Automatic Vessel Segmentation Research Articles

SFNet: Spatial and Frequency Domain Networks for Wide-Field OCT Angiography Retinal Vessel Segmentation.

Automatic segmentation of blood vessels in fundus images is important to assist ophthalmologists in diagnosis. However, automatic segmentation for Optical Coherence Tomography Angiography (OCTA) blood vessels has not been fully investigated due to various difficulties, such as vessel complexity. In addition, there are only a few publicly available OCTA image data sets for training and validating segmentation algorithms. To address these issues, we constructed a wild-field retinal OCTA segmentation data set, the Retinal Vessels Images in OCTA (REVIO) dataset. Second, we propose a new retinal vessel segmentation network based on spatial and frequency domain networks (SFNet). The proposed model are tested on three benchmark data sets including REVIO, ROSE and OCTA-500. The experimental results show superior performance on segmentation tasks compared to the representative methods.

FINE VESSEL SEGMENTATION WITH REFINEMENT GATE IN DEEP LEARNING ARCHITECTURES

Automated vessel segmentation is essential in diagnosing eye-related disorders and monitoring progressive retinal diseases. State-of-the-art methods have achieved excellent results in this field, but very few have considered the post-processing of feature maps. As a result, there is often a lack of small and fine vessels or discontinuities in segmented vessels. To address this issue, this study introduces a novel post-processing technique called the refinement gate, which works with a deep learning model during training. The refinement gate enhances contextual information to extract important features from feature maps better. The proposed technique is applied with U-net architecture and placed after every convolution block in the encoder path. Visual and statistical comparisons demonstrate the robustness of the proposed method using three publicly available datasets, namely: the DRIVE DB, the STARE DB, and CHASE_DB1 datasets, showing significant improvements to segment weak and tiny vessels. The reported results confirm the potential of the model to be used as a segmentation tool in the medical field. This study is the first to propose such a gating mechanism without additional trainable parameters or standalone networks as in other literature.

Automatic Segmentation of Mediastinal Lymph Nodes and Blood Vessels in Endobronchial Ultrasound (EBUS) Images Using Deep Learning.

Endobronchial ultrasound (EBUS) is used in the minimally invasive sampling of thoracic lymph nodes. In lung cancer staging, the accurate assessment of mediastinal structures is essential but challenged by variations in anatomy, image quality, and operator-dependent image interpretation. This study aimed to automatically detect and segment mediastinal lymph nodes and blood vessels employing a novel U-Net architecture-based approach in EBUS images. A total of 1161 EBUS images from 40 patients were annotated. For training and validation, 882 images from 30 patients and 145 images from 5 patients were utilized. A separate set of 134 images was reserved for testing. For lymph node and blood vessel segmentation, the mean ± standard deviation (SD) values of the Dice similarity coefficient were 0.71 ± 0.35 and 0.76 ± 0.38, those of the precision were 0.69 ± 0.36 and 0.82 ± 0.22, those of the sensitivity were 0.71 ± 0.38 and 0.80 ± 0.25, those of the specificity were 0.98 ± 0.02 and 0.99 ± 0.01, and those of the F1 score were 0.85 ± 0.16 and 0.81 ± 0.21, respectively. The average processing and segmentation run-time per image was 55 ± 1 ms (mean ± SD). The new U-Net architecture-based approach (EBUS-AI) could automatically detect and segment mediastinal lymph nodes and blood vessels in EBUS images. The method performed well and was feasible and fast, enabling real-time automatic labeling.

VESCL: an open source 2D vessel contouring library.

VESCL (pronounced 'vessel') is a novel vessel contouring library for computer-assisted 2D vessel contouring and segmentation. VESCL facilitates manual vessel segmentation in 2D medical images to generate gold-standard datasets for training, testing, and validating automatic vessel segmentation. VESCL is an open-source C++ library designed for easy integration into medical image processing systems. VESCL provides an intuitive interface for drawing variable-width parametric curves along vessels in 2D images. It includes highly optimized localized filtering to automatically fit drawn curves to the nearest vessel centerline and automatically determine the varying vessel width along each curve. To support a variety of segmentation paradigms, VESCL can export multiple segmentation representations including binary segmentations, occupancy maps, and distance fields. VESCL provides sub-pixel resolution for vessel centerlines and vessel widths. It is optimized to segment small vessels with single- or sub-pixel widths that are visible to the human eye but hard to segment automatically via conventional filters. When tested on neurovascular digital subtraction angiography (DSA), VESCL's intuitive hand-drawn input with automatic curve fitting increased the speed of fully manual segmentation by 22× over conventional methods and by 3× over the best publicly available computer-assisted manual segmentation method. Accuracy was shown to be within the range of inter-operator variability of gold standard manually segmented data from a publicly available dataset of neurovascular DSA images as measured using Dice scores. Preliminary tests showed similar improvements for segmenting DSA of coronary arteries and RGB images of retinal arteries. VESCL is an open-source C++ library for contouring vessels in 2D images which can be used to reduce the tedious, labor-intensive process of manually generating gold-standard segmentations for training, testing, and comparing automatic segmentation methods.

Patient-specific placental vessel segmentation with limited data

A major obstacle in applying machine learning for medical fields is the disparity between the data distribution of the training images and the data encountered in clinics. This phenomenon can be explained by inconsistent acquisition techniques and large variations across the patient spectrum. The result is poor translation of the trained models to the clinic, which limits their implementation in medical practice. Patient-specific trained networks could provide a potential solution. Although patient-specific approaches are usually infeasible because of the expenses associated with on-the-fly labeling, the use of generative adversarial networks enables this approach. This study proposes a patient-specific approach based on generative adversarial networks. In the presented training pipeline, the user trains a patient-specific segmentation network with extremely limited data which is supplemented with artificial samples generated by generative adversarial models. This approach is demonstrated in endoscopic video data captured during fetoscopic laser coagulation, a procedure used for treating twin-to-twin transfusion syndrome by ablating the placental blood vessels. Compared to a standard deep learning segmentation approach, the pipeline was able to achieve an intersection over union score of 0.60 using only 20 annotated images compared to 100 images using a standard approach. Furthermore, training with 20 annotated images without the use of the pipeline achieves an intersection over union score of 0.30, which, therefore, corresponds to a 100% increase in performance when incorporating the pipeline. A pipeline using GANs was used to generate artificial data which supplements the real data, this allows patient-specific training of a segmentation network. We show that artificial images generated using GANs significantly improve performance in vessel segmentation and that training patient-specific models can be a viable solution to bring automated vessel segmentation to the clinic.

Applications and implementation of generative artificial intelligence in cardiovascular imaging with a focus on ethical and legal considerations: what cardiovascular imagers need to know!

Abstract Machine learning (ML) and deep learning (DL) have potential applications in medicine. This overview explores the applications of AI in cardiovascular imaging, focusing on echocardiography, cardiac MRI (CMR), coronary CT angiography (CCTA), and CT morphology and function. AI, particularly DL approaches like convolutional neural networks, enhances standardization in echocardiography. In CMR, undersampling techniques and DL-based reconstruction methods, such as variational neural networks, improve efficiency and accuracy. ML in CCTA aids in diagnosing coronary artery disease, assessing stenosis severity, and analyzing plaque characteristics. Automatic segmentation of cardiac structures and vessels using AI is discussed, along with its potential in congenital heart disease diagnosis and 3D printing applications. Overall, AI integration in cardiovascular imaging shows promise for enhancing diagnostic accuracy and efficiency across modalities. The growing use of Generative Adversarial Networks in cardiovascular imaging brings substantial advancements but raises ethical concerns. The “black box” problem in DL models poses challenges for interpretability crucial in clinical practice. Evaluation metrics like ROC curves, image quality, clinical relevance, diversity, and quantitative performance assess GAI models. Automation bias highlights the risk of unquestioned reliance on AI outputs, demanding careful implementation and ethical frameworks. Ethical considerations involve transparency, respect for persons, beneficence, and justice, necessitating standardized evaluation protocols. Health disparities emerge if AI training lacks diversity, impacting diagnostic accuracy. AI language models, like GPT-4, face hallucination issues, posing ethical and legal challenges in healthcare. Regulatory frameworks and ethical governance are crucial for fair and accountable AI. Ongoing research and development are vital to evolving AI ethics.

Introduction to static vessel analysis

This talk will provide you with an overview of static retinal vessel analyses. Digital photography, automated vessel segmentation and analyses have revolutionized the field of retinal vessel analyses rapidly. Retinal vessel calibre changes allow for non‐invasive study of the microcirculation in vivo. The study of the retinal microcirculation has shown its potential in cardiovascular risk assessment, allowed for observation of changes in systemic health and capture changes in physical activity. This session is aimed for those new to retinal vessel analyses using fundus images.

Segmentation and quantification of venous structures and perivascular spaces in the thalamus in epilepsy using 7 Tesla MRI

Background and purposeEpilepsy is a complex neurological disorder affecting 50 million people worldwide. Persistent seizures may correlate with neural network, microstructural, and vascular changes within the thalamus. These thalamic changes may result from seizure activity or broader alterations involving neuronal vasculature and neuroinflammatory processes linked to glymphatic drainage. Improved resolution with Ultra-high field (UHF) magnetic resonance imaging (MRI) may be useful in identifying possible thalamic vascular abnormalities not otherwise detectable at lower field strengths. Materials and methodsWe outline a novel method which leverages UHF neuroimaging for detection and quantification of vessels and perivascular spaces (PVS) within the thalamus in 25 epilepsy patients and 16 controls, to uncover possible underlying imaging biomarkers of epilepsy. In our analysis, we optimize a MATLAB-based Frangi-based detection tool called Perivascular Space Semi-Automated Segmentation (PVSSAS) to detect thalamic PVSs, and additionally use a second Frangi-based segmentation tool method to automate detection of vascular structures in the thalamus. The resulting PVS and vessel masks were used to quantify differences in the number of vessels, PVS, overlaps, and number of PVS overlaps per vessel detected between groups, using a Hessian detection filter linked on an 18-connected network. ResultsWe found significantly more thalamic PVS (p = 0.0307) and a significant increase in the number of thalamic vessels (p = 0.038) in patients compared to controls. ConclusionHere we have developed a novel process which leverages UHF MRI to quantify and detect thalamic vessels and PVS that may provide a potential neuroimaging biomarker of epilepsy. Statement of SignificanceWe use 7T, ultra-high field MRI and employed an innovative combination of semi-automated perivascular space segmentation and automated vessel segmentation to visualize and quantify vessels and perivascular spaces (PVS) within the thalamus, a highly cited region of interest in epilepsy. To our knowledge, this is the first study to semi-automatically visualize and segment PVS in the thalamus and automatically detect thalamic vessels. We uncovered detectable differences in thalamic vasculature and PVS. These findings suggests that increases in the number of thalamic PVS and vessels may be a potential neuroimaging biomarker in epilepsy. This tool may be useful in the detection of subtle vascular changes in other regions of the brain related to epilepsy or can be employed in other neurological conditions.

Visualization and quantization of 3D retinal vessels in OCTA images.

Optical coherence tomography angiography (OCTA) has been increasingly used in the analysis of ophthalmic diseases in recent years. Automatic vessel segmentation in 2D OCTA projection images is commonly used in clinical practice. However, OCTA provides a 3D volume of the retinal blood vessels with rich spatial distribution information, and it is incomplete to segment retinal vessels only in 2D projection images. Here, considering that it is difficult to manually label 3D vessels, we introduce a 3D vessel segmentation and reconstruction method for OCTA images with only 2D vessel labels. We implemented 3D vessel segmentation in the OCTA volume using a specially trained 2D vessel segmentation model. The 3D vessel segmentation results are further used to calculate 3D vessel parameters and perform 3D reconstruction. The experimental results on the public dataset OCTA-500 demonstrate that 3D vessel parameters have higher sensitivity to vascular alteration than 2D vessel parameters, which makes it meaningful for clinical analysis. The 3D vessel reconstruction provides vascular visualization in different retinal layers that can be used to monitor the development of retinal diseases. Finally, we also illustrate the use of 3D reconstruction results to determine the relationship between the location of arteries and veins.

PE-Net: a parallel framework for 3D inferior mesenteric artery segmentation.

The structural morphology of mesenteric artery vessels is of significant importance for the diagnosis and treatment of colorectal cancer. However, developing automated vessel segmentation methods for this purpose remains challenging. Existing convolution-based segmentation methods have limitations in capturing long-range dependencies, while transformer-based models require large datasets, making them less suitable for tasks with limited training samples. Moreover, over-segmentation, mis-segmentation, and vessel discontinuity are common challenges in vessel segmentation tasks. To address these issues, we propose a parallel encoding architecture that combines transformers and convolutions to retain the advantages of both approaches. The model effectively learns position deviations and enhances robustness for small-scale datasets. Additionally, we introduce a vessel edge capture module to improve vessel continuity and topology. Extensive experimental results demonstrate the improved performance of our model, with Dice Similarity Coefficient and Average Hausdorff Distance scores of 81.64% and 7.7428, respectively.

PEA-Net: A progressive edge information aggregation network for vessel segmentation

Automatic vessel segmentation is a critical area of research in medical image analysis, as it can greatly assist doctors in accurately and efficiently diagnosing vascular diseases. However, accurately extracting the complete vessel structure from images remains a challenge due to issues such as uneven contrast and background noise. Existing methods primarily focus on segmenting individual pixels and often fail to consider vessel features and morphology. As a result, these methods often produce fragmented results and misidentify vessel-like background noise, leading to missing and outlier points in the overall segmentation. To address these issues, this paper proposes a novel approach called the progressive edge information aggregation network for vessel segmentation (PEA-Net). The proposed method consists of several key components. First, a dual-stream receptive field encoder (DRE) is introduced to preserve fine structural features and mitigate false positive predictions caused by background noise. This is achieved by combining vessel morphological features obtained from different receptive field sizes. Second, a progressive complementary fusion (PCF) module is designed to enhance fine vessel detection and improve connectivity. This module complements the decoding path by combining features from previous iterations and the DRE, incorporating nonsalient information. Additionally, segmentation-edge decoupling enhancement (SDE) modules are employed as decoders to integrate upsampling features with nonsalient information provided by the PCF. This integration enhances both edge and segmentation information. The features in the skip connection and decoding path are iteratively updated to progressively aggregate fine structure information, thereby optimizing segmentation results and reducing topological disconnections. Experimental results on multiple datasets demonstrate that the proposed PEA-Net model and strategy achieve optimal performance in both pixel-level and topology-level metrics.

3D pulmonary vessel segmentation based on improved residual attention u-net

Automatic segmentation of pulmonary vessels is a fundamental and essential task for the diagnosis of various pulmonary vessels diseases. The accuracy of segmentation is suffering from the complex vascular structure. In this paper, an Improved Residual Attention U-Net (IRAU-Net) aiming to segment pulmonary vessel in 3D is proposed. To extract more vessel structure information, the Squeeze and Excitation (SE) block is embedded in the down sampling stage. And in the up sampling stage, the global attention module (GAM) is used to capture target features in both high and low levels. These two stages are connected by Atrous Spatial Pyramid Pooling (ASPP) which can sample in various receptive fields with a low computational cost. By the evaluation experiment, the better performance of IRAU-Net on the segmentation of terminal vessel is indicated. It is expected to provide robust support for clinical diagnosis and treatment.

Multi-level spatial-temporal and attentional information deep fusion network for retinal vessel segmentation

Objective. Automatic segmentation of fundus vessels has the potential to enhance the judgment ability of intelligent disease diagnosis systems. Even though various methods have been proposed, it is still a demanding task to accurately segment the fundus vessels. The purpose of our study is to develop a robust and effective method to segment the vessels in human color retinal fundus images. Approach. We present a novel multi-level spatial-temporal and attentional information deep fusion network for the segmentation of retinal vessels, called MSAFNet, which enhances segmentation performance and robustness. Our method utilizes the multi-level spatial-temporal encoding module to obtain spatial-temporal information and the Self-Attention module to capture feature correlations in different levels of our network. Based on the encoder and decoder structure, we combine these features to get the final segmentation results. Main results. Through abundant experiments on four public datasets, our method achieves preferable performance compared with other SOTA retinal vessel segmentation methods. Our Accuracy and Area Under Curve achieve the highest scores of 96.96%, 96.57%, 96.48% and 98.78%, 98.54%, 98.27% on DRIVE, CHASE_DB1, and HRF datasets. Our Specificity achieves the highest score of 98.58% and 99.08% on DRIVE and STARE datasets. Significance. The experimental results demonstrate that our method has strong learning and representation capabilities and can accurately detect retinal blood vessels, thereby serving as a potential tool for assisting in diagnosis.

Automatic vessel segmentation in MR angiography

Abstract Funding Acknowledgements Type of funding sources: None. Background Vessel-to-background contrast (VBC), the ratio of signal intensities (SI) within a vessel segment compared to the surrounding tissue, is an essential objective parameter for assessing image quality in contrast-enhanced MR angiography (CE-MRA) of peripheral arteries. (1–3) Manual evaluation by the measurement of SIs within the vessel and in the surrounding area is labour-intensive and tedious. (2) An algorithm for automatic segmentation of the vessel lumen and background signal reduces the workload significantly. It has already been successfully applied to CE-MRA in peripheral arterial disease (PAD). (3) However, whether this algorithm is equal to a manual evaluation is not yet known. Purpose To compare the manual and automatic evaluation of VBC in CE-MRA in patients with PAD. Methods A patient with suspected peripheral arterial disease underwent first-pass subtraction MRA on a 3T scanner. Region-of-interests (ROIs) were drawn on predefined vessel segments (Figure 1), manually inside and outside the vessel lumen and additionally a comprehensive ROI with the vessel under investigation in the centre. The algorithm segmented the comprehensive ROI into the vessel and background by identifying a seed point for the vessel and then using a flood fill algorithm. The remaining vessel branches were removed in order to approximate the true background signal. SI of the manual and the automatic segmentation were compared to define the VBC using the formula SI lumen - SI background divided by the standard deviation (SD) of the background for both methods. Results 43 vessel segments were evaluated. An example is given in Figure 2 with A) showing an axial view of the right thigh with a comprehensive ROI with the superficial femoral artery in the centre (green), and manual outlining of the lumen (red) and background (blue). Background signal (B1) and vessel lumen (B2) were used to evaluate the manuall values. Automated segmentation with C1) shows the ROI with the femoral artery in the centre. C2) shows the full extent of the vessel as identified by the flood-fill algorithm, and C3) the resulting mask. Background signal was calculated from SIs remaining in C4 after subtraction of the local maximas shown in C5). The mean SI in the vessel lumen was 4171 (standard deviation, SD 954) for the manual evaluation and 5014 (SD 830) for the automatic evaluation (p = 0.001) and in the background 601 (SD 542) for the manual evaluation and 493 (SD 229) for the automatic evaluation ((p = 0.231). The VBC was 10.3 (SD 5.9) for manual evaluation and 12.2 (SD 5.6) for automatic evaluation (p = 0.139). Conclusion Automatic segmentation of vessel lumen and background signal allows assessment of the VBC in CE-MRA in patients with PAD. The mean SI of the vessel lumen tends to be overestimated about 20% by the automatic algorithm. Higher patient numbers are required to further clarify the applicability of this algorithm for image quality evaluation.

IterMiUnet: A lightweight architecture for automatic blood vessel segmentation

The automatic segmentation of blood vessels in fundus images can help analyze the condition of retinal vasculature, which is crucial for identifying various systemic diseases like hypertension, diabetes, etc. Despite the success of Deep Learning-based models in this segmentation task, most of them are heavily parametrized and thus have limited use in practical applications. This paper proposes IterMiUnet, a new lightweight convolution-based segmentation model that requires significantly fewer parameters and yet delivers performance similar to existing models. The model makes use of the excellent segmentation capabilities of Iternet architecture but overcomes its heavily parametrized nature by incorporating the encoder-decoder structure of MiUnet model within it. Thus, the new model reduces parameters without any compromise with the network's depth, which is necessary to learn abstract hierarchical concepts in deep models. This lightweight segmentation model speeds up training and inference time and is potentially helpful in the medical domain where data is scarce and, therefore, heavily parametrized models tend to overfit. The proposed model was evaluated on three publicly available datasets: DRIVE, STARE, and CHASE-DB1. Further cross-training and inter-rater variability evaluations have also been performed. The proposed model has a lot of potential to be utilized as a tool for the early diagnosis of many diseases.

A lightweight network guided with differential matched filtering for retinal vessel segmentation

The geometric morphology of retinal vessels reflects the state of cardiovascular health, and fundus images are important reference materials for ophthalmologists. Great progress has been made in automated vessel segmentation, but few studies have focused on thin vessel breakage and false-positives in areas with lesions or low contrast. In this work, we propose a new network, differential matched filtering guided attention UNet (DMF-AU), to address these issues, incorporating a differential matched filtering layer, feature anisotropic attention, and a multiscale consistency constrained backbone to perform thin vessel segmentation. The differential matched filtering is used for the early identification of locally linear vessels, and the resulting rough vessel map guides the backbone to learn vascular details. Feature anisotropic attention reinforces the vessel features of spatial linearity at each stage of the model. Multiscale constraints reduce the loss of vessel information while pooling within large receptive fields. In tests on multiple classical datasets, the proposed model performed well compared with other algorithms on several specially designed criteria for vessel segmentation. DMF-AU is a high-performance, lightweight vessel segmentation model. The source code is at https://github.com/tyb311/DMF-AU.

LMSA‐Net: A lightweight multi‐scale aware network for retinal vessel segmentation

AbstractRetinal vessel segmentation is an essential part of ocular disease diagnosis. However, due to complex vascular structure, large‐scale variations of retinal vessels, as well as inefficiency of vessel segmentation speed, accurate and fast automatic vessel segmentation for retinal images is still technically challenging. To tackle these issues, we present a lightweight multi‐scale‐aware network (LMSA‐Net) for retinal vessel segmentation. The network leverages the encoder‐decoder structure that was used in U‐Net. In the encoder, we propose a ghosted sandglass residual (GSR) block, aiming at greatly reducing the parameters and computational cost while obtaining richer semantic information. After that, a multi‐scale feature‐aware aggregation (MFA) module is designed to perceive multi‐scale semantic information for effective information extraction. Then, a global adaptive upsampling (GAU) module is proposed to guide the effective fusion of high‐ and low‐level semantic information in the decoder. Experiments are conducted on three public datasets, including DRIVE, CHASE_DB1, and STARE. The experimental results indicate the effectiveness of the LMSA‐Net, which can achieve better segmentation performance than other state‐of‐the‐art methods.

End-to-End Automatic Classification of Retinal Vessel Based on Generative Adversarial Networks with Improved U-Net

The retinal vessels in the human body are the only ones that can be observed directly by non-invasive imaging techniques. Retinal vessel morphology and structure are the important objects of concern for physicians in the early diagnosis and treatment of related diseases. The classification of retinal vessels has important guiding significance in the basic stage of diagnostic treatment. This paper proposes a novel method based on generative adversarial networks with improved U-Net, which can achieve synchronous automatic segmentation and classification of blood vessels by an end-to-end network. The proposed method avoids the dependency of the segmentation results in the multiple classification tasks. Moreover, the proposed method builds on an accurate classification of arteries and veins while also classifying arteriovenous crossings. The validity of the proposed method is evaluated on the RITE dataset: the accuracy of image comprehensive classification reaches 96.87%. The sensitivity and specificity of arteriovenous classification reach 91.78% and 97.25%. The results verify the effectiveness of the proposed method and show the competitive classification performance.

Retinal image blood vessel classification using hybrid deep learning in cataract diseased fundus images

With recent advanced technologies, various automated diagnosis tools were developed to prevent retinal diseases. The automatic segmentation of blood vessels can help detect various retinal diseases and also assist in reducing doctors’ workload. In existing, numerous techniques have been established to segment RBV automatically. But, they failed to provide better accuracy because of higher computational complexity and lower efficiency. Thus, the proposed work introduces an effective hybrid deep learning technique for RBV segmentation and classification of vessels. Initially, the retinal images are pre-processed to enhance the image quality by performing two steps such as image cropping and colour channel conversion. From the pre-processed images, the most important regions are segmented through a new Enhanced Fuzzy C-Means (EFCM) clustering scheme. During the segmentation process, the retinal images are clustered according to the thickness of the blood vessels, which helps minimize the computational complexity. After segmentation, a hybrid deep learning technique like DenseNet and ShuffleNet is introduced to perform feature extraction and classification. The experimental setup is done by the Python platform using the databases like DRIVE (Digital Retinal Images for Vessel Extraction), STARE (STructured Analysis of the Retina), and HRF (High Resolution fundus). Using the DRIVE dataset, the proposed model achieves an accuracy of 99%, the attained accuracy value of the STARE dataset is 98%, and the HRF dataset obtains an accuracy of 98%. The result analysis proves that the proposed hybrid deep learning technique is more efficient than the state-of-the-art techniques.

High-throughput 3DRA segmentation of brain vasculature and aneurysms using deep learning

Background and Objectives:Automatic segmentation of the cerebral vasculature and aneurysms facilitates incidental detection of aneurysms. The assessment of aneurysm rupture risk assists with pre-operative treatment planning and enables in-silico investigation of cerebral hemodynamics within and in the vicinity of aneurysms. However, ensuring precise and robust segmentation of cerebral vessels and aneurysms in neuroimaging modalities such as three-dimensional rotational angiography (3DRA) is challenging. The vasculature constitutes a small proportion of the image volume, resulting in a large class imbalance (relative to surrounding brain tissue). Additionally, aneurysms and vessels have similar image/appearance characteristics, making it challenging to distinguish the aneurysm sac from the vessel lumen.Methods:We propose a novel multi-class convolutional neural network to tackle these challenges and facilitate the automatic segmentation of cerebral vessels and aneurysms in 3DRA images. The proposed model is trained and evaluated on an internal multi-center dataset and an external publicly available challenge dataset.Results:On the internal clinical dataset, our method consistently outperformed several state-of-the-art approaches for vessel and aneurysm segmentation, achieving an average Dice score of 0.81 (0.15 higher than nnUNet) and an average surface-to-surface error of 0.20 mm (less than the in-plane resolution (0.35 mm/pixel)) for aneurysm segmentation; and an average Dice score of 0.91 and average surface-to-surface error of 0.25 mm for vessel segmentation. In 223 cases of a clinical dataset, our method accurately segmented 190 aneurysm cases.Conclusions:The proposed approach can help address class imbalance problems and inter-class interference problems in multi-class segmentation. Besides, this method performs consistently on clinical datasets from four different sources and the generated results are qualified for hemodynamic simulation. Code available at https://github.com/cistib/vessel-aneurysm-segmentation.