Automatic pharynx and larynx cancer segmentation framework (PLCSF) on contrast enhanced MR images. (March 2017)

• Record Type: Journal Article
• Authors: Doshi, Trushali; Soraghan, John; Petropoulakis, Lykourgos; Di Caterina, Gaetano; Grose, Derek; MacKenzie, Kenneth; Wilson, Christina
• Is Part Of: Biomedical signal processing and control. Volume 33(2017)
• Page Start: 178
• Page End: 188
• Publication Date: 2017-03
• Subjects: BoT base of tongue -- Cov coefficient of variation -- DSC dice similarity coefficient -- FCM/MFCM fuzzy c-means/modified fuzzy c-means -- IIH intensity inhomogeneity -- LSM level set method -- MHD modified Hausdorff distance -- MR/MRI magnetic resonance/magnetic resonance imaging -- MS mean shift -- Ncut normalised cut -- PCC Pearson correlation coefficient -- PLCSF pharyngeal and laryngeal cancer segmentation framework -- RO radiation oncologist -- ROI region of interest -- RTP radiotherapy treatment planning -- SUSAN smallest univalue segment assimilating nucleus -- T1 + Gd gadolinium enhanced T1-weighted; Head and neck cancer -- Automatic segmentation -- Magnetic resonance imaging (MRI) -- Fuzzy c-means clustering -- Fuzzy rules -- Level set method -- Radiotherapy; Signal processing -- Periodicals; Biomedical engineering -- Periodicals; Signal Processing, Computer-Assisted -- Periodicals; Image Processing, Computer-Assisted -- Periodicals; Biomedical Engineering -- Periodicals; 610.28
• DOI: 10.1016/j.bspc.2016.12.001
• Languages: English
• ISSNs: 1746-8094

Abstract

Highlights: A new fully automatic framework (PLCSF) for MRI pharynx and larynx cancer segmentation. Validation of the performance of the proposed framework with approach used in current clinical practice using well-established quality metrics. Applicability to MR images obtained from different MRI scanners with different imaging protocols. This study might provide a support tool for the clinicians in tumour delineation for radiotherapy treatment planning. Abstract: A novel and effective pharynx and larynx cancer segmentation framework (PLCSF) is presented for automatic base of tongue and larynx cancer segmentation from gadolinium-enhanced T1-weighted magnetic resonance images (MRI). The aim of the proposed PLCSF is to assist clinicians in radiotherapy treatment planning. The initial processing of MRI data in PLCSF includes cropping of region of interest; reduction of artefacts and detection of the throat region for the location prior. Further, modified fuzzy c-means clustering is developed to robustly separate candidate cancer pixels from other tissue types. In addition, region-based level set method is evolved to ensure spatial smoothness for the final segmentation boundary after noise removal using non-linear and morphological filtering. Validation study of PLCSF on 102 axial MRI slices demonstrate mean dice similarity coefficient of 0.79 and mean modified Hausdorff distance of 2.2 mm when compared with manual segmentations. Comparison of PLCSF with other algorithms validates the robustness of the PLCSF. Inter- and intra-variability calculations from manual segmentations suggest that PLCSF can help to reduce the human subjectivity.