A quantification strategy for missing bone mass in case of osteolytic bone lesions

Area of Science:

• Biomedical Engineering
• Preclinical Research
• Medical Imaging

Background:

• Breast cancer bone metastases are often destructive, necessitating accurate measurement for treatment assessment.
• Manual measurement of bone lesions in preclinical models is time-consuming and lacks reproducibility.
• Automated methods are crucial for reliable quantification of bone loss in metastatic disease.

Purpose of the Study:

• To develop and validate an automated segmentation method for measuring missing bone mass in a preclinical rat model of breast cancer bone metastases.
• To establish a reproducible technique for assessing bone lesion volumes in hind leg bones using 3D CT scans.
• To enable accurate monitoring of treatment response for bone remodeling therapies.

Main Methods:

• Utilized a preclinical rat model with induced bone metastases in the right hind leg.
• Employed the contralateral (left) hind leg as a healthy bone model.
• Segmented healthy bone using statistical shape and volume growing models, then mirrored and registered to the affected side for comparison.
• Identified osteolytic lesions by comparing gray values between healthy and affected bone models.

Main Results:

• The automated segmentation method successfully reconstructed missing bone structures.
• A mean ratio of reconstructed bone volume to healthy model bone volume (vr/vm) of 1.07 indicated good reconstruction accuracy.
• The method demonstrated reproducibility in identifying and measuring bone loss.

Conclusions:

• Comparing a modified bone structure to a healthy contralateral model is a reproducible approach for quantifying missing bone mass.
• The developed automated segmentation method provides a reliable tool for preclinical research on bone metastases.
• This technique supports reproducible assessment of bone remodeling therapies in metastatic breast cancer models.