the tubules and following them through the expanse of the slide deck.
The Model: Rhino
Using Rhinoceros, a commercial 3D computer graphics and computer-aided design application software, the segmentation from 3D Slicer is imported and a pipe function is applied to make the center points into the tubules. The tubules must be formed to fit a goal volume for both distal and proximal. All functions and volume calculations are run through the visual programming language, Grasshopper. My specific task was space filling residual tubules, those which are not immediately attached to the glomeruli, until the goal volume is achieved. I have performed this task on two separate ROIs.