A fully automatic method for biological target volume segmentation of brain metastases

Abstract

Leksell Gamma Knife is a mini-invasive technique to obtain a complete destruction of cerebral lesions delivering a single high dose radiation beam. Positron Emission Tomography (PET) imaging is increasingly utilized for radiation treatment planning. Nevertheless, lesion volume delineation in PET datasets is challenging because of the low spatial resolution and high noise level of PET images. Nowadays, the biological target volume (BTV) is manually contoured on PET studies. This procedure is time expensive and operator-dependent. In this article, a fully automatic algorithm for the BTV delineation based on random walks (RW) on graphs is proposed. The results are compared with the outcomes of the original RW method, 40% thresholding method, region growing method, and fuzzy c-means clustering method. To validate the effectiveness of the proposed approach in a clinical environment, BTV segmentation on 18 patients with cerebral metastases is performed. Experimental results show that the segmentation algorithm is accurate and has real-time performance satisfying the physician requirements in a radiotherapy environment

MedGA: A novel evolutionary method for image enhancement in medical imaging systems

“…The proposed method improves and combines two computerassisted and operator-independent methods to segment BTV and GTV from PET [584][585][586] and MR [413,525] images, respectively. Each single method has been already validated [413,584], comparing each of them against the most common literature segmentation methods. The former uses the graph-based RW algorithm [221] and the latter is based on unsupervised FCM clustering [52,53].…”

A Survey on Nature-Inspired Medical Image Analysis: A Step Further in Biomedical Data Integration

“…In medical imaging, especially when dealing with Positron Emission Tomography (PET), the RW method can localise weak edges as part of consistent boundaries. To obtain the Biological Target Volume (BTV) delineation on metabolic PET images, the RW incorporates the PET information, by replacing the pixel intensities with the Standardised Uptake Value (SUV) and computing the arc weights accordingly [30]. In the original method [29], the known nodes are marked by user input, whilst an automatic seed-selection procedure makes the method more reliable [31].…”

A smart and operator independent system to delineate tumours in Positron Emission Tomography scans

“…Further, in the case of brain lesions, as reported in Ref. [16], a fully automatic method can be implemented by taking advantage of the great sensitivity and specificity of MET radiotracers in the discrimination between healthy versus cancer. Hence initial user input, which we require in FDG-PET studies, could be completely avoided in brain studies.…”

“…A typical example is the brain, where high uptake of glucose is a normal condition. It has been reported that the extent of tumour cell invasion in brain metastases can be detected by MET PET even more clearly than by CT or MRI [16,17]. For this reason, we considered MET as an alternative radio-tracer.…”