Preclinical Positron Emission Tomography with Body Conforming Animal Molds for Cloud-Based Automated Image Analysis in Mice

We sure demonstrate how to perform mouse in vivo PET imaging using these body-conforming molds and how to navigate the associated data analysis platform. This allows us to keep all animals in the same physical position across time points, and can significantly accelerate and standardize downstream image analysis by automation. In vivo imaging presents multiple advantages compared to traditional non-imaging experiments.

For example, imaging allows systemic, holistic evaluation of a drug's distribution and pharmacological effects in the entire subject, which enables not only monitoring of efficacy at the target site, but also the detection of unexpected offsite activities. Obtaining highly reproducible and quantitative data is critical for drug development. For example, a PET-based bowel distribution study generates a massive amount of valuable information on a compound's kinetics, tissue accumulation, and target binding.

However, there is a lack of efficient automated processes for analyzing those data. We here show that the PET CT images acquired with these BCAMs are highly uniform and can be analyzed in batches by the cloud-based automated segmentation tool, which saves a tremendous amount of time. Furthermore, we can show that the automated analysis yielded satisfactory results consistent with our manual method.

Our next step is to conduct a more comprehensive evaluation of the performance of this automated tool. We will closely compare the ROI results generated by the SAS with those generated by human analysts with various experience levels, and compare in vivo imaging results to ground truth ex vivo quantitation. To begin, attach a body-conforming animal mold, or BCAM, of the appropriate size to the preclip with the top flipped up 90 degrees.

Place an anesthetized mouse on the base and extend its limbs. Then gently place a BCAM tumor injection template mold on the dorsal side of the mouse. Carefully create a temporary circular reference mark at the desired cutout location using a marker.

Once the mark is made, remove the template mold. Using the temporary marks as a reference, lift the skin with fingers or forceps to create a tent. Slowly insert the needle and inject tumor cell suspension subcutaneously at the desired location.

Discard the needle in a biohazard sharps container. Carefully draw up a target dose of fluorine 18 FDG into a sterile syringe. Place the syringe into the dose calibrator well chamber, and lower the dipper to measure the radioactive dose.

Take a stable reading from the dose calibrator. Record the date and time. Before injection, record the mouse weight.

Transfer the mouse to a suitable restrainer and locate the tail vein. Then inject the dose into the conscious mouse and record the injection time. Place the syringe aside, and clean excess blood from the tail using gauze.

Then remove the mouse from the restrainer and place it back in the appropriate housing. Now place the syringe back in the dose calibrator chamber, and record any residual readings for dose calculation. Record the date and time of the residual activity reading.

60 minutes after the intravenous administration of fluorine 18 FDG, transfer the anesthetized mouse into an appropriately sized BCAM based on its body weight. Gently pick up the mouse by the nape of the neck. Insert the tail into the tail slot, and wrap it under the BCAM.

Then place the tail on the tail platform and secure it with tape. Ensure the spine is straight, and gently close the BCAM. Position all four limbs on the BCAM paw platforms, and secure with tape as needed.

Then gently snap the BCAM into the imaging bed shuttle by inserting the front end first. Push down on the back of the BCAM until a click indicates it is secured to the G8 shuttle. To begin open the G8 PET acquisition software.

Enter the study details into the software, including the mouse's weight, date, time, initial dose amount, injection details, and residual activity readings. Next, insert the imaging bed shuttle with the mouse into the scanner opening. Select the appropriate PET CT parameters, and acquire imaging data.

After the acquisition is complete, remove the shuttle with the mouse and insert it into the dock. Press the BCAM vertical tab to release it from the G8 shuttle. Gently pull up and remove the BCAM with the mouse.

Press on the two retaining tabs to flip up and open the top of the BCAM. Then gently remove the mouse. For image analysis, log in to app.invivo.

ax, and create a project. Next, click on the red Upload tab at the top right of the window. Select the imaging system and reporter used in the study.

Then choose the folder with the imaging data. Navigate to the Annotation tab within the project folder. Click on Annotate to add relevant scan information including subject name, sex, BCAM size, subject weight, cohort name, time point, and injected dose value.

For data analysis, click on scans to navigate through the project, group, and individual subject levels. Then select Analyze. At the top right of the analysis ribbon window under Project, select the plus symbol and choose Organ Probability MAP, or OPM ROI.

Linear regression analysis showed that the brain had the highest correlation between the automated OPM analysis and manual methods, reflecting high accuracy. The right kidney and heart showed a moderate correlation in the OPM analysis compared to the manual analysis. The spleen had the lowest correlation, likely due to challenges in manual segmentation, making it difficult to distinguish from nearby soft tissues.