Leksell Gamma Knife is a mini-invasive technique toobtain a complete destruction of cerebral lesions delivering a singlehigh dose radiation beam. Positron Emission Tomography (PET)imaging is increasingly utilized for radiation treatment planning.Nevertheless, lesion volume delineation in PET datasets is challeng-ing because of the low spatial resolution and high noise level of PETimages. Nowadays, the biological target volume (BTV) is manuallycontoured on PET studies. This procedure is time expensive andoperator-dependent. In this article, a fully automatic algorithm for theBTV delineation based on random walks (RW) on graphs is proposed.The results are compared with the outcomes of the original RWmethod, 40% thresholding method, region growing method, andfuzzy c-means clustering method. To validate the effectiveness of theproposed approach in a clinical environment, BTV segmentation on18 patients with cerebral metastases is performed. Experimentalresults show that the segmentation algorithm is accurate and hasreal-time performance satisfying the physician requirements in aradiotherapy environment.
A fully automatic method for biological target volume segmentation of brain metastases
Stefano A;Russo G;Gilardi;
2016
Abstract
Leksell Gamma Knife is a mini-invasive technique toobtain a complete destruction of cerebral lesions delivering a singlehigh dose radiation beam. Positron Emission Tomography (PET)imaging is increasingly utilized for radiation treatment planning.Nevertheless, lesion volume delineation in PET datasets is challeng-ing because of the low spatial resolution and high noise level of PETimages. Nowadays, the biological target volume (BTV) is manuallycontoured on PET studies. This procedure is time expensive andoperator-dependent. In this article, a fully automatic algorithm for theBTV delineation based on random walks (RW) on graphs is proposed.The results are compared with the outcomes of the original RWmethod, 40% thresholding method, region growing method, andfuzzy c-means clustering method. To validate the effectiveness of theproposed approach in a clinical environment, BTV segmentation on18 patients with cerebral metastases is performed. Experimentalresults show that the segmentation algorithm is accurate and hasreal-time performance satisfying the physician requirements in aradiotherapy environment.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.