A Cascaded Multi-Task Generative Framework for Detecting Aortic Dissection on 3-D Non-Contrast-Enhanced Computed Tomography

Xiong, Xiangyu; Ding, Yan; Sun, Chuanqi; Zhang, Zhuoneng; Guan, Xiuhong; Chen, Hao; Liu, Hongyan; Cheng, Zhangbo; Zhao, Lei; Ma, Xiaohai; Xie, Guoxi

doi:10.1109/jbhi.2022.3190293

Cited by 7 publications

(13 citation statements)

References 49 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The ACC, sensitivity, and specificity of the model in the internal testing cohort were 0.897, 0.862, and 0.923, respectively, and reached 0.730, 0.978, and 0.554, respectively, in the external testing cohort. Xiong et al [ 21 ] developed and evaluated a cascaded deep-learning framework that consisted of a 3D segmentation network and a synthesis network. A conditional generative adversarial network was used to map NCCT images to contrast-enhanced CT images of the aortic region non-linearly to assist physicians in diagnosis.…”

Section: Discussionmentioning

confidence: 99%

Diagnosis of Acute Aortic Syndromes on Non-Contrast CT Images with Radiomics-Based Machine Learning

Jin

Zhang

et al. 2023

Biology

View full text Add to dashboard Cite

We aimed to detect acute aortic syndromes (AAS) on non-contrast computed tomography (NCCT) images using a radiomics-based machine learning model. A total of 325 patients who underwent aortic CT angiography (CTA) were enrolled retrospectively from 2 medical centers in China to form the internal cohort (230 patients, 60 patients with AAS) and the external testing cohort (95 patients with AAS). The internal cohort was divided into the training cohort (n = 135), validation cohort (n = 49), and internal testing cohort (n = 46). The aortic mask was manually delineated on NCCT by a radiologist. Least Absolute Shrinkage and Selection Operator regression (LASSO) was used to filter out nine feature parameters; the Support Vector Machine (SVM) model showed the best performance. In the training and validation cohorts, the SVM model had an area under the curve (AUC) of 0.993 (95% CI, 0.965–1); accuracy (ACC), 0.946 (95% CI, 0.877–1); sensitivity, 0.9 (95% CI, 0.696–1); and specificity, 0.964 (95% CI, 0.903–1). In the internal testing cohort, the SVM model had an AUC of 0.997 (95% CI, 0.992–1); ACC, 0.957 (95% CI, 0.945–0.988); sensitivity, 0.889 (95% CI, 0.888–0.889); and specificity, 0.973 (95% CI, 0.959–1). In the external testing cohort, the ACC was 0.991 (95% CI, 0.937–1). This model can detect AAS on NCCT, reducing misdiagnosis and improving examinations and prognosis.

show abstract

Section: Discussionmentioning

confidence: 99%

Diagnosis of Acute Aortic Syndromes on Non-Contrast CT Images with Radiomics-Based Machine Learning

Jin

Zhang

et al. 2023

Biology

View full text Add to dashboard Cite

show abstract

“…Pixelto-pixel [80] achieved image style transfer by training a conditional GAN whose generator and discriminator were both based on the input images. Many 3D medical applications take advantage of the conditional generative model, including conditional synthesis [33,109,140,148,215,219,220], segmentation [27,158,158,226,231], denoising [155,202,203,205,236], detection [59,147,188,208], and registration [11,154,230,234,238].…”

Section: Unconditional and Conditional Generative Modelsmentioning

confidence: 99%

“…Then, a U-Net generator and a discriminator were adversarially trained with the real/fake full-dose PET images. Heart U-Net [161], cGAN [128] Own [208] Sen, Spe, Pre, ACC SSIM, PSNR, MAE, RMSE DSC, USR, OSR, Jaccard Pinaya et al [147] (2022) Brain VQ-VAE [190] MedNIST [216], BRATS [9] Dice, AUPRC, AUROC Autoregressive Transformer [45] WMH [97], MSLUB [106] FPR80, FPR95, FPR99 UK Biobank [177] Own: the dataset was originally collected and processed by the authors of the paper. Sen: sensitivity.…”

Section: Denoisingmentioning

confidence: 99%

“…After training, the reconstruction loss (i.e., 𝐿 2 ) between the real and reconstructed three slices is computed to detect anomalies. A multi-task generative framework was proposed in [208] to detect aortic dissection using non-contrast-enhanced CT (NCE-CT) volumes. The framework consists of a 3D nnU-Net to perform aortas segmentation of NCE-CT volumes and a 3D Multi-Task Generative Architecture (MTGA) to perform CE-CT synthesis, true & false lumen segmentation, and simultaneous aortic dissection detection.…”

Section: Detectionmentioning

confidence: 99%

See 1 more Smart Citation

3D Brain and Heart Volume Generative Models: A Survey

Liu¹,

Dwivedi²,

Boussaid³

et al. 2022

Preprint

View full text Add to dashboard Cite

Generative models such as generative adversarial networks and autoencoders have gained a great deal of attention in the medical field due to their excellent data generation capability. This paper provides a comprehensive survey of generative models for three-dimensional (3D) volumes, focusing on the brain and heart. A new and elaborate taxonomy of unconditional and conditional generative models is proposed to cover diverse medical tasks for the brain and heart: unconditional synthesis, classification, conditional synthesis, segmentation, denoising, detection, and registration. We provide relevant background, examine each task and also suggest potential future directions. A list of the latest publications will be updated on Github to keep up with the rapid influx of papers at https://github.com/csyanbin/3D-Medical-Generative-Survey.

show abstract

“…[3][4][5][6][7] Several methods for automated segmentation of the aorta and pulmonary artery from non-contrast-enhanced CT (NCE-CT) images have been proposed. [8][9][10][11][12][13][14][15] Isgum et al 8 proposed multi-atlas-based segmentation of the cardiac area and aorta in low-dose NCE-CT images. Kurugol et al 9 proposed an automated aorta segmentation and aortic calcification detection method using the aorta circular Hough transformation and refinement using three-dimensional (3D) level sets.…”

Section: Introductionmentioning

confidence: 99%

Aorta and main pulmonary artery segmentation using stacked U‐Net and localization on non‐contrast‐enhanced computed tomography images

et al. 2023

View full text Add to dashboard Cite

BackgroundThe contact between the aorta, main pulmonary artery (MPA), main pulmonary vein, vena cava (VC), and esophagus affects segmentation of the aorta and MPA in non‐contrast‐enhanced computed tomography (NCE‐CT) images.PurposeA two‐stage stacked U‐Net and localization of the aorta and MPA were developed for the segmentation of the aorta and MPA in NCE‐CT images.MethodsNormal‐dose NCE‐CT images of 24 subjects with chronic thromboembolic pulmonary hypertension (CTEPH) and low‐dose NCE‐CT images of 100 subjects without CTEPH were used in this study. The aorta is in contact with the ascending aorta (AA) and MPA, the AA with the VC, the aortic arch (AR) with the VC and esophagus, and the descending aorta (DA) with the esophagus. These contact surfaces were manually annotated. The contact surfaces were quantified using the contact surface ratio (CSR). Segmentation of the aorta and MPA in NCE‐CT images was performed by localization of the aorta and MPA and a two‐stage stacked U‐Net. Localization was performed by extracting and processing the trachea and main bronchus. The first stage of the stacked U‐Net consisted of a 2D U‐Net, 2D U‐Net with a pre‐trained VGG‐16 encoder, and 2D attention U‐Net. The second stage consisted of a 3D U‐Net with four input channels: the CT volume and three segmentation results of the first stage. The model was trained and tested using 10‐fold cross‐validation. Segmentation of the entire volume was evaluated using the Dice similarity coefficient (DSC). Segmentation of the contact area was also assessed using the mean surface distance (MSD). The statistical analysis of the evaluation underwent a multi‐comparison correction. CTEPH and non‐CTEPH cases were classified based on the vessel diameters measured from the segmented MPA.ResultsFor the noncontact surfaces of AA, the MSD of stacked U‐Net was 0.31 ± 0.10 mm (p < 0.05) and 0.32 ± 0.13 mm (p < 0.05) for non‐CTEPH and CTEPH cases, respectively. For contact surfaces with a CSR of 0.4 or greater in AA, the MSD was 0.52 ± 0.23 mm (p < 0.05), and 0.68 ± 0.29 mm (p > 0.05) for non‐CTEPH and CTEPH cases, respectively. MSDs were lower than those of 2D and 3D U‐Nets for contact and noncontact surfaces; moreover, MSDs increased slightly with larger CSRs. However, the stacked U‐Net achieved MSDs of approximately 1 pixel for a wide contact surface. The area under the receiver operating characteristic curve for CTEPH and non‐CTEPH classification using the right main pulmonary artery (RMPA) diameter was 0.97 (95% confidence interval [CI]: 0.94–1.00).ConclusionsSegmentation of the aorta and MPA on NCE‐CT images were affected by vascular and esophageal contact. The application of stacked U‐Net and localization techniques for non‐CTEPH and CTEPH cases mitigated the impact of contact, suggesting its potential for diagnosing CTEPH.

show abstract

A Cascaded Multi-Task Generative Framework for Detecting Aortic Dissection on 3-D Non-Contrast-Enhanced Computed Tomography

Cited by 7 publications

References 49 publications

Diagnosis of Acute Aortic Syndromes on Non-Contrast CT Images with Radiomics-Based Machine Learning

Diagnosis of Acute Aortic Syndromes on Non-Contrast CT Images with Radiomics-Based Machine Learning

3D Brain and Heart Volume Generative Models: A Survey

Aorta and main pulmonary artery segmentation using stacked U‐Net and localization on non‐contrast‐enhanced computed tomography images

Contact Info

Product

Resources

About