This observational study will investigate the properties of image files standardly collected during radiation therapy treatment in a cross-section of liver cancer patients who received stereotactic ablative body radiation therapy (SABR) after trans-catheter arterial chemo emobilisation (TACE). Specifically, it will determine whether the radio-opaque contrast agents in the image files can be detected by tumour-tracking software (KIM).
* Sites will collect image files from the Cancer Radiation Therapy Systems and the medical records of eligible participants. For patient who do not have implanted markers, research personnel (a radiation oncologist, medical physicist or radiation therapist) will manually outline the visible contrast mass. This outline is considered the 'ground truth' of the location of the contrast mass. To ensure its accuracy, the ground truth contrast mass location will be checked by research personnel from another study site. * Image and ground truth data collected from 80% of participants will be used by University of Sydney researchers to develop the KIM algorithm to detect and track the radio-opaque contrast agents with sufficient robustness to provide justification for the prospective use of KIM as a real-time image guided radiation therapy tool for liver cancer patients. * The developed KIM software will be applied by site Medical Physicist(s) to the treatment imaging data from the remaining 20% of participants the images. Tumour positions produced by KIM will be compared with a 'ground truth' manual delineation.
Study Type
OBSERVATIONAL
Enrollment
50
Calvary Mater Newcastle Hospital
Waratah, New South Wales, Australia
NOT_YET_RECRUITINGWestmead Hospital
Westmead, New South Wales, Australia
NOT_YET_RECRUITINGPrincess Alexandra Hospital
Woolloongabba, Queensland, Australia
RECRUITINGProportion of images that have contrast mass detection by software
Proportion of images in which kilovoltage intrafraction monitoring software detects contrast agent mass on CBCT and intra-treatment images
Time frame: 2 weeks
Average difference between software and ground truth measure
Average difference (mm) of location of contrast mass detection on CBCT and treatment images by KIM software compared to the ground truth in each of the horizontal and vertical directions.
Time frame: 2 weeks
Standard deviation of difference between software and ground truth measure
Standard deviation difference (mm) of location of contrast mass detection on CBCT and treatment images by KIM software compared to the ground truth in each of the horizontal and vertical directions.
Time frame: 2 weeks
Modelling of patient or treatment features that contributed to the success or failure of the KIM method
Using the KIM success rate (primary hypothesis) and patient CT imaging features (contrast agent density, size, shape and location, patient size etc) a generalised linear model, or appropriate alternative, will be used to identity univariate and multivariate patient or treatment features that contributed to the success or failure of the KIM method. Selected features will be identified and developed into a model. If the model predicts the likelihood of KIM success in at least 85% of the patients then the study will be deemed to have achieved the hypothesis.
Time frame: 2 weeks
This platform is for informational purposes only and does not constitute medical advice. Always consult a qualified healthcare professional.
Peter MacCallum Cancer Centre
Melbourne, Victoria, Australia
NOT_YET_RECRUITINGThe Austin Hospital
Melbourne, Victoria, Australia
NOT_YET_RECRUITING