Autor(es):
Torres, Helena R. ; Queirós, Sandro Filipe Monteiro ; Morais, Pedro André Gonçalves ; Oliveira, Bruno ; Fonseca, João Luís Gomes ; Mota, Paulo ; Lima, Estêvão Augusto Rodrigues de ; D'hooge, Jan ; Fonseca, Jaime C. ; Vilaça, João L.
Data: 2021
Identificador Persistente: https://hdl.handle.net/1822/71322
Origem: RepositóriUM - Universidade do Minho
Assunto(s): 3D ultrasound; B-spline Explicit Actives Surfaces; Feature detection; Kidney segmentation; 3-D ultrasound (US); B-spline explicit active surfaces (BEAS); Kidney; Image segmentation; Image edge detection; Three-dimensional displays; Deformable models; Ultrasonic imaging; Transforms
Descrição
Renal ultrasound imaging is the primary imaging modality for the assessment of the kidney’s condition and is essential for diagnosis, treatment and surgical intervention planning, and follow-up. In this regard, kidney delineation in three-dimensional ultrasound images represents a relevant and challenging task in clinical practice. In this paper, a novel framework is proposed to accurately segment the kidney in 3D ultrasound images. The proposed framework can be divided into two stages: 1) initialization of the segmentation method; and 2) kidney segmentation. Within the initialization stage, a phase-based feature detection method is used to detect edge points at kidney boundaries, from which the segmentation is automatically initialized. In the segmentation stage, the B-Spline Explicit Active Surface framework is adapted to obtain the final kidney contour. Here, a novel hybrid energy functional that combines localized region-based and edge-based terms is used during segmentation. For the edge term, a fast signed phase-based detection approach is applied. The proposed framework was validated in two distinct datasets: (1) 15 3D challenging poor-quality ultrasound images used for experimental development, parameters assessment, and evaluation; and (2) 42 3D ultrasound images (both healthy and pathologic kidneys) used to unbiasedly assess its accuracy. Overall, the proposed method achieved a Dice overlap around 81% and an average point-to-surface error of ~2.8 mm. These results demonstrate the potential of the proposed method for clinical usage.
