Introduction
This user manual describes the use of Canon Medical’s CT volumetry system for measuring malignant pleural mesothelioma (MPM) tumours.
Background
MPM is an incurable cancer associated with asbestos exposure, and its complex tumour morphology makes radiological response assessment challenging. The current standard, modified Response Evaluation Criteria in Solid Tumours (mRECIST), has limitations, including poor reproducibility and oversimplification of true tumour burden.
Methodology
The system uses a fully automated Convolutional Neural Network (CNN) for volumetric MPM segmentation, which was trained and validated using a multicentre retrospective cohort study.
The study included 183 CT datasets from 108 patients, with 80 fully annotated datasets used for training and 60 datasets used for external validation.
Results
Human and AI volumes were strongly correlated, with a mean bias of +31 cm³ (p=0.182) and 95% limits of -345 to +407 cm³.
The system demonstrated excellent intraobserver agreement (ICC 0.997, p=<0.0001) and moderate interobserver agreement (ICC 0.732, p=0.001).
Conclusion
The Canon Medical CT volumetry system provides a fully automated and accurate tool for measuring MPM tumours, with potential to improve response assessment and patient care.
Manual:
Download: Canon Medical CT Volumetry User Manual
Chat: Powered By VoiceSphere