Attention-Guided Deep Neural Network With Multi-Scale Feature Fusion for Liver Vessel Segmentation

Yan, Qingsen; Bo, Wang; Zhang, Wei; Luo, Chuan; Xu, Wei; Xu, Zhengqing; Zhang, Yanning; Shi, Qinfeng; Zhang, Liang; You, Zheng

doi:10.1109/jbhi.2020.3042069

Cited by 86 publications

(28 citation statements)

References 58 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For instance, [63] designed a novel neural network called "LVSNet" for liver vessel segmentation. Within the designed network, they have introduced a multi-scale feature fusion block that evenly divides the features into 4 filter groups; those filter groups are then connected in hierarchical residual learning; in this way, the number of scales in one block can be enhanced.…”

Section: Multi-channel Cnn For Cad Designmentioning

confidence: 99%

Multi-channel convolutional neural network architectures for thyroid cancer detection

et al. 2022

View full text Add to dashboard Cite

Early detection of malignant thyroid nodules leading to patient-specific treatments can reduce morbidity and mortality rates. Currently, thyroid specialists use medical images to diagnose then follow the treatment protocols, which have limitations due to unreliable human false-positive diagnostic rates. With the emergence of deep learning, advances in computer-aided diagnosis techniques have yielded promising earlier detection and prediction accuracy; however, clinicians’ adoption is far lacking. The present study adopts Xception neural network as the base structure and designs a practical framework, which comprises three adaptable multi-channel architectures that were positively evaluated using real-world data sets. The proposed architectures outperform existing statistical and machine learning techniques and reached a diagnostic accuracy rate of 0.989 with ultrasound images and 0.975 with computed tomography scans through the single input dual-channel architecture. Moreover, the patient-specific design was implemented for thyroid cancer detection and has obtained an accuracy of 0.95 for double inputs dual-channel architecture and 0.94 for four-channel architecture. Our evaluation suggests that ultrasound images and computed tomography (CT) scans yield comparable diagnostic results through computer-aided diagnosis applications. With ultrasound images obtained slightly higher results, CT, on the other hand, can achieve the patient-specific diagnostic design. Besides, with the proposed framework, clinicians can select the best fitting architecture when making decisions regarding a thyroid cancer diagnosis. The proposed framework also incorporates interpretable results as evidence, which potentially improves clinicians’ trust and hence their adoption of the computer-aided diagnosis techniques proposed with increased efficiency and accuracy.

show abstract

Section: Multi-channel Cnn For Cad Designmentioning

confidence: 99%

Multi-channel convolutional neural network architectures for thyroid cancer detection

et al. 2022

View full text Add to dashboard Cite

show abstract

“…Reference [25] presented a deep neural network employing an attention-guided concatenation (AGC) module and allowing an adaptive selection of context features from low-level features guided by high-level features to produce a detailed structure of the liver vessel. It is important that the segmentation extracts continuous liver vessels.…”

Section: Supervisedmentioning

confidence: 99%

“…As a result, 31 publications were selected and are shown in Table 1. [8] 2018 CTA (Section 5) Lu et al [9] 2017 MR Cheng et al [10] 2015 CTA Shang et al [11] 2010 CTA Guo et al [12] 2020 CT Tracking methods Lebre et al [13] 2019 CT and MR (Section 6) Zeng et al [14] 2018 CTA Sangsefidi et al [15] 2018 CT Yang et al [16] 2018 CT Zeng et al [17] 2017 CTA Yan et al [18] 2017 CTA Chi et al [19] 2010 CT Bauer et al [20] 2010 CT Alhonnoro et al [21] 2010 CT Esneault et al [22] 2009 CT Kaftan et al [23] 2009 CT Nazir et al [24] 2021 CT and CTA Machine learning methods Yan et al [25] 2020 CT (Section 7) Thomson et al [26] 2020 USG Xu et al [27] 2020 CT Keshwani et al [28] 2020 CT Kitrungrotsakul et al [29] 2019 CT Huang et al [30] 2018 CT Zhang et al [31] 2018 CT Mishra et al [32] 2018 USG Gocer et al [33] 2017 MR Ibragimov et al [34] 2017 CT Zeng et al [35] 2016 CT Wang et al [36] 2016 CT Oliveira et al [37] 2011 CT Bruyninckx et al [38] 2010 CT…”

Section: Introductionmentioning

confidence: 99%

Computational Methods for Liver Vessel Segmentation in Medical Imaging: A Review

Ciecholewski

Kassjański

2021

Sensors

View full text Add to dashboard Cite

The segmentation of liver blood vessels is of major importance as it is essential for formulating diagnoses, planning and delivering treatments, as well as evaluating the results of clinical procedures. Different imaging techniques are available for application in clinical practice, so the segmentation methods should take into account the characteristics of the imaging technique. Based on the literature, this review paper presents the most advanced and effective methods of liver vessel segmentation, as well as their performance according to the metrics used. This paper includes results available for four imaging methods, namely: computed tomography (CT), computed tomography angiography (CTA), magnetic resonance (MR), and ultrasonography (USG). The publicly available datasets used in research are also presented. This paper may help researchers gain better insight into the available materials and methods, making it easier to develop new, more effective solutions, as well as to improve existing approaches. This article analyzes in detail various segmentation methods, which can be divided into three groups: active contours, tracking-based, and machine learning techniques. For each group of methods, their theoretical and practical characteristics are discussed, and the pros and cons are highlighted. The most advanced and promising approaches are also suggested. However, we conclude that liver vasculature segmentation is still an open problem, because of the various deficiencies and constraints researchers need to address and try to eliminate from the solutions used.

show abstract

“…With the aid of this technique, the main structures, such as the microvasculature (8,9), bile ducts (10), and hepatic sinusoid (11,12) in the liver, can be clearly distinguished on CT images at micrometer-scale resolution. In addition, the development of advanced vascular segmentation technology provides a solid foundation for 3D vascular reconstruction and analysis (13)(14)(15)(16)(17). Thus, PCCT can reveal the 3D network structure of hepatic vessels in the whole liver, providing novel insights for the accurate evaluation of intrahepatic circulation in liver fibrosis.…”

Section: Introductionmentioning

confidence: 99%

Use of the volume-averaged Murray’s deviation method for the characterization of branching geometry in liver fibrosis: a preliminary study on vascular circulation

Lv¹,

Jian²,

Liu³

et al. 2022

Quant Imaging Med Surg

View full text Add to dashboard Cite

Attention-Guided Deep Neural Network With Multi-Scale Feature Fusion for Liver Vessel Segmentation

Cited by 86 publications

References 58 publications

Multi-channel convolutional neural network architectures for thyroid cancer detection

Multi-channel convolutional neural network architectures for thyroid cancer detection

Computational Methods for Liver Vessel Segmentation in Medical Imaging: A Review

Use of the volume-averaged Murray’s deviation method for the characterization of branching geometry in liver fibrosis: a preliminary study on vascular circulation

Contact Info

Product

Resources

About