High Accuracy Pulmonary Vessel Segmentation for Contrast and Non-contrast CT Images and Clinical Evaluation
Listen
Radio episode about this paper
Transcript
Introduction to the show: ident: Electrical Engineering. The day in electrical engineering: signals, systems and the hardware that carries them.
Rosa: I'm Rosa.
Dev: And I'm Dev, here with Taro.
Rosa: We're looking at the paper: "High Accuracy Pulmonary Vessel Segmentation for Contrast and Non-contrast CT Images and Clinical Evaluation".
Dev: Here's the gist of the paper. Accurate segmentation of pulmonary vessels plays a very critical role in diagnosing and assessing various lung diseases<ref:2503.16988#pg2> This study proposes a 3D image segmentation algorithm for automated pulmonary vessel segmentation from both contrast-enhanced and non-contrast CT images<ref:2503.16988#pg2> The network designed includes a Vessel Lumen Structure Optimization Module (VLSOM), which extracts the centerline (Cl) of vessels and adjusts the weights based on the positional information and adds a Cl-Dice Loss to supervise the stability of the vessels structure<ref:2503.16988#pg2> The model achieved Cl-DICE, Cl-Recall, and Recall values of 0.892, 0.861, 0.924 for CTPA data and 0.925, 0.903, 0.949 for NCCT data<ref:2503.16988#pg2> Data and Annotation The dataset consisted of two main sources: 112 cases from the publicly available LIDC (Lung Image Database Consortium) dataset and 315 cases from multiple clinical institutions with the approval of the institutional review boards<ref:2503.16988#pg4> The dataset included 213 cases of CTPA and 214 cases of NCCT cases, which facilitated the development of a versatile model capable of handling both contrast-enhanced and non-contrast imaging data<ref:2503.16988#pg5> Ground truth (GT) annotations were performed by two radiologists with more than ten years of working experience to ensure high-quality supervision during model training<ref:2503.16988#pg6> The annotation standard required labeling vessels as small as 1 mm in diameter, ensuring that even fine vascular structures were captured<ref:2503.16988#pg7> A data annotation scheme was designed to simultaneously obtain high-precision GT results for both CTPA and NCCT, enabling the trained segmentation model to support automatic segmentation for both types of data<ref:2503.16988#pg8> Model Architecture and Loss Function The backbone of the model is based on the 3D U-Net architecture, which has demonstrated strong performance in medical image segmentation tasks due to its encoder-decoder structure with skip connections<ref:2503.16988#pg5> The Vessel Lumen Structure Optimization Module (VLSOM) enhances the traditional 3D U-Net architecture by extracting vessel centerlines (Cl) and applying weight-adjusted supervision for vessel lumens based on their proximity to the centerlines<ref:2503.16988#pg3> The loss function design is defined as Loss = CE Loss + DICE Loss + 0.5 CLDICE Loss, where the weights are specified as Weight CE Loss = (wclass = 3, wcl = 15), Weight DICE Loss = (wclass = 3, wcl = 15, in lung mask), and Weight CL DICE Loss = (wclass = 3, wcl = 15, in lung mask)<ref:2503.16988#pg10> Performance Comparison Compared with the State of the Art (SOTA) method nnUNet-v2, our model demonstrates superior performance in pulmonary vessel segmentation for both CTPA and NCCT data<ref:2503.16988#pg5> The results show that our experimental model achieved Cl-DICE, Cl-Recall and Recall values of 0.892, 0.861, 0.924 for CTPA and 0.925, 0.903, 0.949 for NCCT data<ref:2503.16988#pg2> These metrics indicate that the model can segment more complete and intricate vascular tree branches compared to nnUNet-v2<ref:2503.16988#pg2> Furthermore, the method can segment more complete pulmonary vessel segmentation results when compared with public datasets like Parse2022 and HiPaS model<ref:2503.16988#pg8> Clinical Validation Four objective indicators were used to conduct the clinical visual evaluation: accuracy and robustness, branch abundance, assistance for diagnosis, and vascular continuity<ref:2503.16988#pg8> The average score for accuracy and robustness was 4.26 while the full score is 5<ref:2503.16988#pg8> The consistency of clinical evaluation by two radiologists showed good consistency across the four indicators, with a Kappa value of 0.916, 0.744, 0.674 and 0.796 respectively<ref:2503.16988#pg9> Future Directions The remaining challenges for pulmonary vessel segmentation model include insufficient segmentation accuracy for extreme small vessels (those with diameters less than 0.8 mm) and partial errors in artery-vein classification, particularly in complex or low-contrast regions<ref:2503.16988#pg10> Future improvements could focus on incorporating advanced techniques such as multi-scale feature extraction or attention mechanisms to better detect small vessels<ref:2503.16988#pg10> Additionally, expanding the training dataset and integrating anatomical priors could help reduce classification errors and improve robustness<ref:2503.16988#pg10> Further clinical studies should research the clinical values for the quantitative parameters derived from the segmented vessels such as vessel length, vessel volume, and tortuosity<ref:2503.16988#pg10> Pruning redundant reasoning tokens The study introduced a 3D UNet segmentation network with VLSOM module to enhance pulmonary vessel segmentation performance<ref:2503.16988#pg2> By leveraging combined learning from 213 CTPA and 214 NCCT data, the proposed method generated a single model which achieved high-accuracy and high-integrity automatic vascular segmentation on both 27 CTPA and 29 non-contrast CT images with Cl-Recall, ClDICE and Recall values of 0.879, 0.909, 0.934 for CTPA and 0.928, 0.936, 0.955 for NCCT respectively<ref:2503.16988#pg2> Furthermore, we proposed four visual assessment metrics for subjective scoring in clinical evaluation and the model demonstrated good performance on 21 independent external evaluation CT images from 21 patients with six different lung diseases (PV, PE, ILD, COPD, SLE and GGO), indicating potential for clinical pulmonary vessel analysis and disease diagnosis<ref:2503.16988#pg2> How the training signal is built The Vessel Lumen Structure Optimization Module (VLSOM) extracts the centerlines of the ground truth (GT) vessels and incorporates them as additional supervisory signals during training<ref:2503.16988#pg3> The three main modules within VLSOM include: 1.
Rosa: Vessel Centerline Extraction Outline, which processes GT vascular masks to extract centerlines as a structural cue<ref:2503.16988#pg5> 2.
Dev: Weighted Supervision, which applies different weights to the vessel regions based on their proximity to the centerlines<ref:2503.16988#pg5> 3.
Rosa: Loss function design, which incorporates CL-DICE Loss based on the vessel contour structure<ref:2503.16988#pg10> Where the gains come from The gains in performance stem from the structural cues provided by the VLSOM, specifically through extracting centerlines and applying weighted supervision<ref:2503.16988#pg3> The post-processing algorithm further refines results by performing Connected Component Analysis, removing outliers, selecting largest components, and iteratively refining connections between small disconnected components<ref:2503.16988#pg8> This iterative refinement process ensures that the overall vascular tree morphology is highly intact<ref:2503.16988#pg8> Pruning redundant reasoning tokens The study introduced a 3D UNet segmentation network with VLSOM module to enhance pulmonary vessel segmentation performance<ref:2503.16988#pg2> By leveraging combined learning from 213 CTPA and 214 NCCT data, the proposed method generated a single model which achieved high-accuracy and high-integrity automatic vascular segmentation on both 27 CTPA and 29 non-contrast CT images with Cl-Recall, ClDICE and Recall values of 0.879, 0.909, 0.934 for CTPA and 0.928, 0.936, 0.955 for NCCT data<ref:2503.16988#pg2> Furthermore, we proposed four visual assessment metrics for subjective scoring in clinical evaluation and the model demonstrated good performance on 21 independent external evaluation CT images from 21 patients with six different lung diseases (PV, PE, ILD, COPD, SLE and GGO), indicating potential for clinical pulmonary vessel analysis and disease diagnosis<ref:2503.16988#pg2> Where the gains come from The gains in performance stem from the structural cues provided by the VLSOM, specifically through extracting centerlines and applying weighted supervision<ref:2503.16988#pg3> The post-processing algorithm further refines results by performing Connected Component Analysis, removing outliers, selecting largest components, and iteratively refining connections between small disconnected components<ref:2503.16988#pg8> This iterative refinement process ensures that the overall vascular tree morphology is highly intact<ref:2503.
Paper summary: Rosa: So, to recap, this paper proposes a three dee image segmentation algorithm specifically built for automating pulmonary vessel segmentation using both CTPA and NCCT images > <ref:2503.16988#pg1,a 3D image segmentation algorithm>
Dev: The core idea is the Vessel Lumen Structure Optimization Module, or VLSOM, which uses extracted vessel centerlines to guide the segmentation process through weighted supervision and a specialized loss function called CL-DICE Loss >
Taro: They trained this model on a dataset that included both contrast-enhanced CTPA cases and non-contrast NCCT cases, with careful annotation done by two experienced radiologists to ensure high quality ground truth data >
Rosa: They achieved solid performance metrics across both modalities, hitting recall values up to zero point nine five for the NCCT data when compared against existing methods like nnUNet-v2 > <ref:2503.16988#pg1>
Dev: The dataset they used was quite balanced, with two hundred thirteen cases of CTPA and two hundred fourteen cases of NCCT, which allowed them to develop this versatile model that supports both imaging types > <ref:2503.16988#pg3,213 cases of CTPA and 214 cases of NCCT>
Taro: But the paper also lays out some limitations you have to consider. They flag that there's still a challenge with insufficient segmentation accuracy for very small vessels, those with diameters less than zero point eight millimeters >
Rosa: That means even if the model is good overall, it might struggle when trying to precisely capture those extremely fine vascular structures >
Dev: And another limitation they mentioned is partial errors in classifying arteries versus veins, especially in complex or low-contrast regions within the scans >
Conclusion: Rosa: So looking at the title, "High Accuracy Pulmonary Vessel Segmentation for Contrast and Non-contrast CT Images and Clinical Evaluation," it really captures the breadth of what they did here >
Dev: It means they successfully created a way to achieve high accuracy in segmenting pulmonary vessels regardless of whether the scan uses contrast or not, which is important for clinical work where you might face both scenarios >
Taro: The implication is that this unified approach could help speed up diagnosis and treatment planning because doctors would have a tool that works on more types of imaging data than before >
Rosa: It’s about moving beyond just focusing on one modality to create a system that is robust enough for real-world clinical use across different patient situations >
Dev: And the authors demonstrated this by showing their model performs well on both CTPA and NCCT data, which addresses the gap in research where NCCT segmentation has been less studied >
Ying Ming, Shaoze Luo, Longfei Zhao, Ruijie Zhao, Bing Li, Qiqi Xu, *Wei Song
Department of Radiology, Peking Union Medical College Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing 100730, China · Research and Development Center, Canon Medical Systems (China), No.3, Xinyuan South Road, Chaoyang District, Beijing 100027
eess.IV, cs.CV
Submitted: 2025-03-21
Updated: 2025-05-19
Comments: Visual clinical evaluation results were added in v2 Comparison with the latest techniques were added Authors were updated
Journal ref: Med Phys. 2026; 53:e70751
DOI: 10.1002/mp.70751
License: http://creativecommons.org/licenses/by/4.0/
Importance score: 92/100
The gist: Accurate segmentation of pulmonary vessels plays a very critical role in diagnosing and assessing various lung diseases<ref:2503.16988#pg2> This study proposes a 3D image segmentation algorithm for…
Key concepts
- 3D U-Net Architecture
- This is the core structure of the deep learning model used for segmentation. It's an encoder-decoder network with skip connections, which helps the model capture both broad context and fine details in complex medical images like CT scans. This design is highly effective for accurately delineating structures in 3D space.
- Vessel Lumen Structure Optimization Module (VLSOM)
- This specialized module enhances the standard 3D U-Net by focusing on vessel centerlines (Cl). It extracts these centerlines from ground truth data and uses them to guide the training process, applying weighted supervision based on how close a vessel segment is to its true centerline. This helps stabilize the segmentation of vessels.
- Cl-DICE Loss
- This is a specific loss function added to the model's training process. It supervises the stability and structure of the segmented vessels by incorporating information about their extracted centerlines (Cl). By minimizing this loss, the model learns to generate vessel segments that not only match the overall shape but also maintain a realistic and continuous vascular structure.
Terminology
Summary
Accurate segmentation of pulmonary vessels plays a very critical role in diagnosing and assessing various lung diseases<ref:2503. p. 0 Deep Learning Pulmonary Vessel Segmentation for ContrastEnhanced and Non-contrast CT Images and Clinical Evaluation
Ying Ming1, Shaoze Luo2, Longfei Zhao2, Ruijie Zhao1, Bing Li2, Qiqi Xu2, and Wei Song1*
1 Department of Radiology, Peking Union Medical College Hospital, Chinese Academy of Medical
Sciences and Peking Union Medical College, Beijing 100730, China
2 Research and Development Center, Canon Medical Systems (China), No. 3, Xinyuan South Road,
Chaoyang District, Beijing 100027, China
ABSTRACT
Accurate segmentation of pulmonary vessels plays a very critical role in
diagnosing and assessing various lung diseases. However, the segmentation precision of these methods
is insufficient, and support for NCCT (Non-Contrast Computed Tomography) types
of data is also a requirement in some clinical scenarios. In this study, we propose a
3D image segmentation algorithm for automated pulmonary vessel segmentation
from both contrast-enhanced and non-contrast CT images. In this study, we propose a
3D image segmentation algorithm for automated pulmonary vessel segmentation
from both contrast-enhanced and non-contrast CT images. In the network, we
designed a Vessel Lumen Structure Optimization Module (VLSOM), which extracts
the centerline (Cl) of vessels and adjusts the weights based on the positional
information and adds a Cl-Dice Loss to supervise the stability of the vessels structure. In the network, we
designed a Vessel Lumen Structure Optimization Module (VLSOM), which extracts
the centerline (Cl) of vessels and adjusts the weights based on the positional
information and adds a Cl-Dice Loss to supervise the stability of the vessels structure. We used 427 sets of high-precision annotated CT data from multiple vendors and
countries to train the model and achieved Cl-DICE, Cl-Recall, and Recall values of
0. We used 427 sets of high-precision annotated CT data from multiple vendors and
countries to train the model and achieved Cl-DICE, Cl-Recall, and Recall values of
0. 924 for CTPA data and 0. 924 for CTPA data and 0. 949 for NCCT data. 924 for CTPA data and 0. 949 for NCCT data.Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Data and Annotation
The dataset consisted of two main sources: 112 cases from the publicly available LIDC (Lung Image Database Consortium) dataset and 315 cases from multiple clinical institutions with the approval of the institutional review boards<ref:2503.16988#pg4> The dataset included 213 cases of CTPA and 214 cases of NCCT cases, which facilitated the development of a versatile model capable of handling both contrast-enhanced and non-contrast imaging data<ref:2503. p. 2 To ensure comprehensive support for
both CTPA and NCCT data, the dataset included 213 cases of CTPA and 214 cases
of NCCT cases. This balanced distribution across modalities facilitated the
development of a versatile model capable of handling both contrast-enhanced and
non-contrast imaging data. 5 mm, the Z-direction resolution of NCCT
ranges from 0. Internal Data Annotation
Ground truth (GT) annotations were performed by two radiologists with more
than ten years of working experience to ensure high-quality supervision during
model training. in diameter, ensuring that even fine vascular structures were captured. This
meticulous annotation process was critical for achieving high precision in vessel
segmentation and preserving the completeness of the vascular network. In terms of ground truth
(GT) annotation, we have designed an annotation scheme that allows us to
simultaneously obtain high-precision GT results for both CTPA and NCCT. This
enables the trained segmentation model to support automatic segmentation for both
types of data.Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Model Architecture and Loss Function
The backbone of the model is based on the 3D U-Net architecture, which has demonstrated strong performance in medical image segmentation tasks due to its encoder-decoder structure with skip connections<ref:2503.16988#pg5> The Vessel Lumen Structure Optimization Module (VLSOM) enhances the traditional 3D U-Net architecture by extracting vessel centerlines (Cl) and applying weight-adjusted supervision for vessel lumens based on their proximity to the centerlines<ref:2503. p. 1 A key
innovation of our approach lies in the design of a Vessel Lumen Structure
Optimization Module (VLSOM). This module enhances the traditional 3D U-Net
architecture by extracting vessel centerlines (Cl) and applying weight-adjusted
supervision for vessel lumens based on their proximity to the centerlines. to guide the model in identifying and preserving the fine details of vessel lumen
structures; (2) Weighted Supervision: Based on clinical experience and preexperimental results, the VLSOM applies different weights to the vessel regions
based on their proximity to the centerlines. This ensures that the model pays more
attention to the centerline regions during training, thereby improving the precision
of vessel segmentation; (3) Loss function design:
𝐿𝑜𝑠𝑠 = 𝐶𝐸 𝐿𝑜𝑠𝑠 + 𝐷𝐼𝐶𝐸 𝐿𝑜𝑠𝑠 + 0. The notation "in lung mask" indicates that this weighting logic is only
effective within the lung region and no additional weighting is applied outside the
lung area. The above DICE Loss[16]
and Cl DICE Loss[17]
are both loss functions
based on the vessel contour structure.Where the paper says it
Where the paper says it
Where the paper says it
Performance Comparison
Compared with the State of the Art (SOTA) method nnUNet-v2, our model demonstrates superior performance in pulmonary vessel segmentation for both CTPA and NCCT data<ref:2503.16988#pg5> The results show that our experimental model achieved Cl-DICE, Cl-Recall and Recall values of 0. p. 0 In the network, we
designed a Vessel Lumen Structure Optimization Module (VLSOM), which extracts
the centerline (Cl) of vessels and adjusts the weights based on the positional
information and adds a Cl-Dice Loss to supervise the stability of the vessels structure. We used 427 sets of high-precision annotated CT data from multiple vendors and
countries to train the model and achieved Cl-DICE, Cl-Recall, and Recall values of
0. We used 427 sets of high-precision annotated CT data from multiple vendors and
countries to train the model and achieved Cl-DICE, Cl-Recall, and Recall values of
0. 924 for CTPA data and 0. We then used our self-developed annotation tool for doctors to
carefully correct and supplement annotations (S0), ultimately obtaining a more
complete and accurate annotation of the vascular branches in CTPA images (C1). To
support the model's application to NCCT data, we trained a CTPA-based
segmentation network using S1 and applied this network to NCCT data to obtain an
initial vascular tree (N0). Figure1 Iterative generation process of GT annotations of CTPA-NCCT
In summary, the dataset used in this study was carefully curated to include a
diverse range of cases from both public and private sources, with balanced
representation of CTPA and NCCT data. The high-quality annotations and
thoughtful data partitioning provided a strong foundation for developing a clinically
relevant and versatile pulmonary vessel segmentation model.Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Clinical Validation
Four objective indicators were used to conduct the clinical visual evaluation: accuracy and robustness, branch abundance, assistance for diagnosis, and vascular continuity<ref:2503.16988#pg8> The average score for accuracy and robustness was 4. p. 0 We finally conducted a clinical
visual assessment on an independent external test dataset. The average score for
accuracy and robustness, branch abundance, assistance for diagnosis and vascular
continuity are 4. The average score for
accuracy and robustness, branch abundance, assistance for diagnosis and vascular
continuity are 4. 83 respectively while the full score is 5. 75: excellent consistency). The entire evaluation process
for each case took approximately 5 minutes, including the evaluation of four
indicators by radiologists. The average score for
accuracy and robustness, branch abundance, assistance for diagnosis and vascular
continuity are 4. 83 respectively while the full score is 5.Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Future Directions
The remaining challenges for pulmonary vessel segmentation model include insufficient segmentation accuracy for extreme small vessels (those with diameters less than 0.8 mm) and partial errors in artery-vein classification, particularly in complex or low-contrast regions<ref:2503. p. 6 2 Notable deficiencies exist in segmentation accuracy and robustness, failing to
accurately delineate the anatomies of the pulmonary artery and vein. 3 Though generally accurate, segmentation contains appreciable errors or
misclassifications between arteries and veins. The specific
preprocessing steps are outlined below: (1) Lung Region Cropping: To focus on the
region of interest (ROI) and reduce unnecessary background information, the CT
images are first cropped based on the bounding box of their lung masks; (2)
Resampling: The cropped images are then resampled to a target isotropic resolution
of 0. 8mm, This step ensures uniformity in voxel size across all
datasets, addressing variations in the original imaging resolutions and facilitating
consistent feature extraction by the model. Deep Learning Pulmonary Vessel Segmentation for ContrastEnhanced and Non-contrast CT Images and Clinical Evaluation
Ying Ming1, Shaoze Luo2, Longfei Zhao2, Ruijie Zhao1, Bing Li2, Qiqi Xu2, and Wei Song1*
1 Department of Radiology, Peking Union Medical College Hospital, Chinese Academy of Medical
Sciences and Peking Union Medical College, Beijing 100730, China
2 Research and Development Center, Canon Medical Systems (China), No. 3, Xinyuan South Road,
Chaoyang District, Beijing 100027, China
ABSTRACT
Accurate segmentation of pulmonary vessels plays a very critical role in
diagnosing and assessing various lung diseases.Where the paper says it
Where the paper says it
Where the paper says it
Pruning redundant reasoning tokens
The study introduced a 3D UNet segmentation network with VLSOM module to enhance pulmonary vessel segmentation performance<ref:2503.16988#pg2> By leveraging combined learning from 213 CTPA and 214 NCCT data, the proposed method generated a single model which achieved high-accuracy and high-integrity automatic vascular segmentation on both 27 CTPA and 29 non-contrast CT images with Cl-Recall, ClDICE and Recall values of 0. p. 1 In terms of evaluation, we added Cl-DICE[17] and Cl-Recall[18] to
evaluate the results, emphasizing the quantitative analysis of the accuracy and
integrity of the vascular lumen structure. The resulting pulmonary vessel segmentation algorithm offers several
advantages: high precision, comprehensive vascular representation, and
compatibility with both CTPA and NCCT data. In terms of evaluation, we added Cl-DICE[17] and Cl-Recall[18] to
evaluate the results, emphasizing the quantitative analysis of the accuracy and
integrity of the vascular lumen structure. The resulting pulmonary vessel segmentation algorithm offers several
advantages: high precision, comprehensive vascular representation, and
compatibility with both CTPA and NCCT data. 3, Xinyuan South Road,
Chaoyang District, Beijing 100027, China
ABSTRACT
Accurate segmentation of pulmonary vessels plays a very critical role in
diagnosing and assessing various lung diseases. Currently, many automated
algorithms are primarily targeted at CTPA (Computed Tomography Pulmonary
Angiography) types of data. 3, Xinyuan South Road,
Chaoyang District, Beijing 100027, China
ABSTRACT
Accurate segmentation of pulmonary vessels plays a very critical role in
diagnosing and assessing various lung diseases. Currently, many automated
algorithms are primarily targeted at CTPA (Computed Tomography Pulmonary
Angiography) types of data. The average score for
accuracy and robustness, branch abundance, assistance for diagnosis and vascular
continuity are 4. 83 respectively while the full score is 5. To demonstrate the applicability of the proposed method to NCCT, we also collected
15 cases of patients who underwent NCCT, including 4 cases of interstitial lung
disease (ILD), 4 cases of chronic obstructive pulmonary disease (COPD), 3 cases of
systemic lupus erythematosus (SLE) and 4 patients with ground-glass opacity
(GGO). All CTPA and NCCT scans were acquired using three different CT systems:
Somatom Definition Flash and Somatom Force (Siemens Healthineers, Forchheim,
Germany); Philips IQon CT (Philips Medical Systems, Best, Netherlands); Aquilion
ONE GENESIS (Canon Medical Systems, Tokyo, Japan). To demonstrate the applicability of the proposed method to NCCT, we also collected
15 cases of patients who underwent NCCT, including 4 cases of interstitial lung
disease (ILD), 4 cases of chronic obstructive pulmonary disease (COPD), 3 cases of
systemic lupus erythematosus (SLE) and 4 patients with ground-glass opacity
(GGO). All CTPA and NCCT scans were acquired using three different CT systems:
Somatom Definition Flash and Somatom Force (Siemens Healthineers, Forchheim,
Germany); Philips IQon CT (Philips Medical Systems, Best, Netherlands); Aquilion
ONE GENESIS (Canon Medical Systems, Tokyo, Japan).Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
How the training signal is built
The Vessel Lumen Structure Optimization Module (VLSOM) extracts the centerlines of the ground truth (GT) vessels and incorporates them as additional supervisory signals during training<ref:2503.16988#pg3> The three main modules within VLSOM include:
-
Vessel Centerline Extraction Outline, which processes GT vascular masks to extract centerlines as a structural cue<ref:2503. p. 7
16988#pg5>Where the paper says it
The three main modules include: (1) Vessel Centerline Extraction Outline: The GT vascular masks are processed to extract their centerlines. These centerlines serve as a structural cue
-
Weighted Supervision, which applies different weights to the vessel regions based on their proximity to the centerlines<ref:2503. p. 8
16988#pg5>Where the paper says it
to guide the model in identifying and preserving the fine details of vessel lumen structures; (2) Weighted Supervision: Based on clinical experience and preexperimental results, the VLSOM applies different weights to the vessel regions based on their proximity to the centerlines. This ensures that the model pays more attention to the centerline regions during training, thereby improving the precision of vessel segmentation; (3) Loss function design: 𝐿𝑜𝑠𝑠 = 𝐶𝐸 𝐿𝑜𝑠𝑠 + 𝐷𝐼𝐶𝐸 𝐿𝑜𝑠𝑠 + 0.
-
Loss function design, which incorporates CL-DICE Loss based on the vessel contour structure<ref:2503. p. 8
16988#pg10>Where the paper says it
The notation "in lung mask" indicates that this weighting logic is only effective within the lung region and no additional weighting is applied outside the lung area. The above DICE Loss[16] and Cl DICE Loss[17] are both loss functions based on the vessel contour structure.
Where the gains come from
The gains in performance stem from the structural cues provided by the VLSOM, specifically through extracting centerlines and applying weighted supervision<ref:2503.16988#pg3> The post-processing algorithm further refines results by performing Connected Component Analysis, removing outliers, selecting largest components, and iteratively refining connections between small disconnected components<ref:2503. p. 8 3Post-Processing
After model inference on the cropped volume, we designed a post-processing
algorithm to remove outliers and adjust disconnected vessel branches. The steps of
the post-processing algorithm are as follows: (1) Connected Component Analysis:
Calculate the connected components for both the artery and vein domains. 3Post-Processing
After model inference on the cropped volume, we designed a post-processing
algorithm to remove outliers and adjust disconnected vessel branches. The steps of
the post-processing algorithm are as follows: (1) Connected Component Analysis:
Calculate the connected components for both the artery and vein domains. 5 Vascular continuity is well maintained, with almost no noticeable fragmentation. The overall vascular tree morphology is highly intact. 5 Vascular continuity is well maintained, with almost no noticeable fragmentation. The overall vascular tree morphology is highly intact.Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
The study introduced a 3D UNet segmentation network with VLSOM module to enhance pulmonary vessel segmentation performance<ref:2503.16988#pg2> By leveraging combined learning from 213 CTPA and 214 NCCT data, the proposed method generated a single model which achieved high-accuracy and high-integrity automatic vascular segmentation on both 27 CTPA and 29 non-contrast CT images with Cl-Recall, ClDICE and Recall values of 0.879, 0.909, 0.934 for CTPA and 0.928, 0.936, 0.955 for NCCT data<ref:2503. p. 0 Currently, many automated
algorithms are primarily targeted at CTPA (Computed Tomography Pulmonary
Angiography) types of data. However, the segmentation precision of these methods
is insufficient, and support for NCCT (Non-Contrast Computed Tomography) types
of data is also a requirement in some clinical scenarios.Where the paper says it
The gains in performance stem from the structural cues provided by the VLSOM, specifically through extracting centerlines and applying weighted supervision<ref:2503.16988#pg3> The post-processing algorithm further refines results by performing Connected Component Analysis, removing outliers, selecting largest components, and iteratively refining connections between small disconnected components<ref:2503.16988#pg8> This iterative refinement process ensures that the overall vascular tree morphology is highly intact<ref:2503.
Green — the paper says it, and the page is printed. Red — no passage carries it, so treat that sentence as unverified. The same check runs on your own text: check a summary against its source, $5.
Improvements for AI systems
-
Vessel Lumen Structure Optimization Module (VLSOM) integration allows for extraction of vessel centerlines (Cl) and adjustment of weights based on positional information to
supervise the stability of the vessels structure.
This enables high segmentation accuracy while preserving completeness by focusing attention on centerline regions during training, as described in Section 3.2. -
Unified model architecture supports both contrast-enhanced CTPA and non-contrast NCCT data by generating a single model trained on combined data, demonstrating
performance transfer from CTPA to NCCT
to extend applicability to NCCT data, as noted in the Results section and Abstract. -
The loss function design incorporates a novel term:
Cl DICE Loss,
which is used alongside standard CE Loss and DICE Loss, specifically targeting the integrity of the vascular lumen structure based on centerline metrics, improving segmentation beyond global statistical metrics. -
Post-processing algorithms are designed to handle structural inconsistencies by using Connected Component Analysis and iterative refinement steps to
remove outliers
andHandle Disconnected Artery Components
orHandle Disconnected Vein Components.
This ensures a more complete vascular tree morphology after model inference. -
The system can achieve superior performance compared to the State of the Art nnUNet-v2, achieving higher Cl-Recall and Cl-Dice values (e.g., 0.892 for CTPA), which indicates that
the model in this paper can segment more complete and intricate vascular tree branches.
-
The system provides objective clinical utility by quantifying results using four metrics—accuracy and robustness, branch abundance, assistance for diagnosis, and vascular continuity—which were validated by radiologists using a five-point scale. This allows clinicians to assess if the segmentation
can guide surgical planning and diagnosis with few requiring modifications.
-
Future improvements can focus on incorporating advanced techniques like multi-scale feature extraction or attention mechanisms to better detect small vessels with diameters less than 0.8 mm, addressing the current limitation of
insufficient segmentation accuracy for extreme small vessels.
Abstract
Accurate segmentation of pulmonary vessels plays a very critical role in diagnosing and assessing various lung diseases. Currently, many automated algorithms are primarily targeted at CTPA (Computed Tomography Pulmonary Angiography) types of data. However, the segmentation precision of these methods is insufficient, and support for NCCT (Non-Contrast Computed Tomography) types of data is also a requirement in some clinical scenarios. In this study, we propose a 3D image segmentation algorithm for automated pulmonary vessel segmentation from both contrast-enhanced and non-contrast CT images. In the network, we designed a Vessel Lumen Structure Optimization Module (VLSOM), which extracts the centerline (Cl) of vessels and adjusts the weights based on the positional information and adds a Cl-Dice Loss to supervise the stability of the vessels structure. We used 427 sets of high-precision annotated CT data from multiple vendors and countries to train the model and achieved Cl-DICE, Cl-Recall, and Recall values of 0.892, 0.861, 0.924 for CTPA data and 0.925, 0.903, 0.949 for NCCT data. This shows that our model has achieved good performance in both accuracy and completeness of pulmonary vessel segmentation. We finally conducted a clinical visual assessment on an independent external test dataset. The average score for accuracy and robustness, branch abundance, assistance for diagnosis and vascular continuity are 4.26, 4.17, 4.33, 3.83 respectively while the full score is 5. These results highlight the great potential of this method in clinical application.
Sources
Related papers
- HDNet: High-resolution Dual-domain Learning for Spectral Compressive Imaging
- Mask-guided Spectral-wise Transformer for Efficient Hyperspectral Image Reconstruction
- FDIR: Fidelity-Anchored Decompressed Image Restoration for Human and Machine Perception
- A Disease-Centric Vision-Language Foundation Model for Precision Oncology in Kidney Cancer
- CADS: A Comprehensive Anatomical Dataset and Segmentation for Whole-Body Anatomy in Computed Tomography
- Task-Aware Bitrate Thresholds for Codec-Robust Ultrasound Segmentation