Author: Bryan Bednarz, Malick Bio Idrissou, Campbell Haasch, Reinier Hernandez 👨🔬
Affiliation: Department of Medical Physics, School of Medicine and Public Health, University of Wisconsin - Madison, Department of Medical Physics, School of Medicine and Public Health, University of Wisconsin–Madison, University of Wisconsin - Madison 🌍
Purpose: Quantitative optical imaging is a powerful tool in murine models for assessing tumor growth and metastatic spread using bioluminescence imaging (BLI) and for detecting radiopharmaceutical uptake of challenging-to-image radiotracers using the emerging technology of Cherenkov luminescence imaging (CLI). However, accurate serial signal quantification in BLI/CLI is hindered by its dependence on shifting anatomical structures not visible in planar imaging. To overcome these limitations, we developed an intuitive, GUI-based platform for registration of optical images with 3D anatomical data, deblurring of optical images using a Geant4-based point spread function (PSF), and quantification of optical signal sources. The platform was evaluated by measuring In vivo activity in tumor-bearing mice injected with ⁸⁶Y-NM600 through CLI.
Methods: A Python-based platform was developed to process, register, deblur, and quantify optical images with uCT. A calibration procedure was established using a 96-well plate filled with a tissue-mimicking intralipid/blood mixture and varying amounts of the optical signal source to calibrate radiance as a function of source quantity and depth in tissue. Four nude mice bearing MC38 right-flank tumors were injected with 250 µCi of ⁸⁶Y-NM600 and imaged using CLI and PET/CT at 3-, 24-, 48-, and 96-hours post-injection. Tumor and liver activity were quantified at each timepoint using the developed platform, with PET imaging serving as a reference for comparison.
Results: Quantification of liver activity using the developed platform differed by an average of 7% from PET-derived values. While CLI-based tumor activity measurements were on average 30% higher than PET values, the CLI measurements maintained proportionality across all mice and timepoint
Conclusion: The developed GUI-based platform provides a user-friendly solution for the quantification of optical signals in preclinical models. With plans for open-source release, this platform has the potential to become a widely accessible tool for optical signal quantification in preclinical research.