1
|
Aghapanah H, Rasti R, Kermani S, Tabesh F, Banaem HY, Aliakbar HP, Sanei H, Segars WP. CardSegNet: An adaptive hybrid CNN-vision transformer model for heart region segmentation in cardiac MRI. Comput Med Imaging Graph 2024; 115:102382. [PMID: 38640619 DOI: 10.1016/j.compmedimag.2024.102382] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 10/01/2023] [Revised: 03/08/2024] [Accepted: 04/10/2024] [Indexed: 04/21/2024]
Abstract
Cardiovascular MRI (CMRI) is a non-invasive imaging technique adopted for assessing the blood circulatory system's structure and function. Precise image segmentation is required to measure cardiac parameters and diagnose abnormalities through CMRI data. Because of anatomical heterogeneity and image variations, cardiac image segmentation is a challenging task. Quantification of cardiac parameters requires high-performance segmentation of the left ventricle (LV), right ventricle (RV), and left ventricle myocardium from the background. The first proposed solution here is to manually segment the regions, which is a time-consuming and error-prone procedure. In this context, many semi- or fully automatic solutions have been proposed recently, among which deep learning-based methods have revealed high performance in segmenting regions in CMRI data. In this study, a self-adaptive multi attention (SMA) module is introduced to adaptively leverage multiple attention mechanisms for better segmentation. The convolutional-based position and channel attention mechanisms with a patch tokenization-based vision transformer (ViT)-based attention mechanism in a hybrid and end-to-end manner are integrated into the SMA. The CNN- and ViT-based attentions mine the short- and long-range dependencies for more precise segmentation. The SMA module is applied in an encoder-decoder structure with a ResNet50 backbone named CardSegNet. Furthermore, a deep supervision method with multi-loss functions is introduced to the CardSegNet optimizer to reduce overfitting and enhance the model's performance. The proposed model is validated on the ACDC2017 (n=100), M&Ms (n=321), and a local dataset (n=22) using the 10-fold cross-validation method with promising segmentation results, demonstrating its outperformance versus its counterparts.
Collapse
Affiliation(s)
- Hamed Aghapanah
- School of Advanced Technologies in Medicine, Isfahan University of Medical Sciences, Isfahan, Iran
| | - Reza Rasti
- Department of Biomedical Engineering, Faculty of Engineering, University of Isfahan, Isfahan, Iran; Department of Biomedical Engineering, Duke University, Durham, NC 27708, USA.
| | - Saeed Kermani
- School of Advanced Technologies in Medicine, Isfahan University of Medical Sciences, Isfahan, Iran.
| | - Faezeh Tabesh
- Cardiovascular Research Institute, Isfahan University of Medical Sciences, Isfahan, Iran
| | - Hossein Yousefi Banaem
- Skull Base Research Center, Loghman Hakim Hospital, Shahid Beheshti University of Medical Sciences, Tehran, Iran
| | - Hamidreza Pour Aliakbar
- Rajaie Cardiovascular Medical and Research Center, Iran University of Medical Sciences, Tehran, Iran
| | - Hamid Sanei
- Cardiovascular Research Institute, Isfahan University of Medical Sciences, Isfahan, Iran
| | - William Paul Segars
- Department of Biomedical Engineering, Faculty of Engineering, University of Isfahan, Isfahan, Iran
| |
Collapse
|
2
|
Shoieb DA, Fathalla KM, Youssef SM, Younes A. CAT-Seg: cascaded medical assistive tool integrating residual attention mechanisms and Squeeze-Net for 3D MRI biventricular segmentation. Phys Eng Sci Med 2024; 47:153-168. [PMID: 37999903 DOI: 10.1007/s13246-023-01352-2] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 03/31/2023] [Accepted: 10/31/2023] [Indexed: 11/25/2023]
Abstract
Cardiac image segmentation is a critical step in the early detection of cardiovascular disease. The segmentation of the biventricular is a prerequisite for evaluating cardiac function in cardiac magnetic resonance imaging (CMRI). In this paper, a cascaded model CAT-Seg is proposed for segmentation of 3D-CMRI volumes. CAT-Seg addresses the problem of biventricular confusion with other regions and localized the region of interest (ROI) to reduce the scope of processing. A modified DeepLabv3+ variant integrating SqueezeNet (SqueezeDeepLabv3+) is proposed as a part of CAT-Seg. SqueezeDeepLabv3+ handles the different shapes of the biventricular through the different cardiac phases, as the biventricular only accounts for small portion of the volume slices. Also, CAT-Seg presents a segmentation approach that integrates attention mechanisms into 3D Residual UNet architecture (3D-ResUNet) called 3D-ARU to improve the segmentation results of the three major structures (left ventricle (LV), Myocardium (Myo), and right ventricle (RV)). The integration of the spatial attention mechanism into ResUNet handles the fuzzy edges of the three structures. The proposed model achieves promising results in training and testing with the Automatic Cardiac Diagnosis Challenge (ACDC 2017) dataset and the external validation using MyoPs. CAT-Seg demonstrates competitive performance with state-of-the-art models. On ACDC 2017, CAT-Seg is able to segment LV, Myo, and RV with an average minimum dice symmetry coefficient (DSC) performance gap of 1.165%, 4.36%, and 3.115% respectively. The average maximum improvement in terms of DSC in segmenting LV, Myo and RV is 4.395%, 6.84% and 7.315% respectively. On MyoPs external validation, CAT-Seg outperformed the state-of-the-art in segmenting LV, Myo, and RV with an average minimum performance gap of 6.13%, 5.44%, and 2.912% respectively.
Collapse
Affiliation(s)
- Doaa A Shoieb
- Computer Engineering Department, Arab Academy for Science, Technology and Maritime Transport (AASTMT), Alexandria, 1029, Egypt.
| | - Karma M Fathalla
- Computer Engineering Department, Arab Academy for Science, Technology and Maritime Transport (AASTMT), Alexandria, 1029, Egypt
| | - Sherin M Youssef
- Computer Engineering Department, Arab Academy for Science, Technology and Maritime Transport (AASTMT), Alexandria, 1029, Egypt
| | - Ahmed Younes
- Department of Mathematics and Computer Science, Faculty of Science, Alexandria University, Alexandria, Egypt
| |
Collapse
|
3
|
Liu X, Su S, Gu W, Yao T, Shen J, Mo Y. Super-Resolution Reconstruction of CT Images Based on Multi-scale Information Fused Generative Adversarial Networks. Ann Biomed Eng 2024; 52:57-70. [PMID: 38064116 DOI: 10.1007/s10439-023-03412-w] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 06/27/2023] [Accepted: 11/16/2023] [Indexed: 01/04/2024]
Abstract
The popularization and widespread use of computed tomography (CT) in the field of medicine evocated public attention to the potential radiation exposure endured by patients. Reducing the radiation dose may lead to scattering noise and low resolution, which can adversely affect the radiologists' judgment. Hence, this paper introduces a new network called PANet-UP-ESRGAN (PAUP-ESRGAN), specifically designed to obtain low-dose CT (LDCT) images with high peak signal-to-noise ratio (PSNR) and high resolution (HR). The model was trained on synthetic medical image data based on a Generative Adversarial Network (GAN). A degradation modeling process was introduced to accurately represent realistic degradation complexities. To reconstruct image edge textures, a pyramidal attention model call PANet was added before the middle of the multiple residual dense blocks (MRDB) in the generator to focus on high-frequency image information. The U-Net discriminator with spectral normalization was also designed to improve its efficiency and stabilize the training dynamics. The proposed PAUP-ESRGAN model was evaluated on the abdomen and lung image datasets, which demonstrated a significant improvement in terms of robustness of model and LDCT image detail reconstruction, compared to the latest real-esrgan network. Results showed that the mean PSNR increated by 19.1%, 25.05%, and 21.25%, the mean SSIM increated by 0.4%, 0.4%, and 0.4%, and the mean NRMSE decreated by 0.25%, 0.25%, and 0.35% at 2[Formula: see text], 4[Formula: see text], and 8[Formula: see text] super-resolution scales, respectively. Experimental results demonstrate that our method outperforms the state-of-the-art super-resolution methods on restoring CT images with respect to peak signal-to-noise ratio (PSNR), structural similarity (SSIM) and normalized root-mean-square error (NRMSE) indices.
Collapse
Affiliation(s)
- Xiaobao Liu
- Faculty of Mechanical and Electrical Engineering, Kunming University of Science and Technology, No.727, Jingming South Road, Chenggong District, Kunming, 650500, China.
| | - Shuailin Su
- Faculty of Mechanical and Electrical Engineering, Kunming University of Science and Technology, No.727, Jingming South Road, Chenggong District, Kunming, 650500, China
| | - Wenjuan Gu
- Faculty of Mechanical and Electrical Engineering, Kunming University of Science and Technology, No.727, Jingming South Road, Chenggong District, Kunming, 650500, China
| | - Tingqiang Yao
- Faculty of Mechanical and Electrical Engineering, Kunming University of Science and Technology, No.727, Jingming South Road, Chenggong District, Kunming, 650500, China
| | - Jihong Shen
- The First Department of Urology, The First Affiliated Hospital of Kunming Medical University, 295 Xichang Road, Chenggong District, Kunming, 650032, China
| | - Yin Mo
- The First Department of Urology, The First Affiliated Hospital of Kunming Medical University, 295 Xichang Road, Chenggong District, Kunming, 650032, China
| |
Collapse
|
4
|
Yan Z, Su Y, Sun H, Yu H, Ma W, Chi H, Cao H, Chang Q. SegNet-based left ventricular MRI segmentation for the diagnosis of cardiac hypertrophy and myocardial infarction. COMPUTER METHODS AND PROGRAMS IN BIOMEDICINE 2022; 227:107197. [PMID: 36351349 DOI: 10.1016/j.cmpb.2022.107197] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Subscribe] [Scholar Register] [Received: 06/18/2022] [Revised: 10/18/2022] [Accepted: 10/20/2022] [Indexed: 06/16/2023]
Abstract
OBJECTIVE A set of cardiac MRI short-axis image dataset is constructed, and an automatic segmentation based on an improved SegNet model is developed to evaluate its performance based on deep learning techniques. METHODS The Affiliated Hospital of Qingdao University collected 1354 cardiac MRI between 2019 and 2022, and the dataset was divided into four categories: for the diagnosis of cardiac hypertrophy and myocardial infraction and normal control group by manual annotation to establish a cardiac MRI library. On the basis, the training set, validation set and test set were separated. SegNet is a classical deep learning segmentation network, which borrows part of the classical convolutional neural network, that pixelates the region of an object in an image division of levels. Its implementation consists of a convolutional neural network. Aiming at the problems of low accuracy and poor generalization ability of current deep learning frameworks in medical image segmentation, this paper proposes a semantic segmentation method based on deep separable convolutional network to improve the SegNet model, and trains the data set. Tensorflow framework was used to train the model and the experiment detection achieves good results. RESULTS In the validation experiment, the sensitivity and specificity of the improved SegNet model in the segmentation of left ventricular MRI were 0.889, 0.965, Dice coefficient was 0.878, Jaccard coefficient was 0.955, and Hausdorff distance was 10.163 mm, showing good segmentation effect. CONCLUSION The segmentation accuracy of the deep learning model developed in this paper can meet the requirements of most clinical medicine applications, and provides technical support for left ventricular identification in cardiac MRI.
Collapse
Affiliation(s)
- Zhisheng Yan
- Department of Cardiovascular Surgery, The Affiliated Hospital of Qingdao University, No. 1677 Wutai mountain Road, Huangdao, Qingdao, Shandong 266000, China
| | - Yujing Su
- Pediatric Clinic, Qingdao Municipal Hospital, Qingdao, Shandong, China
| | - Haixia Sun
- Healthcare Clinic, The Affiliated Hospital of Qingdao University, Qingdao, Shandong, China
| | - Haiyang Yu
- Department of Radiology, The Affiliated Hospital of Qingdao University, Qingdao, Shandong, China
| | - Wanteng Ma
- Department of Cardiovascular Surgery, The Affiliated Hospital of Qingdao University, No. 1677 Wutai mountain Road, Huangdao, Qingdao, Shandong 266000, China
| | - Honghui Chi
- Department of Cardiovascular Surgery, The Affiliated Hospital of Qingdao University, No. 1677 Wutai mountain Road, Huangdao, Qingdao, Shandong 266000, China
| | - Huihui Cao
- Department of Cardiovascular Surgery, The Affiliated Hospital of Qingdao University, No. 1677 Wutai mountain Road, Huangdao, Qingdao, Shandong 266000, China
| | - Qing Chang
- Department of Cardiovascular Surgery, The Affiliated Hospital of Qingdao University, No. 1677 Wutai mountain Road, Huangdao, Qingdao, Shandong 266000, China.
| |
Collapse
|