Monitoring brain tumor response to therapy using MRI segmentation

QA: Quality Assurance, Accuracy, Precision and Phantoms

“…Sometimes this is called the test-retest performance. It has been measured for many parameters, for example spectroscopy (Charles et al, 1996;Marshall et al, 1996;Simmons et al, 1998;Bartha et al, 2000;Chard et al, 2002), dynamic Gd imaging (Buckley, 2002;Galbraith et al, 2002;Padhani et al, 2002), fMRI (Tegeler et al, 1999;Loubinoux et al, 2001), tissue volume (Fox and Freeborough, 1997;Lemieux et al, 2000;Cardenas et al, 2001;Gasperini et al, 2001), lesion volume (Grimaud et al, 1996;Vaidyanathan et al, 1997;Rovaris et al, 1998;Filippi et al, 1998b), lesion counting (Rovaris et al, 1999;Wei et al, 2002), spinal cord cross-sectional area (Leary et al, 1999) and clinical scores (Cohen et al, 2000). Its value depends on the method used to measure the parameter, and is often very sensitive to the precise details of the data collection procedure (such as patient positioning and pre-scan procedure) and data analysis (particularly ROI placement).…”

Brain tumor target volume determination for radiation treatment planning through automated MRI segmentation

“…Both the kNN and KG segmentation methods have been clinically applied as a technique for more accurately measuring tumor volume variation in the brain (15,18). This work evaluates the performance of kNN as a representative of operator-assisted semiautomated segmentation and KG as a promising candidate for fully automated GTV determination.…”

Connectivity-based local adaptive thresholding for carotid artery segmentation using MRA images

“…Validation is typically performed using one of two different types of truth models. The most straightforward approach to validation is by comparing the automated segmentations with manually obtained segmentations [41][42]. This approach, besides suffering from the drawbacks outlined (complex and time consuming), does not guarantee a perfect truth model since an operator's performance can also be flawed.…”