Research output: Contribution to journal › Article › peer-review
The Robust Vessel Segmentation and Centerline Extraction: One-Stage Deep Learning Approach. / Epifanov, Rostislav; Fedotova, Yana; Dyachuk, Savely et al.
In: Journal of Imaging, Vol. 11, No. 7, 209, 26.06.2025.Research output: Contribution to journal › Article › peer-review
}
TY - JOUR
T1 - The Robust Vessel Segmentation and Centerline Extraction: One-Stage Deep Learning Approach
AU - Epifanov, Rostislav
AU - Fedotova, Yana
AU - Dyachuk, Savely
AU - Gostev, Alexandr
AU - Karpenko, Andrei
AU - Mullyadzhanov, Rustam
N1 - The work of Epifanov R., Fedotova Y., and Gostev A. is supported by the Russian Science Foundation grant No. 23-75-10047. The work of Dyachuk S. and Mullyadzhanov R. was supported by the Mathematical Center in Akademgorodok under the Agreement No. 075-15-2025-349 with the Ministry of Science and Higher Education of the Russian Federation.
PY - 2025/6/26
Y1 - 2025/6/26
N2 - The accurate segmentation of blood vessels and centerline extraction are critical in vascular imaging applications, ranging from preoperative planning to hemodynamic modeling. This study introduces a novel one-stage method for simultaneous vessel segmentation and centerline extraction using a multitask neural network. We designed a hybrid architecture that integrates convolutional and graph layers, along with a task-specific loss function, to effectively capture the topological relationships between segmentation and centerline extraction, leveraging their complementary features. The proposed end-to-end framework directly predicts the centerline as a polyline with real-valued coordinates, thereby eliminating the need for post-processing steps commonly required by previous methods that infer centerlines either implicitly or without ensuring point connectivity. We evaluated our approach on a combined dataset of 142 computed tomography angiography images of the thoracic and abdominal regions from LIDC-IDRI and AMOS datasets. The results demonstrate that our method achieves superior centerline extraction performance (Surface Dice with threshold of 3 mm: 97.65% (Formula presented.) 2.07%) compared to state-of-the-art techniques, and attains the highest subvoxel resolution (Surface Dice with threshold of 1 mm: 72.52% (Formula presented.) 8.96%). In addition, we conducted a robustness analysis to evaluate the model stability under small rigid and deformable transformations of the input data, and benchmarked its robustness against the widely used VMTK toolkit.
AB - The accurate segmentation of blood vessels and centerline extraction are critical in vascular imaging applications, ranging from preoperative planning to hemodynamic modeling. This study introduces a novel one-stage method for simultaneous vessel segmentation and centerline extraction using a multitask neural network. We designed a hybrid architecture that integrates convolutional and graph layers, along with a task-specific loss function, to effectively capture the topological relationships between segmentation and centerline extraction, leveraging their complementary features. The proposed end-to-end framework directly predicts the centerline as a polyline with real-valued coordinates, thereby eliminating the need for post-processing steps commonly required by previous methods that infer centerlines either implicitly or without ensuring point connectivity. We evaluated our approach on a combined dataset of 142 computed tomography angiography images of the thoracic and abdominal regions from LIDC-IDRI and AMOS datasets. The results demonstrate that our method achieves superior centerline extraction performance (Surface Dice with threshold of 3 mm: 97.65% (Formula presented.) 2.07%) compared to state-of-the-art techniques, and attains the highest subvoxel resolution (Surface Dice with threshold of 1 mm: 72.52% (Formula presented.) 8.96%). In addition, we conducted a robustness analysis to evaluate the model stability under small rigid and deformable transformations of the input data, and benchmarked its robustness against the widely used VMTK toolkit.
KW - computed tomography angiography images
KW - multitask neural network
KW - one-stage centerline reconstruction
KW - vascular modeling toolkit
KW - vessel centerline extraction
KW - vessel segmentation
UR - https://www.mendeley.com/catalogue/add42b1f-dd22-38c7-8ece-22fca57544e0/
UR - https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=105011415375&origin=inward
U2 - 10.3390/jimaging11070209
DO - 10.3390/jimaging11070209
M3 - Article
C2 - 40710596
VL - 11
JO - Journal of Imaging
JF - Journal of Imaging
SN - 2313-433X
IS - 7
M1 - 209
ER -
ID: 68614281