IVUS Image Segmentation Using Superpixel-Wise Fuzzy Clustering and Level Set Evolution

Xia, Menghua; Yan, Wenjun; Huang, Yi; Guo, Yi; Zhou, Guohui; Wang, Yuanyuan

doi:10.3390/app9224967

Cited by 10 publications

(6 citation statements)

References 39 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Traditional methods, such as statistically driven approaches (Cardinal et al 2006, Unal et al 2008, graph-cut algorithms (Sun et al 2013), active contour models (Jodas et al 2017, Xia et al 2019, Wang et al 2019b, the holistic approach (Ciompi et al 2012), the extremal region selection (Faraji et al 2018), and iterative random walks (China et al 2019), were advantaged by their exemption of training annotations. However, it was cumbersome for traditional methods to identify ambiguous membrane boundaries in IVUS images as they exploited low-level image features with few semantic meanings, such as intensities and edges.…”

Section: Related Workmentioning

confidence: 99%

3D pyramidal densely connected network with cross-frame uncertainty guidance for intravascular ultrasound sequence segmentation

et al. 2023

View full text Add to dashboard Cite

Objective: Automatic extraction of external elastic membrane border (EEM) and lumen-intima border (LIB) in intravascular ultrasound (IVUS) sequences aids atherosclerosis diagnosis. Existing IVUS segmentation networks ignored longitudinal relations among sequential images and neglected that IVUS images of different vascular conditions vary largely in intricacy and informativeness. As a result, they suffered from performance degradation in complicated parts in IVUS sequences. Approach: In this paper, we develop a 3D Pyramidal Densely-connected Network (PDN) with Adaptive learning and post-Correction guided by a novel cross-frame uncertainty (CFU). The proposed method is named PDN-AC. Specifically, the PDN enables the longitudinal information exploitation and the effective perception of size-varied vessel regions in IVUS samples, by pyramidally connecting multi-scale 3D dilated convolutions. Additionally, the CFU enhances the robustness of the method to complicated pathology from the frame-level (f-CFU) and pixel-level (p-CFU) via exploiting cross-frame knowledge in IVUS sequences. The f-CFU weighs the complexity of IVUS frames and steers an adaptive sampling during the PDN training. The p-CFU visualizes uncertain pixels probably misclassified by the PDN and guides an active contour-based post-correction. Main results: Human and animal experiments were conducted on IVUS datasets acquired from atherosclerosis patients and pigs. Results showed that the f-CFU weighted adaptive sampling reduced the Hausdorff distance (HD) by 10.53%/7.69% in EEM/LIB detection. Improvements achieved by the p-CFU guided post-correction were 2.94%/5.56%. Significance: The PDN-AC attained mean Jaccard values of 0.90/0.87 and HD values of 0.33/0.34 mm in EEM/LIB detection, preferable to state-of-the-art IVUS segmentation methods.

show abstract

Section: Related Workmentioning

confidence: 99%

3D pyramidal densely connected network with cross-frame uncertainty guidance for intravascular ultrasound sequence segmentation

et al. 2023

View full text Add to dashboard Cite

show abstract

“…The segmentation of IVUS arterial borders (lumen and media-adventitia) was presented as a challenge at MICCAI 2011 conference. Since then, it is highly studied in literature with improvements made till now [ 12 – 21 ]. Recent studies depict that deep learning and artificial intelligence aid in improving the outcome of the medical imaging [ 23 – 25 ].…”

Section: Introductionmentioning

confidence: 99%

CADNet: an advanced architecture for automatic detection of coronary artery calcification and shadow border in intravascular ultrasound (IVUS) images

et al. 2023

View full text Add to dashboard Cite

Intravascular Ultrasound (IVUS) is a medical imaging modality widely used for the detection and treatment of coronary heart disease. The detection of vascular structures is extremely important for accurate treatment procedures. Manual detection of lumen and calcification is very time-consuming and requires technical experience. Ultrasound imaging suffers from the generation of artifacts which obstructs the clear delineation among structures. Considering, the need, to provide special attention to crucial areas, convolutional block attention modules (CBAM) is integrated into an encoder-decoder-based U-Net architecture along with Atrous Spatial Pyramid Pooling (ASPP) to detect vessel components: lumen, calcification and shadow borders. The attention modules prove effective in dealing with areas of special attention by assigning additional weights to crucial channels and preserving spatial features. The IVUS data of 12 patients undergoing the treatment is taken for this study. The novelty of the model design is such that it is able to detect the lumen area in the presence/absence of calcification and bifurcation artifacts too. Also, the model efficiently detects the calcification area even in case of severely complex lesions with shadows behind them. The main contribution of the work is that IVUS images of varying degrees of calcification till 360° are also considered in this work, which is usually neglected in previous studies. The experimental results of 1097 IVUS images of 12 patients resulted in meanIoU (0.7894 ± 0.011), Dice Coefficient (0.8763 ± 0.070), precision (0.8768 ± 0.069) and recall (0.8774 ± 0.071) of the proposed model CADNet which show the model’s effectiveness relative to other state-of-the art methods.

show abstract

“…Image segmentation [5] refers to the process of partitioning a digital image into smaller segments based on shared visual characteristics such as brightness, structure, texture, or color. This procedure simplifies the task of extracting features like structure, color, and shape from digital images [6] by grouping pixels into clusters based on the similarity of pixel intensity levels in the original image. Consequently, images are divided into clusters of pixels sharing common characteristics.…”

Section: Introductionmentioning

confidence: 99%

Hierarchical Spatial Feature-CNN Employing Grad-CAM for Enhanced Segmentation and Classification in Alzheimer's and Parkinson's Disease Diagnosis via MRI

Asokan,

Seshadri

2023

View full text Add to dashboard Cite

The precise diagnosis of neurodegenerative disorders, notably Parkinson's disease (PD) and Alzheimer's disease (AD), presents a formidable challenge, often necessitating several years for definitive determination. Given the increasing prevalence of PD and AD in aging populations of affluent nations, there is an urgent need for advanced technology and more precise diagnostic methodologies, particularly for early disease stages. In recent years, segmentation has seen a surge in application for processing Magnetic Resonance (MR) brain data, emerging as a valuable and indispensable tool. To capture comprehensive images of the brain for the diagnosis and classification of these neurodegenerative disorders, Magnetic Resonance Imaging (MRI) is employed. However, early detection and classification of PD and AD utilizing MRI datasets pose significant complexities. Owing to the inherent subjectivity of human observation, automated segmentation of MRI images has become a crucial asset for healthcare professionals. The primary focus of this study is to devise an effective image segmentation approach and classification techniques for the detection and categorization of AD and PD. Initially, Hierarchical Spatial Feature-CNN is employed to segment abnormal traces of PD and AD in MRIs. Subsequently, the Gradient-weighted Class Activation Mapping (Grad-CAM) method is used for disease classification. In Grad-CAM, each neuron is assigned prioritization weights based on their contribution to the classification of interest, using gradient information flowing into the final convolutional layer of the Convolutional Neural Network (CNN). Thus, the combination of Grad-CAM with CNN is applied to address the classification challenges inherent in PD and AD. Extensive experiments were conducted on the proposed model, resulting in a classification accuracy exceeding 98.17%. In addition, the proposed integration of Grad-CAM with CNN outperformed existing state-of-the-art approaches across all performance measures. This study underscores the potential of the proposed model in enhancing the diagnostic process for neurodegenerative disorders, offering promise for more efficient and accurate detection and classification of PD and AD.

show abstract

IVUS Image Segmentation Using Superpixel-Wise Fuzzy Clustering and Level Set Evolution

Cited by 10 publications

References 39 publications

3D pyramidal densely connected network with cross-frame uncertainty guidance for intravascular ultrasound sequence segmentation

3D pyramidal densely connected network with cross-frame uncertainty guidance for intravascular ultrasound sequence segmentation

CADNet: an advanced architecture for automatic detection of coronary artery calcification and shadow border in intravascular ultrasound (IVUS) images

Hierarchical Spatial Feature-CNN Employing Grad-CAM for Enhanced Segmentation and Classification in Alzheimer's and Parkinson's Disease Diagnosis via MRI

Contact Info

Product

Resources

About