Tissue-specific and interpretable sub-segmentation of whole tumour burden on CT images by unsupervised fuzzy clustering

• Published in: Computers in biology and medicine Vol. 120; p. 103751
• Main Authors: Rundo, Leonardo, Beer, Lucian, Ursprung, Stephan, Martin-Gonzalez, Paula, Markowetz, Florian, Brenton, James D., Crispin-Ortuzar, Mireia, Sala, Evis, Woitek, Ramona
• Format: Journal Article
• Language: English
• Published: United States Elsevier Ltd 01.05.2020; Elsevier Limited; Elsevier
• Subjects: Algorithms; Biomarkers; Cancer; Cluster Analysis; Clustering; Computed tomography; Computer applications; Fuzzy Logic; Heterogeneity; Image processing; Image Processing, Computer-Assisted; Image segmentation; Internal Medicine; Kidney cancer; Medical imaging; Other; Ovarian cancer; Problem solving; Radiomics; Renal cell carcinoma; Soft tissues; Spatial data; Tissue-specific segmentation; Tomography, X-Ray Computed; Tumor Burden; Tumors; Unsupervised fuzzy clustering; Vector quantization; Unsupervised fuzzy clustering; Tissue-specific segmentation; Computed tomography; Renal cell carcinoma; Ovarian cancer; Radiomics
• ISSN: 0010-4825; 1879-0534; 1879-0534
• DOI: 10.1016/j.compbiomed.2020.103751

Cancer typically exhibits genotypic and phenotypic heterogeneity, which can have prognostic significance and influence therapy response. Computed Tomography (CT)-based radiomic approaches calculate quantitative features of tumour heterogeneity at a mesoscopic level, regardless of macroscopic areas of hypo-dense (i.e., cystic/necrotic), hyper-dense (i.e., calcified), or intermediately dense (i.e., soft tissue) portions. With the goal of achieving the automated sub-segmentation of these three tissue types, we present here a two-stage computational framework based on unsupervised Fuzzy C-Means Clustering (FCM) techniques. No existing approach has specifically addressed this task so far. Our tissue-specific image sub-segmentation was tested on ovarian cancer (pelvic/ovarian and omental disease) and renal cell carcinoma CT datasets using both overlap-based and distance-based metrics for evaluation. On all tested sub-segmentation tasks, our two-stage segmentation approach outperformed conventional segmentation techniques: fixed multi-thresholding, the Otsu method, and automatic cluster number selection heuristics for the K-means clustering algorithm. In addition, experiments showed that the integration of the spatial information into the FCM algorithm generally achieves more accurate segmentation results, whilst the kernelised FCM versions are not beneficial. The best spatial FCM configuration achieved average Dice similarity coefficient values starting from 81.94±4.76 and 83.43±3.81 for hyper-dense and hypo-dense components, respectively, for the investigated sub-segmentation tasks. The proposed intelligent framework could be readily integrated into clinical research environments and provides robust tools for future radiomic biomarker validation. [Display omitted] •Intelligent two-stage method for CT tissue-specific image segmentation of whole tumours.•Unsupervised fuzzy clustering framework for interpretable sub-segmentation.•Sub-segmentation results achieved on ovarian and renal cancer are accurate and reliable.•The proposed computational framework considerably outperforms conventional approaches.•Insights gained about intra-tumoural heterogeneity evaluation for radiomics.