Abstract
The diffusion and perfusion magnetic resonance (MR) images can provide functional information about tumour and enable more sensitive detection of the tumour extent. We aimed to develop a fuzzy feature fusion method for auto-segmentation of gliomas in radiotherapy planning using multi-parametric functional MR images including apparent diffusion coefficient (ADC), fractional anisotropy (FA) and relative cerebral blood volume (rCBV). For each functional modality, one histogram-based fuzzy model was created to transform image volume into a fuzzy feature space. Based on the fuzzy fusion result of the three fuzzy feature spaces, regions with high possibility belonging to tumour were generated automatically. The auto-segmentations of tumour in structural MR images were added in final auto-segmented gross tumour volume (GTV). For evaluation, one radiation oncologist delineated GTVs for nine patients with all modalities. Comparisons between manually delineated and auto-segmented GTVs showed that, the mean volume difference was 8.69% (±5.62%); the mean Dice’s similarity coefficient (DSC) was 0.88 (±0.02); the mean sensitivity and specificity of auto-segmentation was 0.87 (±0.04) and 0.98 (±0.01) respectively. High accuracy and efficiency can be achieved with the new method, which shows potential of utilizing functional multi-parametric MR images for target definition in precision radiation treatment planning for patients with gliomas.
Highlights
To avoid missing target and spare critical healthy brain tissue around the target volume in radiotherapy treatment planning of gliomas, cancerous tissue involvement must be correctly defined in the gross tumour volume (GTV) delineation
Accurate automatic segmentation of gliomas in multimodal magnetic resonance (MR) images remains a challenging issue for three main reasons: (1) the tumour heterogeneous nature and the difference in image acquisition method make one tumour shows different shape, size and location in different image modality (Fig. 1); (2) the ambiguous tumour border may deteriorate in the image with decreased resolution, especially in diffusion and perfusion parameter image such as apparent diffusion coefficient (ADC), fractional anisotropy (FA) and relative cerebral blood volume (rCBV); (3) the partial volume effects and the inherent noise in imaging system could produce negative influence to the segmentation results[9]
The MR images were acquired for each patient on a 1.5 T MR scanner (GE Signa Excite), including T1-weighted contrast-enhanced (T1C), T2, diffusion-weighted imaging (DWI), diffusion tensor imaging (DTI) and perfusion weighted (PWI) images
Summary
To avoid missing target and spare critical healthy brain tissue around the target volume in radiotherapy treatment planning of gliomas, cancerous tissue involvement must be correctly defined in the gross tumour volume (GTV) delineation. Dynamic-susceptibility contrast (DSC) and its derived parameter of relative cerebral blood volume (rCBV) can be used to assess the tumour vascularity Those diffusion and perfusion parameters have been investigated to define accurate tumour extent and delineate target volume in radiation treatment planning. Accurate automatic segmentation of gliomas in multimodal MR images remains a challenging issue for three main reasons: (1) the tumour heterogeneous nature and the difference in image acquisition method make one tumour shows different shape, size and location in different image modality (Fig. 1); (2) the ambiguous tumour border may deteriorate in the image with decreased resolution, especially in diffusion and perfusion parameter image such as ADC, FA and rCBV; (3) the partial volume effects and the inherent noise in imaging system could produce negative influence to the segmentation results[9]
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
