SPEQTACLE: a Hilbertian norm generalization of the Fuzzy C-Means algorithm for tumor delineation in PET/CT images

Our results suggest that the functional tumor internal target volume defined as the union of functional gross tumor volumes delineated on motion-corrected binned images is the

Two observers performed manual delineation of the tumors on the CT images and delineated the tumor uptake on the corresponding PET images manually, using a fixed threshold

To segment tumor in PET images using multiple features, the spatial evidential clustering algorithm, e.g., SECM [8], is fully improved by integrating unsupervised distance

Sequentially, this image can be processed by a standard segmentation method, in our case a random- walker approach, for segmenting tumors based on their intrin- sic

The proposed framework con- sists of 3 steps: seed detection giving a set of seeds susceptible to belong to the tumor; seed selection allowing to delete aberrant seeds and to

Index Terms— PET/CT, breast cancer, lymph nodes, tumor seg- mentation, region-based features, feature analysis, Random Forest, U-Net,

Second, we generalize the enhanced fuzzy c-means algorithm [4] to color images (in fact any N -d image) by using a quantization of the color space.. Thus, we keep the ability of

Considering noise and imprecision modeling is of great concern for reliable PET image segmentation [9], in this pa- per, this critical issue is addressed via the theory of