Automatic segmentation and quantification of epicardial adipose tissue from coronary computed tomography angiography

He, Xiuxiu; Guo, Bang Jun; Lei, Yang; Wang, Tonghe; Fu, Yabo; Zhang, Long Jiang; Liu, Tian; Yang, Xiaofeng

doi:10.1088/1361-6560/ab8077

Cited by 25 publications

(44 citation statements)

References 45 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The segmentation of relatively small and complex structures with high inter-patient variability, such as ECF, has been far less successful. Recently, a few deep learning approaches to ECF segmentation have made progress on this problem ( He et al, 2020 ; Rodrigues et al, 2017 ; Commandeur et al, 2018 ). In this paper, we build upon the previous work by presenting a novel deep learning model for 3D segmentation of ECF, We propose a solution of automatic segmentation of ECF volume using a deep learning based approach that is evaluated on both non-contrast and contrast-enhanced CT datasets.…”

Section: Introductionmentioning

confidence: 99%

“… Schlemper et al (2019) demonstrate the performance of the attention U-Net on real-time fetal detection on 2D images and pancreas detection on 3D CT images. He et al (2020) proposed ECF segmentation from CE-CCTA using a modified 3D U-Net approach by adding attention gates (AG). AGs are commonly used in classification tasks ( Wu et al, 2020 ; Sharmin & Chakma, 2021 ; Fei et al, 2021 ; Kelvin et al, 2015 ) and have been applied for various medical image problems such as image classification ( Kelvin et al, 2015 ; Zhao et al, 2017 ), image segmentation ( Schlemper et al, 2019 ; Wu et al, 2020 ; Sharmin & Chakma, 2021 ; Fei et al, 2021 ; Kelvin et al, 2015 ; Zhao et al, 2017 ), and image captioning ( Zhao et al, 2017 ).…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

A 3D deep learning approach to epicardial fat segmentation in non-contrast and post-contrast cardiac CT images

Siriapisith

Kusakunniran

Haddawy

2021

PeerJ Computer Science

View full text Add to dashboard Cite

Epicardial fat (ECF) is localized fat surrounding the heart muscle or myocardium and enclosed by the thin-layer pericardium membrane. Segmenting the ECF is one of the most difficult medical image segmentation tasks. Since the epicardial fat is infiltrated into the groove between cardiac chambers and is contiguous with cardiac muscle, segmentation requires location and voxel intensity. Recently, deep learning methods have been effectively used to solve medical image segmentation problems in several domains with state-of-the-art performance. This paper presents a novel approach to 3D segmentation of ECF by integrating attention gates and deep supervision into the 3D U-Net deep learning architecture. The proposed method shows significant improvement of the segmentation performance, when compared with standard 3D U-Net. The experiments show excellent performance on non-contrast CT datasets with average Dice scores of 90.06%. Transfer learning from a pre-trained model of a non-contrast CT to contrast-enhanced CT dataset was also performed. The segmentation accuracy on the contrast-enhanced CT dataset achieved a Dice score of 88.16%.

show abstract

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

A 3D deep learning approach to epicardial fat segmentation in non-contrast and post-contrast cardiac CT images

Siriapisith

Kusakunniran

Haddawy

2021

PeerJ Computer Science

View full text Add to dashboard Cite

show abstract

“…Recent publications have described the use of machine and deep learning approaches to segment EAT on non-contrast CT calcium score images 13 , contrast CT images 14 , and high-resolution CT angiography (CTA) images 15,16 . Some studies assessed both epicardial and paracardial (external to the pericardium) fat depots 17 , while others distinguished between epicardial and paracardial fat [13][14][15] . Some authors have used methods without learning, including a recent method by De Albuquerque et.…”

Section: Introductionmentioning

confidence: 99%

“…al 16 , which used the floor of the log clustering algorithm and a set of morphological operations. Deep learning is popular using 2D slice 13 and 3D patch 15 data. Zhao et.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

DeepFat: Deep Learning Segmentation and Quantification Method for Assessing Epicardial Adipose Tissue in CT Calcium Score Scans

Hoori¹,

Hu²,

Lee³

et al. 2021

Preprint

View full text Add to dashboard Cite

Epicardial adipose tissue volume (EAT) has been linked to coronary artery disease and the risk of major adverse cardiac events. As manual quantification of EAT is time-consuming, requires specialized training, and is prone to human error, we developed a method (DeepFat) for the automatic assessment of EAT on non-contrast low-dose CT calcium score images using deep learning. We segmented the tissue enclosed by the pericardial sac on axial slices, using two innovations. First, we applied a HU‑attention-window with a window/level 350/40-HU to draw attention to the sac and reduce numerical errors. Second, we applied look ahead slab-of-slices with bisection (“bisect”) in which we split the heart into halves and sequenced the lower half from bottom-to-middle and the upper half from top-to-middle, thereby presenting an always increasing curvature of the sac to the network. EAT volume was obtained by thresholding voxels within the sac in the fat window (-190/-30-HU). Compared to manual segmentation, our algorithm gave excellent results with volume Dice=88.52%±3.3, slice Dice=87.70%±7.5, EAT error=0.5%±8.1, and R=98.52%(p<0.001). HU-attention-window and bisect improved Dice volume scores by 0.49% and 3.2% absolute, respectively. Extensive augmentation improved results. Variability between analysts was comparable to variability with DeepFat. Results compared favorably to those of previous publications.

show abstract

Increased adipose tissue is associated with improved overall survival, independent of skeletal muscle mass in non‐small cell lung cancer

Tao,

Fang,

Chen

et al. 2023

J cachexia sarcopenia muscle

View full text Add to dashboard Cite

BackgroundThe prognostic significance of non‐cancer‐related prognostic factors, such as body composition, has gained extensive attention in oncological research. Compared with sarcopenia, the prognostic significance of adipose tissue for overall survival in non‐small cell lung cancer remains unclear. We investigated the prognostic value of skeletal muscle and adipose tissue in patients with non‐small cell lung cancer.MethodsThis retrospective study included 4434 patients diagnosed with non‐small cell lung cancer between January 2014 and December 2016. Cross‐sectional areas of skeletal muscle and subcutaneous fat were measured, and the pericardial fat volume was automatically calculated. The skeletal muscle index and subcutaneous fat index were calculated as skeletal muscle area and subcutaneous fat area divided by height squared, respectively, and the pericardial fat index was calculated as pericardial fat volume divided by body surface area. The association between body composition and outcomes was evaluated using Cox proportional hazards model.ResultsA total of 750 patients (501 males [66.8%] and 249 females [33.2%]; mean age, 60.9 ± 9.8 years) were included. Sarcopenia (60.8% vs. 52.7%; P < 0.001), decreased subcutaneous fat index (51.4% vs. 25.2%; P < 0.001) and decreased pericardial fat index (55.4% vs. 16.5%; P < 0.001) were more commonly found in the deceased group than survived group. In multivariable Cox regression analysis, after adjusting for clinical variables, increased subcutaneous fat index (hazard ratio [HR] = 0.56, 95% confidence interval [CI]: 0.47–0.66, P < 0.001) and increased pericardial fat index (HR = 0.47, 95% CI: 0.40–0.56, P < 0.001) were associated with longer overall survival. For stage I–III patients, increased subcutaneous fat index (HR = 0.62, 95% CI: 0.48–0.76, P < 0.001) and increased pericardial fat index (HR = 0.43, 95% CI: 0.34–0.54, P < 0.001) were associated with better 5‐year overall survival rate. Similar results were recorded in stage IV patients. For patients with surgery, the prognostic value of increased subcutaneous fat index (HR = 0.60, 95% CI: 0.44–0.80, P = 0.001) and increased pericardial fat index (HR = 0.51, 95% CI: 0.38–0.68, P < 0.001) remained and predicted favourable overall survival. Non‐surgical patients showed similar results as surgical patients. No association was noted between sarcopenia and overall survival (P > 0.05).ConclusionsIncreased subcutaneous fat index and pericardial fat index were associated with a higher 5‐year overall survival rate, independent of sarcopenia, in non‐small cell lung cancer and may indicate a reduced risk of non‐cancer‐related death.

show abstract

Automatic segmentation and quantification of epicardial adipose tissue from coronary computed tomography angiography

Cited by 25 publications

References 45 publications

A 3D deep learning approach to epicardial fat segmentation in non-contrast and post-contrast cardiac CT images

A 3D deep learning approach to epicardial fat segmentation in non-contrast and post-contrast cardiac CT images

DeepFat: Deep Learning Segmentation and Quantification Method for Assessing Epicardial Adipose Tissue in CT Calcium Score Scans

Increased adipose tissue is associated with improved overall survival, independent of skeletal muscle mass in non‐small cell lung cancer

Contact Info

Product

Resources

About