Branching Characteristics Research Articles

Artificial Intelligence-Based Feature Analysis of Pulmonary Vein Morphology on Computed Tomography Scans and Risk of Atrial Fibrillation Recurrence After Catheter Ablation: A Multi-Site Study.

Atrial fibrillation (AF) recurrence (AF+) is common after catheter ablation. Pulmonary vein (PV) isolation is the cornerstone of AF ablation, but PV remodeling has been associated with the risk of AF+. We aimed to evaluate whether artificial intelligence-based morphological features of primary and secondary PV branches on computed tomography images are associated with AF+ post-ablation. Two artificial intelligence models were trained for the segmentation of computed tomography images, enabling the isolation of PV branches. Patients from Cleveland Clinic (n=135) and Vanderbilt University (n=594) were combined and divided into 2 sets for training and cross-validation (D1, n=218) and internal testing (D2, n=511). An independent validation set (D3, n=80) was obtained from University Hospitals of Cleveland. We extracted 48 fractal-based and 12 shape-based radiomic features from primary and secondary PV branches of patients with AF+ and without recurrence after catheter ablation of AF. To predict AF+, 3 Gradient Boosting classification models based on significant features from primary branch PV model (Mp), secondary branch PV model (Ms), and combined primary and secondary branch PV model (Mc) were built. Features relating to primary PVs were found to be associated with AF+. The Mp classifier achieved area under the curve values of 0.73, 0.71, and 0.70 across the 3 datasets. AF+ cases exhibited greater surface complexity in their primary PV area, as evidenced by higher fractal dimension values compared with AF nonrecurrence cases. The Ms classifier results revealed a weaker association with AF+, suggesting higher relevance to AF+ post-ablation from primary PV branch morphology. This largest multi-institutional study to date revealed associations between artificial intelligence-extracted morphological features of the primary PV branches with AF+ in 809 patients from 3 sites. Future work will focus on enhancing the predictive ability of the classifier by integrating clinical, structural, and morphological features, including left atrial appendage and left atrium-related characteristics.

Analysis of deformation evolution characteristics of the northern Lajishan fault before the Jishishan MS6.2 earthquake based on GNSS observations

A MS6.2 earthquake struck Jishishan Country in Gansu Province on 18 December 2023. The earthquake occurred on the eastern branch of the Northern Lajishan fault (NLJSF) which has never experienced such a violent seismic event in recorded history. In this study, multiple periods of GNSS campaign observation data and continuous observation data are collected to reveal the crustal deformation evolution characteristics of the eastern branch of the NLJSF before the earthquake, focusing on the dynamic behavior of the causative fault, regional strain parameter features, and the variation of GNSS baselines across the NLJSF. Velocity field data show that the differential movement on both sides of the fault zone before the earthquake accumulated energy for the occurrence of the earthquake. Cross-fault profile analyses indicate a noticeable weakening of horizontal shortening on both fault flanks, suggesting a nearing fault lockup. The baseline time series and strain parameter time series results both show that the study area is mainly NEE-trending compression deformation. In addition, before the earthquake, multiple baseline results and strain parameter time series deviated from the linear trend and gradually flattened, indicating a strain accumulation in the study area. The overall crustal deformation evolution shows a relatively high earthquake risk before the Lajishan earthquake.

Double red giant branch and red clump features of Galactic disc stellar populations with Gaia gspspec

The bimodality of the Milky Way disc, in the form of a thick short disc and a thinner more radially extended one, encrypts the complex internal evolution of our Galaxy and its interaction with the environment. To disentangle the different competing physical processes at play in Galactic evolution, a detailed chrono-chemical-kinematical and dynamical characterisation of the disc bimodality is necessary, including high number statistics. Here, we make use of an extremely precise sub-sample of the Gaia DR3 catalogue of stellar chemo-physical parameters. The selected database is composed of 408 800 stars with a median uncertainty of 10 K, 0.03, and 0.01 dex in and respectively. The stellar parameter precision allows us to break the age-metallicity degeneracy of disc stars. For the first time, the disc bimodality in the Kiel diagram of giant stars is observed, getting rid of interstellar absortion issues. This bimodality produces double red giant branch sequences and red clump features for mono-metallicity populations. A comparison with BaSTI isochrones allows us to demonstrate that an age gap is needed to explain the evolutionary sequence separation, in agreement with previous age-metallicity relations obtained using sub-giant stars. A bimodal distribution in the stellar mass- plane is observed at constant metallicity. Finally, a selection of stars with 0.03 dex shows that the most metal-rich population in the Milky Way disc presents an important proportion of stars with ages in the range of 5-13 Gyr, in agreement with previous literature findings. This old, extremely metal-rich population is possibly a mix of migrated stars from the internal Galactic regions, and old disc stars formed before the last major merger of the Milky Way. The Gaia Kiel diagrams of disc mono-abundance stellar populations reveal a complex, non-linear age-metallicity relation crafted by internal and external processes of Galactic evolution. Their detailed analysis opens new opportunities to reconstruct the puzzle of the Milky Way disc bimodality.

Fractal Geometry in Vertebrate Respiratory Systems: A Comparative Analysis of Branching Patterns Across Species

This study investigates the application of fractal geometry to the analysis of vertebrate respiratory systems, focusing on the branching patterns of the bronchial tree across diverse species. Using a combination of mathematical modeling and computational visualization, we explore how fractal dimensions and branching characteristics reflect evolutionary adaptations to various habitats and physiological demands. The research encompasses a range of vertebrates, including humans, horses, dolphins, chickens, iguanas, and bullfrogs, representing diverse taxonomic groups and ecological niches. Our analysis reveals a clear gradient of complexity in respiratory structures, correlating strongly with metabolic rates and environmental adaptations. Mammals, particularly those adapted for high‐performance activities like horses, demonstrate the most complex fractal patterns, indicating highly efficient gas exchange systems. In contrast, ectothermic species such as iguanas and bullfrogs exhibit simpler structures, reflecting their lower metabolic demands and alternative respiratory strategies. The study employs fractal dimension calculations, L‐system modeling, and lacunarity analysis to quantify and visualize these differences. Our findings suggest that fractal analysis provides valuable insights into the relationship between form and function in vertebrate respiratory systems, offering a novel perspective on comparative physiology and evolutionary biology. This research not only enhances our understanding of respiratory system evolution but also has potential applications in biomimetic design, medical diagnostics, and ecological physiology. It demonstrates the power of interdisciplinary approaches, combining advanced mathematics with biological principles to uncover fundamental patterns in nature.

IFGAN: Pre- to Post-Contrast Medical Image Synthesis Based on Interactive Frequency GAN

Medical images provide a visual representation of the internal structure of the human body. Injecting a contrast agent can increase the contrast of diseased tissues and assist in the accurate identification and assessment of conditions. Considering the adverse reactions and side effects caused by contrast agents, previous methods synthesized post-contrast images with pre-contrast images to bypass the administration process. However, existing methods pay inadequate attention to reasonable mapping of the lesion area and ignore gaps between post-contrast and real images in the frequency domain. Thus, in this paper, we propose an interactive frequency generative adversarial network (IFGAN) to solve the above problems and synthesize post-contrast images from pre-contrast images. We first designed an enhanced interaction module that is embedded in the generator to focus on the contrast enhancement region. Within it, target and reconstruction branch features interact to control the local contrast enhancement region feature and maintain the anatomical structure. We propose focal frequency loss to ensure the consistency of post-contrast and real images in the frequency domain. The experimental results demonstrated that IFGAN outperforms other sophisticated approaches in terms of preserving the accurate contrast enhancement of lesion regions and anatomical structures. Specifically, our method produces substantial improvements of 7.9% in structural similarity (SSIM), 36.3% in the peak signal-to-noise ratio (PSNR), and 8.5% in multiscale structural similarity (MSIM) compared with recent state-of-the-art methods.

GLIC: Underwater target detection based on global–local information coupling and multi-scale feature fusion

With the rapid development of object detection technology, underwater object detection has attracted widespread attention. Most of the existing underwater target detection methods are built based on convolutional neural networks (CNNs), which still have some limitations in the utilization of global information and cannot fully capture the key information in the images. To overcome the challenge of insufficient global–local feature extraction, an underwater target detector (namely GLIC) based on global–local information coupling and multi-scale feature fusion is proposed in this paper. Our GLIC consists of three main components: spatial pyramid pooling, global–local information coupling, and multi-scale feature fusion. Firstly, we embed spatial pyramid pooling, which improves the robustness of the model while retaining more spatial information. Secondly, we design the feature pyramid network with global–local information coupling. The global context of the transformer branch and the local features of the CNN branch interact with each other to enhance the feature representation. Finally, we construct a Multi-scale Feature Fusion (MFF) module that utilizes balanced semantic features integrated at the same depth for multi-scale feature fusion. In this way, each resolution in the pyramid receives equal information from others, thus balancing the information flow and making the features more discriminative. As demonstrated in comprehensive experiments, our GLIC, respectively, achieves 88.46%, 87.51%, and 74.94% mAP on the URPC2019, URPC2020, and UDD datasets.

Identification of New Cultivar and Different Provenances of Dendrocalamus brandisii (Poaceae: Bambusoideae) Using Simple Sequence Repeats Developed from the Whole Genome.

Dendrocalamus brandisii is a high-quality bamboo species that can be used for both bamboo shoots and wood. The nutritional components and flavors of D. brandisii vary from different geographical provenances. However, the unique biological characteristics of bamboo make morphological classification methods unsuitable for distinguishing them. Although the new cultivar 'Manxie No.1' has significant differences in the branch characteristics and the color of shoot sheaths compared to the D. brandisii, it still lacks precise genetic information at the molecular level. This study identified 231,789 microsatellite markers based on the whole genome of D. brandisii and analyzed their type composition and distribution on chromosomes in detail. Then, using TP-M13-SSR fluorescence-labeling technology, 34 pairs of polymorphic primers were screened to identify the new cultivar 'Manxie No.1' and 11 different geographical provenances of D. brandisii. We also constructed DNA fingerprinting profiles for them. At the same time, we mapped six polymorphic SSRs to the gene of D. brandisii, among which SSR673 was mapped to DhB10G011540, which is related to plant immunity. The specific markers selected in this study can rapidly identify the provenances and the new cultivar of D. brandisii and help explore candidate genes related to some important traits.

Analyzing the Anatomical Characteristics of the Facial Artery Branches: A Human Cadaveric Study.

The facial artery is the predominant branch of the external carotid artery supplying blood to the head and neck regions. Knowledge regarding the anatomical path and distribution of each facial artery's branches is essential. The aim of the study is to assess the levels, establish the reference positions of each branch of the facial artery to reliable landmarks in the face and neck regions, andevaluate the outer diameter and linear measurement. A prospective single-center cadaveric study was done on 60 hemifaces from 30 properly embalmed and formalin-fixed male cadavers. The statistical analysis was carried out using IBM SPSS Statistics for Windows, Version 22 (Released 2013; IBM Corp., Armonk, New York, United States). Descriptive statistics and inferential statistics have been applied. Nonparametric analysis and the tests of normality were used. The Wilcoxon signed-ranktest was used to compare the various parameters between the left and right sides of each branch of the facial artery. The average length and diameter of each branch of the facial artery measured in millimeters are as follows: superior labial, 27.3 ± 9.17 mm and 1.6 ± 0.2 mm; inferior labial, 31.5 ± 7.86 mm and 1.3 ± 0.6 mm; lateral nasal, 10.6 ± 4.55 mm and 0.74 ± 0.2 mm; angular,29.65 ± 7.93 mm and 0.8 ± 0.2 mm; and submental,35.9 ± 5.3 mm and 0.9 ± 0.4 mm. Statistically, there is no remarkable variation between the left and right sides except the submental artery, which shows a remarkable difference based on the origin between the left side and right side with a p-value of 0.039. This study's findings will enable surgeons to more efficiently plan and design reconstructive flaps based on the facial artery, pertaining to this specific ethnic population (Indianorigin), which is not available in published history untilnow.

Classification and extraction method of hidden dangers along railway lines based on semantic segmentation network

ABSTRACT Foreign objects invading high-speed railway lines can cause danger. One existing solution is to use remote sensing images to analyse the dangerous areas along the railway line, thereby providing a certain amount of investigation time. Considering the spatial and temporal resolution characteristics of existing remote sensing technologies in identifying floating objects and the reality of rapid land use changes, this paper identifies areas on the ground where floating objects may be generated by using semantic segmentation techniques oriented to remotely sensed imagery and provides early warnings to staff along the route. However, these regions that need to be analysed have different semantics and scales. To address these challenges, this paper proposes a Dual-branch Parallel Fusion Network (DPFNet) based on Transformer, aimed at enhancing multi-class semantic segmentation in remote sensing images. To leverage global contextual information, we introduce a Swin Transformer-based backbone network, which employs self-attention to capture a comprehensive scene context, facilitating better segmentation by considering the entire scene’s context. For multi-scale semantic features, we propose one approach that involves independent branching feature expression and a Multi-scale Feature Space Fusion Module (MFSFM). The former enriches multi-scale information, while the latter fuses features across different levels to capture diverse semantic features. Experimental results demonstrate that DPFNet can effectively identify the hidden danger area, and the fusion of multi-scale features makes the network more accurately identify and segment the risk area of different sizes, improving the segmentation accuracy and robustness, and is of great significance to the formation of the ‘prevention’ as the core of the railway safety operation.

Предпосылки создания отраслевых индустриальных парков в Тульско-Новомосковской агломерации

The tendencies of the industrial parks formation in the Russian agglomerations are considers in the article. The branch and territorial features of industrial parks foundation in the Tula region are determined on basis of the experience study of the special economic zone (SEZ) «Uzlovaya». The positive influence of the manufacturing SEZ on the branch clusters in agglomerations is confirmed.

Nearshore optical video object detector based on temporal branch and spatial feature enhancement

The computing power of nearshore and ship-borne devices is limited, posing significant challenges for accurately detecting objects in real-time on such devices. We propose a nearshore video object detector (NVID) to tackle these challenges. Considering the abundance of dynamic entities in the nearshore environment, we have developed you can look more (YCLM) to perceive the temporal characteristics of these objects. Furthermore, to improve the ability to detect objects of different sizes of networks, we designed parallel deformable attention (PDA) based on the spatial features of objects. More importantly, we developed fast re-parameterization convolution (FREConv) and faster conv (FConv). Building on these innovations, we proposed a fast re-parameterization network (FRENet) specifically tailored to produce low-parameter, multi-scale feature outputs. With end-to-end training, our pipeline outperforms other state-of-the-art (SOTA) methods on the nearshore objects (NearshoreObjects) dataset (90.4 average precision (AP) 50 (＋4.7), 9.3 parameters (Params) (−1.0M), 24.8 frames per second (FPS) (Jetson Nano) (＋0.6)). In addition, NVID also achieved excellent results in the on board (OnBoard) dataset (90.3 AP50 (＋2.8), 9.3 params (−1.0M), 26.5 FPS (Jetson Nano) (＋0.8)). The source code can be accessed at https://github.com/Yuanlin-Zhao/NVID.

Characteristic of fluoroether gum via non-linear rheological approaches

Abstract Large amplitude oscillatory shear (LAOS) tests were conducted on four types of fluoroether elastomer gum, namely VPL-85540, VPL-75545, VPL-65455, and VPL-55540 in order to study their molecular weight, chain structure, and branching degree from the perspective of rubber rheology. For the setting of shear rate in varied LAOS measurements, frequency and angular velocity controlling modes were attempted and compared. Non-linear rheological parameters showed that VPL-65455 possessed significant polymeric branching characteristics. VPL-55540 possessed the lowest molecular weight, and VPL-75545 was detected with higher long-pendant monomer content in its polymer chain. Rheological properties concluded from the analysis of two shear rate controlling modes of LAOS tests exhibited great consistency, which confirmed the reliability of LAOS tests in the characterization of fluoroether gum.

AMTrack:Transformer tracking via action information and mix-frequency features

Nowadays, Transformer-based visual tracking algorithms have been developing quickly because of the self-attention mechanism of Transformer, which has the capability to model global information. Although the self-attention mechanism in Transformer can effectively capture long-range dependencies in feature space, they only use flattened two-dimensional features and are unable to capture long-range temporal dependencies. Furthermore, since the self-attention in Transformer functions as a low-pass filter, it picks up on low-frequency features of the target while ignoring high-frequency features. This research suggests a Transformer tracker based on action information and mix-frequency features (AMTrack) to address these problems. Specifically, to address the lack of temporal remote dependencies, we introduce the target action aware module and the target action offset module. The target action aware module sets up several pathways to extract spatio-temporal, channel, and motion feature independently. In contrast, the target action offset module computes the target’s offset information by computing relative feature maps. Furthermore, in order to address the imbalance between high and low frequency features, we propose a mix-frequency attention and multi-frequency self-attention convolutional block. The mix-frequency attention uses high-frequency features within partitioned local windows as input for the high-frequency branch and average-pooled low-frequency features as the input for the low-frequency branch, calculating attention scores in both branches respectively. The multi-frequency self-attention convolutional block uses self-attention to capture low-frequency features and convolution to capture high-frequency features. Extensive experiments are carried out on eight challenging tracking datasets (e.g., OTB100 (Object Tracking Benchmark 100), NFS (Need For Speed), UAV123 (Unmanned Aerial Vehicles 123), TC128 (Temple Color 128), VOT2018 (Visual Object Tracking 2018), LaSOT (Large-scale Single Object Tracking), TrackingNet (Tracking Network), GOT-10k (Generic Object Tracking-10k)), and the experimental results show that our tracker achieves excellent tracking performance when compared with several state-of-the-art tracking algorithms. The experimental results show that on LaSOT, the success rates AUC (Area Under Curve), PNorm, and P reach 65.8%, 69.2%, and 68.0%, respectively, where the AUC value is 2.1% higher than the baseline algorithm TrDiMP (Transformer Discriminative Model Prediction). On other datasets, our tracker also achieves excellent tracking performance.

Thermal convection in a Brinkman–Darcy and Kelvin–Voigt fluid of order one with coupled stresses effect

AbstractOur paper thoroughly examines a model for the thermal convection of specific viscoelastic fluids within a porous medium characterized by Brinkman–Darcy properties. These fluids demonstrate first‐order Kelvin–Voigt behavior as they permeate the medium. This investigation considers the influence of couple stresses on the system. We calculate the critical Rayleigh number required for linear instability theory, focusing on both stationary and oscillatory types of convection. Additionally, we conduct the nonlinear stability theory. In the latter cases, it is demonstrated that oscillatory convection can occur. A detailed investigation is conducted into the characteristics of both stationary and oscillatory convection branches, including the points at which the transition occurs.

Superficial temporal artery branching pattern in single stage auricular reconstruction: A novel classification

IntroductionThe two most severe complications of single-stage, porous polyethene microtia reconstruction are flap necrosis/framework exposure and frontal nerve paralysis. To reduce these risks, require a temporoparietal fascia (TPF) flap that includes both the parietal and frontal branches of the superficial temporal artery (STA) while sparing the nerve. We propose a classification that helps minimize said complications. Material and methodsFifty-five TPF flaps of 54 microtia patients who underwent single-stage auricular reconstruction from May 2018 to July 2021 were studied. Flaps were harvested using endoscopic techniques. The parietal and frontal branch characteristics and measurements were obtained using a microscope/endoscope. ResultsThe frontal artery might have 1 to 4 branches. If they were close to Pitanguy's line (≤5 mm), there would be a high risk of nerve damage. Parietal (P) and frontal (F) artery diameters <0.5 mm were risk factors for partial flap necrosis. Based on this observation, we proposed 0.5 mm as the diameter threshold to determine whether an arterial branch is hypoplasia or sufficient.From this study, a new classification of STA branching pattern was proposed with five types: PF1 (23.6 %), PF2 (43.6 %), pF1 (3.6 %), pF2 (12.8 %), and Pf (16.4 %); where P/F indicates sufficient branches, p/f indicates absent or hypoplasia ones, and the number indicates single or multiple frontal artery branching. ConclusionThe risk of flap necrosis and frontal nerve damage is due to abnormalities of the frontal artery of the STA in the TPF flap. Understanding the anatomical classification with clear visualization during flap harvest ensures a successful outcome.

Dual cross-enhancement network for highly accurate dichotomous image segmentation

The existing image segmentation tasks mainly focus on segmenting objects with specific characteristics, such as salient, camouflaged, and meticulous objects, etc. However, the research of highly accurate Dichotomous Image Segmentation (DIS) combining these tasks has just started and still faces problems such as insufficient information interaction between layers and incomplete integration of high-level semantic information and low-level detailed features. In this paper, a new dual cross-enhancement network (DCENet) for highly accurate DIS is proposed, which mainly consists of two new modules: a cross-scaling guidance (CSG) module and a semantic cross-transplantation (SCT) module. Specifically, the CSG module adopts the adjacent-layer cross-scaling guidance method, which can efficiently interact with the multi-scale features of the adjacent layers extracted; the SCT module uses dual-branch features to complement each other. Moreover, in the way of transplantation, the high-level semantic information of the low-resolution branch is used to guide the low-level detail features of the high-resolution branch, and the features of different resolution branches are effectively fused. Finally, experimental results on the challenging DIS5K benchmark dataset show that the proposed network outperforms the 9 state-of-the-art (SOTA) networks in 5 widely used evaluation metrics. In addition, the ablation experiments also demonstrate the effectiveness of the cross-scaling guidance module and the semantic cross-transplantation module.

Semantic Segmentation of Point Cloud Scene via Multi-Scale Feature Aggregation and Adaptive Fusion

Point cloud semantic segmentation is a key step in 3D scene understanding and analysis. In recent years, deep learning–based point cloud semantic segmentation methods have received extensive attention from researchers. Multi-scale neighborhood feature learning methods are suitable for inhomogeneous density point clouds, but different scale branching feature learning increases the computational complexity and makes it difficult to accurately fuse different scale features to express local information. In this study, a point cloud semantic segmentation network based on RandLA-Net with multi-scale local feature aggregation and adaptive fusion is proposed. The designed structure can reduce computational complexity and accurately express local features. The mean intersection-over-union is improved by 1.1% on the SemanticKITTI data set with an inference speed of nine frames per second, while the mean intersection-over-union is improved by 0.9% on the S3DIS data set, compared with RandLA-Net. We also conduct ablation studies to validate the effectiveness of the proposed key structure.

L1-Tree: A novel algorithm for constructing 3D tree models and estimating branch architectural traits using terrestrial laser scanning data

Branch architecture provides crucial information for the understanding of plant trait variability and the adaptive strategies employed by trees in response to their environment. High-fidelity terrestrial laser scanning (TLS) data provide an accurate, efficient, and non-destructive means for constructing three-dimensional (3D) tree models and estimating architectural traits. However, the complex canopy structure of trees in natural forests and the presence of occlusion in TLS data pose significant challenges to achieving this goal. In this study, we present a novel algorithm, L1-Tree, for the construction of 3D tree models and the estimation of architectural traits from TLS data. This algorithm is grounded in the L1-Median algorithm and integrates a tree skeleton optimization procedure that considers the structural characteristics of tree branches. By comparing modeling results and manually derived branch traits for 24 trees of 24 species, we found that the L1-Tree algorithm achieved precision, recall, and F-score values of 0.94 for branch identification, coefficient of determination, root-mean-squared error, and normalized root-mean-squared error of 0.998, 0.068 m, and 0.3 % for branch length estimation, and a respective value of 0.958, 0.257 cm and 0.9 % for branch radius estimation. Additionally, the branch identification accuracy and accuracy in branch architectural trait estimation remained satisfactory across branch orders. Compared to established 3D tree model construction algorithms (e.g., TreeQSM), our L1-Tree algorithm demonstrated a superior capability in handling noisy environments and data gaps, making it a robust tool for TLS data-based tree architecture studies. Leaf-wood separation emerged as a crucial step influencing the performance of the L1-Tree algorithm. We observed significant drop in branch identification accuracy when using an automatic leaf-wood separation algorithm as input, highlighting the urgent need to develop effective leaf-wood separation algorithms to generate high-quality wood point clouds for tree architecture studies.

Order of magnitude increase in power from flow-induced vibrations

Natural hydro environments such as rivers and ocean currents provide abundant and essentially free sources of kinetic energy. Although significant research has been devoted to harnessing this power through traditional turbine systems, there are also studies on the novel use of flow-induced vibration of a bluff body. The large oscillations of an elliptical cylinder resulting from the newly-discovered hyper branch have raised interesting questions about its potential as a source of effective renewable energy. This study investigates the power extraction characteristics of the hyper branch over a range of flow velocities. A factor-of-ten increase in the maximum time-mean power coefficient relative to an established flow-induced vibration system — the vortex induced vibration for aquatic clean energy converter is demonstrated. This suggests the possibility of a new and effective strategy to utilise flow-induced vibrations for energy generation and could help drive the commercial implementation and uptake of oscillating energy converters.

The clinical and imaging data fusion model for single-period cerebral CTA collateral circulation assessment.

The Chinese population ranks among the highest globally in terms of stroke prevalence. In the clinical diagnostic process, radiologists utilize computed tomography angiography (CTA) images for diagnosis, enabling a precise assessment of collateral circulation in the brains of stroke patients. Recent studies frequently combine imaging and machine learning methods to develop computer-aided diagnostic algorithms. However, in studies concerning collateral circulation assessment, the extracted imaging features are primarily composed of manually designed statistical features, which exhibit significant limitations in their representational capacity. Accurately assessing collateral circulation using image features in brain CTA images still presents challenges. To tackle this issue, considering the scarcity of publicly accessible medical datasets, we combined clinical data with imaging data to establish a dataset named RadiomicsClinicCTA. Moreover, we devised two collateral circulation assessment models to exploit the synergistic potential of patients' clinical information and imaging data for a more accurate assessment of collateral circulation: data-level fusion and feature-level fusion. To remove redundant features from the dataset, we employed Levene's test and T-test methods for feature pre-screening. Subsequently, we performed feature dimensionality reduction using the LASSO and random forest algorithms and trained classification models with various machine learning algorithms on the data-level fusion dataset after feature engineering. Experimental results on the RadiomicsClinicCTA dataset demonstrate that the optimized data-level fusion model achieves an accuracy and AUC value exceeding 86% . Subsequently, we trained and assessed the performance of the feature-level fusion classification model. The results indicate the feature-level fusion classification model outperforms the optimized data-level fusion model. Comparative experiments show that the fused dataset better differentiates between good and bad side branch features relative to the pure radiomics dataset. Our study underscores the efficacy of integrating clinical and imaging data through fusion models, significantly enhancing the accuracy of collateral circulation assessment in stroke patients.