1
|
Ou J, Jiang L, Bai T, Zhan P, Liu R, Xiao H. ResTransUnet: An effective network combined with Transformer and U-Net for liver segmentation in CT scans. Comput Biol Med 2024; 177:108625. [PMID: 38823365 DOI: 10.1016/j.compbiomed.2024.108625] [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: 07/23/2023] [Revised: 04/15/2024] [Accepted: 05/18/2024] [Indexed: 06/03/2024]
Abstract
Liver segmentation is a fundamental prerequisite for the diagnosis and surgical planning of hepatocellular carcinoma. Traditionally, the liver contour is drawn manually by radiologists using a slice-by-slice method. However, this process is time-consuming and error-prone, depending on the radiologist's experience. In this paper, we propose a new end-to-end automatic liver segmentation framework, named ResTransUNet, which exploits the transformer's ability to capture global context for remote interactions and spatial relationships, as well as the excellent performance of the original U-Net architecture. The main contribution of this paper lies in proposing a novel fusion network that combines Unet and Transformer architectures. In the encoding structure, a dual-path approach is utilized, where features are extracted separately using both convolutional neural networks (CNNs) and Transformer networks. Additionally, an effective feature enhancement unit is designed to transfer the global features extracted by the Transformer network to the CNN for feature enhancement. This model aims to address the drawbacks of traditional Unet-based methods, such as feature loss during encoding and poor capture of global features. Moreover, it avoids the disadvantages of pure Transformer models, which suffer from large parameter sizes and high computational complexity. The experimental results on the LiTS2017 dataset demonstrate remarkable performance for our proposed model, with Dice coefficients, volumetric overlap error (VOE), and relative volume difference (RVD) values for liver segmentation reaching 0.9535, 0.0804, and -0.0007, respectively. Furthermore, to further validate the model's generalization capability, we conducted tests on the 3Dircadb, Chaos, and Sliver07 datasets. The experimental results demonstrate that the proposed method outperforms other closely related models with higher liver segmentation accuracy. In addition, significant improvements can be achieved by applying our method when handling liver segmentation with small and discontinuous liver regions, as well as blurred liver boundaries. The code is available at the website: https://github.com/Jouiry/ResTransUNet.
Collapse
Affiliation(s)
- Jiajie Ou
- School of Artificial Intelligence, Chongqing University of Technology, Chongqing, China
| | - Linfeng Jiang
- School of Artificial Intelligence, Chongqing University of Technology, Chongqing, China; School of Computing and College of Design and Engineering, National University of Singapore, Singapore.
| | - Ting Bai
- School of Electrical Engineering and Computer Science, KTH Royal Institute of Technology, Stockholm, Sweden
| | - Peidong Zhan
- School of Artificial Intelligence, Chongqing University of Technology, Chongqing, China
| | - Ruihua Liu
- School of Artificial Intelligence, Chongqing University of Technology, Chongqing, China
| | - Hanguang Xiao
- School of Artificial Intelligence, Chongqing University of Technology, Chongqing, China
| |
Collapse
|
2
|
Hassan R, Mondal MRH, Ahamed SI. UDBRNet: A novel uncertainty driven boundary refined network for organ at risk segmentation. PLoS One 2024; 19:e0304771. [PMID: 38885241 PMCID: PMC11182520 DOI: 10.1371/journal.pone.0304771] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [MESH Headings] [Track Full Text] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 03/02/2024] [Accepted: 05/19/2024] [Indexed: 06/20/2024] Open
Abstract
Organ segmentation has become a preliminary task for computer-aided intervention, diagnosis, radiation therapy, and critical robotic surgery. Automatic organ segmentation from medical images is a challenging task due to the inconsistent shape and size of different organs. Besides this, low contrast at the edges of organs due to similar types of tissue confuses the network's ability to segment the contour of organs properly. In this paper, we propose a novel convolution neural network based uncertainty-driven boundary-refined segmentation network (UDBRNet) that segments the organs from CT images. The CT images are segmented first and produce multiple segmentation masks from multi-line segmentation decoder. Uncertain regions are identified from multiple masks and the boundaries of the organs are refined based on uncertainty data. Our method achieves remarkable performance, boasting dice accuracies of 0.80, 0.95, 0.92, and 0.94 for Esophagus, Heart, Trachea, and Aorta respectively on the SegThor dataset, and 0.71, 0.89, 0.85, 0.97, and 0.97 for Esophagus, Spinal Cord, Heart, Left-Lung, and Right-Lung respectively on the LCTSC dataset. These results demonstrate the superiority of our uncertainty-driven boundary refinement technique over state-of-the-art segmentation networks such as UNet, Attention UNet, FC-denseNet, BASNet, UNet++, R2UNet, TransUNet, and DS-TransUNet. UDBRNet presents a promising network for more precise organ segmentation, particularly in challenging, uncertain conditions. The source code of our proposed method will be available at https://github.com/riadhassan/UDBRNet.
Collapse
Affiliation(s)
- Riad Hassan
- Institute of Information and Communication Technology, Bangladesh University of Engineering and Technology, Palashi, Dhaka, Bangladesh
| | - M. Rubaiyat Hossain Mondal
- Institute of Information and Communication Technology, Bangladesh University of Engineering and Technology, Palashi, Dhaka, Bangladesh
| | - Sheikh Iqbal Ahamed
- Department of Computer Science, Marquette University, Milwaukee, Wisconsin, United States of America
| |
Collapse
|
3
|
Raj A, Allababidi A, Kayed H, Gerken ALH, Müller J, Schoenberg SO, Zöllner FG, Rink JS. Streamlining Acute Abdominal Aortic Dissection Management-An AI-based CT Imaging Workflow. JOURNAL OF IMAGING INFORMATICS IN MEDICINE 2024:10.1007/s10278-024-01164-0. [PMID: 38864947 DOI: 10.1007/s10278-024-01164-0] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Grants] [Track Full Text] [Subscribe] [Scholar Register] [Received: 02/23/2024] [Revised: 05/24/2024] [Accepted: 06/04/2024] [Indexed: 06/13/2024]
Abstract
Life-threatening acute aortic dissection (AD) demands timely diagnosis for effective intervention. To streamline intrahospital workflows, automated detection of AD in abdominal computed tomography (CT) scans seems useful to assist humans. We aimed at creating a robust convolutional neural network (CNN)-based pipeline capable of real-time screening for signs of abdominal AD in CT. In this retrospective study, abdominal CT data from AD patients presenting with AD and from non-AD patients were collected (n 195, AD cases 94, mean age 65.9 years, female ratio 35.8%). A CNN-based algorithm was developed with the goal of enabling a robust, automated, and highly sensitive detection of abdominal AD. Two sets from internal (n = 32, AD cases 16) and external sources (n = 1189, AD cases 100) were procured for validation. The abdominal region was extracted, followed by the automatic isolation of the aorta region of interest (ROI) and highlighting of the membrane via edge extraction, followed by classification of the aortic ROI as dissected/healthy. A fivefold cross-validation was employed on the internal set, and an ensemble of the 5 trained models was used to predict the internal and external validation set. Evaluation metrics included receiver operating characteristic curve (AUC) and balanced accuracy. The AUC, balanced accuracy, and sensitivity scores of the internal dataset were 0.932 (CI 0.891-0.963), 0.860, and 0.885, respectively. For the internal validation dataset, the AUC, balanced accuracy, and sensitivity scores were 0.887 (CI 0.732-0.988), 0.781, and 0.875, respectively. Furthermore, for the external validation dataset, AUC, balanced accuracy, and sensitivity scores were 0.993 (CI 0.918-0.994), 0.933, and 1.000, respectively. The proposed automated pipeline could assist humans in expediting acute aortic dissection management when integrated into clinical workflows.
Collapse
Affiliation(s)
- Anish Raj
- Computer Assisted Clinical Medicine, Medical Faculty Mannheim, Heidelberg University, Theodor-Kutzer-Ufer 1-3, D-68167, Mannheim, Germany.
- Mannheim Institute for Intelligent Systems in Medicine, Medical Faculty Mannheim, Heidelberg University, Theodor-Kutzer-Ufer 1-3, D-68167, Mannheim, Germany.
| | - Ahmad Allababidi
- Department of Radiology and Nuclear Medicine, University Medical Center Mannheim, Theodor-Kutzer-Ufer 1-3, D-68167, Mannheim, Germany
| | - Hany Kayed
- Department of Radiology and Nuclear Medicine, University Medical Center Mannheim, Theodor-Kutzer-Ufer 1-3, D-68167, Mannheim, Germany
| | - Andreas L H Gerken
- Department of Surgery, Medical Faculty Mannheim, University Medical Center Mannheim, Heidelberg University, Theodor-Kutzer-Ufer 1-3, D-68167, Mannheim, Germany
| | - Julia Müller
- Mediri GmbH, Eppelheimer Straße 13, D-69115, Heidelberg, Germany
| | - Stefan O Schoenberg
- Department of Radiology and Nuclear Medicine, University Medical Center Mannheim, Theodor-Kutzer-Ufer 1-3, D-68167, Mannheim, Germany
| | - Frank G Zöllner
- Computer Assisted Clinical Medicine, Medical Faculty Mannheim, Heidelberg University, Theodor-Kutzer-Ufer 1-3, D-68167, Mannheim, Germany
- Mannheim Institute for Intelligent Systems in Medicine, Medical Faculty Mannheim, Heidelberg University, Theodor-Kutzer-Ufer 1-3, D-68167, Mannheim, Germany
| | - Johann S Rink
- Department of Radiology and Nuclear Medicine, University Medical Center Mannheim, Theodor-Kutzer-Ufer 1-3, D-68167, Mannheim, Germany
| |
Collapse
|
4
|
Zhang G, Gu W, Wang S, Li Y, Zhao D, Liang T, Gong Z, Ju R. MOTC: Abdominal Multi-objective Segmentation Model with Parallel Fusion of Global and Local Information. JOURNAL OF IMAGING INFORMATICS IN MEDICINE 2024; 37:1-16. [PMID: 38347391 PMCID: PMC11169149 DOI: 10.1007/s10278-024-00978-2] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Subscribe] [Scholar Register] [Received: 10/04/2023] [Revised: 11/19/2023] [Accepted: 11/21/2023] [Indexed: 06/13/2024]
Abstract
Convolutional Neural Networks have been widely applied in medical image segmentation. However, the existence of local inductive bias in convolutional operations restricts the modeling of long-term dependencies. The introduction of Transformer enables the modeling of long-term dependencies and partially eliminates the local inductive bias in convolutional operations, thereby improving the accuracy of tasks such as segmentation and classification. Researchers have proposed various hybrid structures combining Transformer and Convolutional Neural Networks. One strategy is to stack Transformer blocks and convolutional blocks to concentrate on eliminating the accumulated local bias of convolutional operations. Another strategy is to nest convolutional blocks and Transformer blocks to eliminate bias within each nested block. However, due to the granularity of bias elimination operations, these two strategies cannot fully exploit the potential of Transformer. In this paper, a parallel hybrid model is proposed for segmentation, which includes a Transformer branch and a Convolutional Neural Network branch in encoder. After parallel feature extraction, inter-layer information fusion and exchange of complementary information are performed between the two branches, simultaneously extracting local and global features while eliminating the local bias generated by convolutional operations within the current layer. A pure convolutional operation is used in decoder to obtain final segmentation results. To validate the impact of the granularity of bias elimination operations on the effectiveness of local bias elimination, the experiments in this paper were conducted on Flare21 dataset and Amos22 dataset. The average Dice coefficient reached 92.65% on Flare21 dataset, and 91.61% on Amos22 dataset, surpassing comparative methods. The experimental results demonstrate that smaller granularity of bias elimination operations leads to better performance.
Collapse
Affiliation(s)
- GuoDong Zhang
- School of Computer, Shenyang Aerospace University, Daoyi South Street, Shenyang, 110135, Liaoning Province, China
| | - WenWen Gu
- School of Computer, Shenyang Aerospace University, Daoyi South Street, Shenyang, 110135, Liaoning Province, China
| | - SuRan Wang
- School of Computer, Shenyang Aerospace University, Daoyi South Street, Shenyang, 110135, Liaoning Province, China
| | - YanLin Li
- School of Computer, Shenyang Aerospace University, Daoyi South Street, Shenyang, 110135, Liaoning Province, China
| | - DaZhe Zhao
- Key Laboratory of Intelligent Computing in Medical Image, Northeastern University, Wenhua Street, Shenyang, 110819, Liaoning Province, China
| | - TingYu Liang
- School of Computer, Shenyang Aerospace University, Daoyi South Street, Shenyang, 110135, Liaoning Province, China
| | - ZhaoXuan Gong
- School of Computer, Shenyang Aerospace University, Daoyi South Street, Shenyang, 110135, Liaoning Province, China
| | - RongHui Ju
- School of Computer, Shenyang Aerospace University, Daoyi South Street, Shenyang, 110135, Liaoning Province, China.
- Department of Radiology, The Peoples Hospital of Liaoning Province, Wenyi Street, Shenyang, 110016, Liaoning Province, China.
| |
Collapse
|
5
|
Remus R, Sure C, Selkmann S, Uttich E, Bender B. Soft tissue material properties based on human abdominal in vivo macro-indenter measurements. Front Bioeng Biotechnol 2024; 12:1384062. [PMID: 38854855 PMCID: PMC11157078 DOI: 10.3389/fbioe.2024.1384062] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 02/08/2024] [Accepted: 04/22/2024] [Indexed: 06/11/2024] Open
Abstract
Simulations of human-technology interaction in the context of product development require comprehensive knowledge of biomechanical in vivo behavior. To obtain this knowledge for the abdomen, we measured the continuous mechanical responses of the abdominal soft tissue of ten healthy participants in different lying positions anteriorly, laterally, and posteriorly under local compression depths of up to 30 mm. An experimental setup consisting of a mechatronic indenter with hemispherical tip and two time-of-flight (ToF) sensors for optical 3D displacement measurement of the surface was developed for this purpose. To account for the impact of muscle tone, experiments were conducted with both controlled activation and relaxation of the trunk muscles. Surface electromyography (sEMG) was used to monitor muscle activation levels. The obtained data sets comprise the continuous force-displacement data of six abdominal measurement regions, each synchronized with the local surface displacements resulting from the macro-indentation, and the bipolar sEMG signals at three key trunk muscles. We used inverse finite element analysis (FEA), to derive sets of nonlinear material parameters that numerically approximate the experimentally determined soft tissue behaviors. The physiological standard values obtained for all participants after data processing served as reference data. The mean stiffness of the abdomen was significantly different when the trunk muscles were activated or relaxed. No significant differences were found between the anterior-lateral measurement regions, with exception of those centered on the linea alba and centered on the muscle belly of the rectus abdominis below the intertubercular plane. The shapes and areas of deformation of the skin depended on the region and muscle activity. Using the hyperelastic Ogden model, we identified unique material parameter sets for all regions. Our findings confirmed that, in addition to the indenter force-displacement data, knowledge about tissue deformation is necessary to reliably determine unique material parameter sets using inverse FEA. The presented results can be used for finite element (FE) models of the abdomen, for example, in the context of orthopedic or biomedical product developments.
Collapse
Affiliation(s)
- Robin Remus
- Chair of Product Development, Department of Mechanical Engineering, Ruhr-University Bochum, Bochum, Germany
| | | | | | | | | |
Collapse
|
6
|
Li S, Li XG, Zhou F, Zhang Y, Bie Z, Cheng L, Peng J, Li B. Automated segmentation of liver and hepatic vessels on portal venous phase computed tomography images using a deep learning algorithm. J Appl Clin Med Phys 2024:e14397. [PMID: 38773719 DOI: 10.1002/acm2.14397] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 02/27/2024] [Revised: 04/20/2024] [Accepted: 04/29/2024] [Indexed: 05/24/2024] Open
Abstract
BACKGROUND CT-image segmentation for liver and hepatic vessels can facilitate liver surgical planning. However, time-consuming process and inter-observer variations of manual segmentation have limited wider application in clinical practice. PURPOSE Our study aimed to propose an automated deep learning (DL) segmentation algorithm for liver and hepatic vessels on portal venous phase CT images. METHODS This retrospective study was performed to develop a coarse-to-fine DL-based algorithm that was trained, validated, and tested using private 413, 52, and 50 portal venous phase CT images, respectively. Additionally, the performance of the DL algorithm was extensively evaluated and compared with manual segmentation using an independent clinical dataset of preoperative contrast-enhanced CT images from 44 patients with hepatic focal lesions. The accuracy of DL-based segmentation was quantitatively evaluated using the Dice Similarity Coefficient (DSC) and complementary metrics [Normalized Surface Dice (NSD) and Hausdorff distance_95 (HD95) for liver segmentation, Recall and Precision for hepatic vessel segmentation]. The processing time for DL and manual segmentation was also compared. RESULTS Our DL algorithm achieved accurate liver segmentation with DSC of 0.98, NSD of 0.92, and HD95 of 1.52 mm. DL-segmentation of hepatic veins, portal veins, and inferior vena cava attained DSC of 0.86, 0.89, and 0.94, respectively. Compared with the manual approach, the DL algorithm significantly outperformed with better segmentation results for both liver and hepatic vessels, with higher accuracy of liver and hepatic vessel segmentation (all p < 0.001) in independent 44 clinical data. In addition, the DL method significantly reduced the manual processing time of clinical postprocessing (p < 0.001). CONCLUSIONS The proposed DL algorithm potentially enabled accurate and rapid segmentation for liver and hepatic vessels using portal venous phase contrast CT images.
Collapse
Affiliation(s)
- Shengwei Li
- Minimally Invasive Tumor Therapy Center, Beijing Hospital, Peking Union Medical College, Beijing, China
| | - Xiao-Guang Li
- Minimally Invasive Tumor Therapy Center, Beijing Hospital, Peking Union Medical College, Beijing, China
| | - Fanyu Zhou
- Minimally Invasive Tumor Therapy Center, Beijing Hospital, Peking Union Medical College, Beijing, China
| | - Yumeng Zhang
- Minimally Invasive Tumor Therapy Center, Beijing Hospital, Peking Union Medical College, Beijing, China
| | - Zhixin Bie
- Minimally Invasive Tumor Therapy Center, Beijing Hospital, Peking Union Medical College, Beijing, China
| | - Lin Cheng
- Minimally Invasive Tumor Therapy Center, Beijing Hospital, Peking Union Medical College, Beijing, China
| | - Jinzhao Peng
- Minimally Invasive Tumor Therapy Center, Beijing Hospital, Peking Union Medical College, Beijing, China
| | - Bin Li
- Minimally Invasive Tumor Therapy Center, Beijing Hospital, Peking Union Medical College, Beijing, China
| |
Collapse
|
7
|
Li S, Wang H, Meng Y, Zhang C, Song Z. Multi-organ segmentation: a progressive exploration of learning paradigms under scarce annotation. Phys Med Biol 2024; 69:11TR01. [PMID: 38479023 DOI: 10.1088/1361-6560/ad33b5] [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: 06/29/2023] [Accepted: 03/13/2024] [Indexed: 05/21/2024]
Abstract
Precise delineation of multiple organs or abnormal regions in the human body from medical images plays an essential role in computer-aided diagnosis, surgical simulation, image-guided interventions, and especially in radiotherapy treatment planning. Thus, it is of great significance to explore automatic segmentation approaches, among which deep learning-based approaches have evolved rapidly and witnessed remarkable progress in multi-organ segmentation. However, obtaining an appropriately sized and fine-grained annotated dataset of multiple organs is extremely hard and expensive. Such scarce annotation limits the development of high-performance multi-organ segmentation models but promotes many annotation-efficient learning paradigms. Among these, studies on transfer learning leveraging external datasets, semi-supervised learning including unannotated datasets and partially-supervised learning integrating partially-labeled datasets have led the dominant way to break such dilemmas in multi-organ segmentation. We first review the fully supervised method, then present a comprehensive and systematic elaboration of the 3 abovementioned learning paradigms in the context of multi-organ segmentation from both technical and methodological perspectives, and finally summarize their challenges and future trends.
Collapse
Affiliation(s)
- Shiman Li
- Digital Medical Research Center, School of Basic Medical Science, Fudan University, Shanghai Key Lab of Medical Image Computing and Computer Assisted Intervention, Shanghai 200032, People's Republic of China
| | - Haoran Wang
- Digital Medical Research Center, School of Basic Medical Science, Fudan University, Shanghai Key Lab of Medical Image Computing and Computer Assisted Intervention, Shanghai 200032, People's Republic of China
| | - Yucong Meng
- Digital Medical Research Center, School of Basic Medical Science, Fudan University, Shanghai Key Lab of Medical Image Computing and Computer Assisted Intervention, Shanghai 200032, People's Republic of China
| | - Chenxi Zhang
- Digital Medical Research Center, School of Basic Medical Science, Fudan University, Shanghai Key Lab of Medical Image Computing and Computer Assisted Intervention, Shanghai 200032, People's Republic of China
| | - Zhijian Song
- Digital Medical Research Center, School of Basic Medical Science, Fudan University, Shanghai Key Lab of Medical Image Computing and Computer Assisted Intervention, Shanghai 200032, People's Republic of China
| |
Collapse
|
8
|
Liu H, Xu Z, Gao R, Li H, Wang J, Chabin G, Oguz I, Grbic S. COSST: Multi-Organ Segmentation With Partially Labeled Datasets Using Comprehensive Supervisions and Self-Training. IEEE TRANSACTIONS ON MEDICAL IMAGING 2024; 43:1995-2009. [PMID: 38224508 DOI: 10.1109/tmi.2024.3354673] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [MESH Headings] [Track Full Text] [Subscribe] [Scholar Register] [Indexed: 01/17/2024]
Abstract
Deep learning models have demonstrated remarkable success in multi-organ segmentation but typically require large-scale datasets with all organs of interest annotated. However, medical image datasets are often low in sample size and only partially labeled, i.e., only a subset of organs are annotated. Therefore, it is crucial to investigate how to learn a unified model on the available partially labeled datasets to leverage their synergistic potential. In this paper, we systematically investigate the partial-label segmentation problem with theoretical and empirical analyses on the prior techniques. We revisit the problem from a perspective of partial label supervision signals and identify two signals derived from ground truth and one from pseudo labels. We propose a novel two-stage framework termed COSST, which effectively and efficiently integrates comprehensive supervision signals with self-training. Concretely, we first train an initial unified model using two ground truth-based signals and then iteratively incorporate the pseudo label signal to the initial model using self-training. To mitigate performance degradation caused by unreliable pseudo labels, we assess the reliability of pseudo labels via outlier detection in latent space and exclude the most unreliable pseudo labels from each self-training iteration. Extensive experiments are conducted on one public and three private partial-label segmentation tasks over 12 CT datasets. Experimental results show that our proposed COSST achieves significant improvement over the baseline method, i.e., individual networks trained on each partially labeled dataset. Compared to the state-of-the-art partial-label segmentation methods, COSST demonstrates consistent superior performance on various segmentation tasks and with different training data sizes.
Collapse
|
9
|
Kunhimon S, Shaker A, Naseer M, Khan S, Khan FS. Learnable weight initialization for volumetric medical image segmentation. Artif Intell Med 2024; 151:102863. [PMID: 38593682 DOI: 10.1016/j.artmed.2024.102863] [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: 08/31/2023] [Revised: 03/27/2024] [Accepted: 03/30/2024] [Indexed: 04/11/2024]
Abstract
Hybrid volumetric medical image segmentation models, combining the advantages of local convolution and global attention, have recently received considerable attention. While mainly focusing on architectural modifications, most existing hybrid approaches still use conventional data-independent weight initialization schemes which restrict their performance due to ignoring the inherent volumetric nature of the medical data. To address this issue, we propose a learnable weight initialization approach that utilizes the available medical training data to effectively learn the contextual and structural cues via the proposed self-supervised objectives. Our approach is easy to integrate into any hybrid model and requires no external training data. Experiments on multi-organ and lung cancer segmentation tasks demonstrate the effectiveness of our approach, leading to state-of-the-art segmentation performance. Our proposed data-dependent initialization approach performs favorably as compared to the Swin-UNETR model pretrained using large-scale datasets on multi-organ segmentation task. Our source code and models are available at: https://github.com/ShahinaKK/LWI-VMS.
Collapse
Affiliation(s)
- Shahina Kunhimon
- Mohammed Bin Zayed University of Artificial Intelligence, Abu Dhabi, United Arab Emirates.
| | - Abdelrahman Shaker
- Mohammed Bin Zayed University of Artificial Intelligence, Abu Dhabi, United Arab Emirates
| | - Muzammal Naseer
- Mohammed Bin Zayed University of Artificial Intelligence, Abu Dhabi, United Arab Emirates
| | - Salman Khan
- Mohammed Bin Zayed University of Artificial Intelligence, Abu Dhabi, United Arab Emirates
| | - Fahad Shahbaz Khan
- Mohammed Bin Zayed University of Artificial Intelligence, Abu Dhabi, United Arab Emirates; Linkoping University, Sweden
| |
Collapse
|
10
|
Chen Z, Yao L, Liu Y, Han X, Gong Z, Luo J, Zhao J, Fang G. Deep learning-aided 3D proxy-bridged region-growing framework for multi-organ segmentation. Sci Rep 2024; 14:9784. [PMID: 38684904 PMCID: PMC11059262 DOI: 10.1038/s41598-024-60668-5] [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: 05/13/2023] [Accepted: 04/25/2024] [Indexed: 05/02/2024] Open
Abstract
Accurate multi-organ segmentation in 3D CT images is imperative for enhancing computer-aided diagnosis and radiotherapy planning. However, current deep learning-based methods for 3D multi-organ segmentation face challenges such as the need for labor-intensive manual pixel-level annotations and high hardware resource demands, especially regarding GPU resources. To address these issues, we propose a 3D proxy-bridged region-growing framework specifically designed for the segmentation of the liver and spleen. Specifically, a key slice is selected from each 3D volume according to the corresponding intensity histogram. Subsequently, a deep learning model is employed to pinpoint the semantic central patch on this key slice, to calculate the growing seed. To counteract the impact of noise, segmentation of the liver and spleen is conducted on superpixel images created through proxy-bridging strategy. The segmentation process is then extended to adjacent slices by applying the same methodology iteratively, culminating in the comprehensive segmentation results. Experimental results demonstrate that the proposed framework accomplishes segmentation of the liver and spleen with an average Dice Similarity Coefficient of approximately 0.93 and a Jaccard Similarity Coefficient of around 0.88. These outcomes substantiate the framework's capability to achieve performance on par with that of deep learning methods, albeit requiring less guidance information and lower GPU resources.
Collapse
Affiliation(s)
- Zhihong Chen
- Institute of Computing Science and Technology, Guangzhou University, Guangzhou, 510006, China
- Guangdong Provincial Key Laboratory of Artificial Intelligence in Medical Image Analysis and Application, Guangzhou, 510080, China
| | - Lisha Yao
- Guangdong Provincial Key Laboratory of Artificial Intelligence in Medical Image Analysis and Application, Guangzhou, 510080, China
- School of Medicine, South China University of Technology, Guangzhou, 510180, China
| | - Yue Liu
- Institute of Computing Science and Technology, Guangzhou University, Guangzhou, 510006, China
- School of Information Engineering, Jiangxi College of Applied Technology, Ganzhou, 341000, China
| | - Xiaorui Han
- Department of Radiology, School of Medicine, Guangzhou First People's Hospital, South China University of Technology, Guangzhou, 510180, China
| | - Zhengze Gong
- Information and Data Centre, School of Medicine, Guangzhou First People's Hospital, South China University of Technology Guangdong, Guangzhou, 510180, China
| | - Jichao Luo
- Institute of Computing Science and Technology, Guangzhou University, Guangzhou, 510006, China
- Guangdong Provincial Key Laboratory of Artificial Intelligence in Medical Image Analysis and Application, Guangzhou, 510080, China
| | - Jietong Zhao
- Institute of Computing Science and Technology, Guangzhou University, Guangzhou, 510006, China
| | - Gang Fang
- Institute of Computing Science and Technology, Guangzhou University, Guangzhou, 510006, China.
- Guangdong Provincial Key Laboratory of Artificial Intelligence in Medical Image Analysis and Application, Guangzhou, 510080, China.
| |
Collapse
|
11
|
Yang H, Wang Q, Zhang Y, An Z, Liu C, Zhang X, Zhou SK. Lung Nodule Segmentation and Uncertain Region Prediction With an Uncertainty-Aware Attention Mechanism. IEEE TRANSACTIONS ON MEDICAL IMAGING 2024; 43:1284-1295. [PMID: 37966939 DOI: 10.1109/tmi.2023.3332944] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [MESH Headings] [Track Full Text] [Subscribe] [Scholar Register] [Indexed: 11/17/2023]
Abstract
Radiologists possess diverse training and clinical experiences, leading to variations in the segmentation annotations of lung nodules and resulting in segmentation uncertainty. Conventional methods typically select a single annotation as the learning target or attempt to learn a latent space comprising multiple annotations. However, these approaches fail to leverage the valuable information inherent in the consensus and disagreements among the multiple annotations. In this paper, we propose an Uncertainty-Aware Attention Mechanism (UAAM) that utilizes consensus and disagreements among multiple annotations to facilitate better segmentation. To this end, we introduce the Multi-Confidence Mask (MCM), which combines a Low-Confidence (LC) Mask and a High-Confidence (HC) Mask. The LC mask indicates regions with low segmentation confidence, where radiologists may have different segmentation choices. Following UAAM, we further design an Uncertainty-Guide Multi-Confidence Segmentation Network (UGMCS-Net), which contains three modules: a Feature Extracting Module that captures a general feature of a lung nodule, an Uncertainty-Aware Module that produces three features for the annotations' union, intersection, and annotation set, and an Intersection-Union Constraining Module that uses distances between the three features to balance the predictions of final segmentation and MCM. To comprehensively demonstrate the performance of our method, we propose a Complex-Nodule Validation on LIDC-IDRI, which tests UGMCS-Net's segmentation performance on lung nodules that are difficult to segment using common methods. Experimental results demonstrate that our method can significantly improve the segmentation performance on nodules that are difficult to segment using conventional methods.
Collapse
|
12
|
Huang Y, Yang J, Sun Q, Yuan Y, Li H, Hou Y. Multi-residual 2D network integrating spatial correlation for whole heart segmentation. Comput Biol Med 2024; 172:108261. [PMID: 38508056 DOI: 10.1016/j.compbiomed.2024.108261] [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: 12/19/2023] [Revised: 02/21/2024] [Accepted: 03/06/2024] [Indexed: 03/22/2024]
Abstract
Whole heart segmentation (WHS) has significant clinical value for cardiac anatomy, modeling, and analysis of cardiac function. This study aims to address the WHS accuracy on cardiac CT images, as well as the fast inference speed and low graphics processing unit (GPU) memory consumption required by practical clinical applications. Thus, we propose a multi-residual two-dimensional (2D) network integrating spatial correlation for WHS. The network performs slice-by-slice segmentation on three-dimensional cardiac CT images in a 2D encoder-decoder manner. In the network, a convolutional long short-term memory skip connection module is designed to perform spatial correlation feature extraction on the feature maps at different resolutions extracted by the sub-modules of the pre-trained ResNet-based encoder. Moreover, a decoder based on the multi-residual module is designed to analyze the extracted features from the perspectives of multi-scale and channel attention, thereby accurately delineating the various substructures of the heart. The proposed method is verified on a dataset of the multi-modality WHS challenge, an in-house WHS dataset, and a dataset of the abdominal organ segmentation challenge. The dice, Jaccard, average symmetric surface distance, Hausdorff distance, inference time, and maximum GPU memory of the WHS are 0.914, 0.843, 1.066 mm, 15.778 mm, 9.535 s, and 1905 MB, respectively. The proposed network has high accuracy, fast inference speed, minimal GPU memory consumption, strong robustness, and good generalization. It can be deployed to clinical practical applications for WHS and can be effectively extended and applied to other multi-organ segmentation fields. The source code is publicly available at https://github.com/nancy1984yan/MultiResNet-SC.
Collapse
Affiliation(s)
- Yan Huang
- Key Laboratory of Intelligent Computing in Medical Image, Ministry of Education, Northeastern University, Shenyang, Liaoning, China; School of Computer Science and Engineering, Northeastern University, Shenyang, Liaoning, China
| | - Jinzhu Yang
- Key Laboratory of Intelligent Computing in Medical Image, Ministry of Education, Northeastern University, Shenyang, Liaoning, China; School of Computer Science and Engineering, Northeastern University, Shenyang, Liaoning, China; National Frontiers Science Center for Industrial Intelligence and Systems Optimization, Shenyang, Liaoning, China.
| | - Qi Sun
- Key Laboratory of Intelligent Computing in Medical Image, Ministry of Education, Northeastern University, Shenyang, Liaoning, China; School of Computer Science and Engineering, Northeastern University, Shenyang, Liaoning, China
| | - Yuliang Yuan
- Key Laboratory of Intelligent Computing in Medical Image, Ministry of Education, Northeastern University, Shenyang, Liaoning, China; School of Computer Science and Engineering, Northeastern University, Shenyang, Liaoning, China
| | - Honghe Li
- Key Laboratory of Intelligent Computing in Medical Image, Ministry of Education, Northeastern University, Shenyang, Liaoning, China; School of Computer Science and Engineering, Northeastern University, Shenyang, Liaoning, China
| | - Yang Hou
- Department of Radiology, Shengjing Hospital of China Medical University, Shenyang, Liaoning, China
| |
Collapse
|
13
|
Zhang Y, Shen Z, Jiao R. Segment anything model for medical image segmentation: Current applications and future directions. Comput Biol Med 2024; 171:108238. [PMID: 38422961 DOI: 10.1016/j.compbiomed.2024.108238] [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: 08/07/2023] [Revised: 02/06/2024] [Accepted: 02/25/2024] [Indexed: 03/02/2024]
Abstract
Due to the inherent flexibility of prompting, foundation models have emerged as the predominant force in the fields of natural language processing and computer vision. The recent introduction of the Segment Anything Model (SAM) signifies a noteworthy expansion of the prompt-driven paradigm into the domain of image segmentation, thereby introducing a plethora of previously unexplored capabilities. However, the viability of its application to medical image segmentation remains uncertain, given the substantial distinctions between natural and medical images. In this work, we provide a comprehensive overview of recent endeavors aimed at extending the efficacy of SAM to medical image segmentation tasks, encompassing both empirical benchmarking and methodological adaptations. Additionally, we explore potential avenues for future research directions in SAM's role within medical image segmentation. While direct application of SAM to medical image segmentation does not yield satisfactory performance on multi-modal and multi-target medical datasets so far, numerous insights gleaned from these efforts serve as valuable guidance for shaping the trajectory of foundational models in the realm of medical image analysis. To support ongoing research endeavors, we maintain an active repository that contains an up-to-date paper list and a succinct summary of open-source projects at https://github.com/YichiZhang98/SAM4MIS.
Collapse
Affiliation(s)
- Yichi Zhang
- School of Data Science, Fudan University, Shanghai, China.
| | - Zhenrong Shen
- School of Biomedical Engineering, Shanghai Jiao Tong University, Shanghai, China
| | - Rushi Jiao
- School of Biomedical Engineering, Shanghai Jiao Tong University, Shanghai, China
| |
Collapse
|
14
|
Anghel C, Grasu MC, Anghel DA, Rusu-Munteanu GI, Dumitru RL, Lupescu IG. Pancreatic Adenocarcinoma: Imaging Modalities and the Role of Artificial Intelligence in Analyzing CT and MRI Images. Diagnostics (Basel) 2024; 14:438. [PMID: 38396476 PMCID: PMC10887967 DOI: 10.3390/diagnostics14040438] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 01/10/2024] [Revised: 02/10/2024] [Accepted: 02/14/2024] [Indexed: 02/25/2024] Open
Abstract
Pancreatic ductal adenocarcinoma (PDAC) stands out as the predominant malignant neoplasm affecting the pancreas, characterized by a poor prognosis, in most cases patients being diagnosed in a nonresectable stage. Image-based artificial intelligence (AI) models implemented in tumor detection, segmentation, and classification could improve diagnosis with better treatment options and increased survival. This review included papers published in the last five years and describes the current trends in AI algorithms used in PDAC. We analyzed the applications of AI in the detection of PDAC, segmentation of the lesion, and classification algorithms used in differential diagnosis, prognosis, and histopathological and genomic prediction. The results show a lack of multi-institutional collaboration and stresses the need for bigger datasets in order for AI models to be implemented in a clinically relevant manner.
Collapse
Affiliation(s)
- Cristian Anghel
- Faculty of Medicine, Department of Medical Imaging and Interventional Radiology, Carol Davila University of Medicine and Pharmacy Bucharest, 020021 Bucharest, Romania; (C.A.); (R.L.D.); (I.G.L.)
- Department of Radiology and Medical Imaging, Fundeni Clinical Institute, 022328 Bucharest, Romania; (D.A.A.); (G.-I.R.-M.)
| | - Mugur Cristian Grasu
- Faculty of Medicine, Department of Medical Imaging and Interventional Radiology, Carol Davila University of Medicine and Pharmacy Bucharest, 020021 Bucharest, Romania; (C.A.); (R.L.D.); (I.G.L.)
- Department of Radiology and Medical Imaging, Fundeni Clinical Institute, 022328 Bucharest, Romania; (D.A.A.); (G.-I.R.-M.)
| | - Denisa Andreea Anghel
- Department of Radiology and Medical Imaging, Fundeni Clinical Institute, 022328 Bucharest, Romania; (D.A.A.); (G.-I.R.-M.)
| | - Gina-Ionela Rusu-Munteanu
- Department of Radiology and Medical Imaging, Fundeni Clinical Institute, 022328 Bucharest, Romania; (D.A.A.); (G.-I.R.-M.)
| | - Radu Lucian Dumitru
- Faculty of Medicine, Department of Medical Imaging and Interventional Radiology, Carol Davila University of Medicine and Pharmacy Bucharest, 020021 Bucharest, Romania; (C.A.); (R.L.D.); (I.G.L.)
- Department of Radiology and Medical Imaging, Fundeni Clinical Institute, 022328 Bucharest, Romania; (D.A.A.); (G.-I.R.-M.)
| | - Ioana Gabriela Lupescu
- Faculty of Medicine, Department of Medical Imaging and Interventional Radiology, Carol Davila University of Medicine and Pharmacy Bucharest, 020021 Bucharest, Romania; (C.A.); (R.L.D.); (I.G.L.)
- Department of Radiology and Medical Imaging, Fundeni Clinical Institute, 022328 Bucharest, Romania; (D.A.A.); (G.-I.R.-M.)
| |
Collapse
|
15
|
Selle M, Kircher M, Schwennen C, Visscher C, Jung K. Dimension reduction and outlier detection of 3-D shapes derived from multi-organ CT images. BMC Med Inform Decis Mak 2024; 24:49. [PMID: 38355504 PMCID: PMC10865689 DOI: 10.1186/s12911-024-02457-8] [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: 03/24/2023] [Accepted: 02/08/2024] [Indexed: 02/16/2024] Open
Abstract
BACKGROUND Unsupervised clustering and outlier detection are important in medical research to understand the distributional composition of a collective of patients. A number of clustering methods exist, also for high-dimensional data after dimension reduction. Clustering and outlier detection may, however, become less robust or contradictory if multiple high-dimensional data sets per patient exist. Such a scenario is given when the focus is on 3-D data of multiple organs per patient, and a high-dimensional feature matrix per organ is extracted. METHODS We use principal component analysis (PCA), t-distributed stochastic neighbor embedding (t-SNE) and multiple co-inertia analysis (MCIA) combined with bagplots to study the distribution of multi-organ 3-D data taken by computed tomography scans. After point-set registration of multiple organs from two public data sets, multiple hundred shape features are extracted per organ. While PCA and t-SNE can only be applied to each organ individually, MCIA can project the data of all organs into the same low-dimensional space. RESULTS MCIA is the only approach, here, with which data of all organs can be projected into the same low-dimensional space. We studied how frequently (i.e., by how many organs) a patient was classified to belong to the inner or outer 50% of the population, or as an outlier. Outliers could only be detected with MCIA and PCA. MCIA and t-SNE were more robust in judging the distributional location of a patient in contrast to PCA. CONCLUSIONS MCIA is more appropriate and robust in judging the distributional location of a patient in the case of multiple high-dimensional data sets per patient. It is still recommendable to apply PCA or t-SNE in parallel to MCIA to study the location of individual organs.
Collapse
Affiliation(s)
- Michael Selle
- Institute of Animal Genomics, University of Veterinary Medicine Hannover, Hannover, Germany.
| | - Magdalena Kircher
- Institute of Animal Genomics, University of Veterinary Medicine Hannover, Hannover, Germany
| | - Cornelia Schwennen
- Institute for Animal Nutrition, University of Veterinary Medicine Hannover, Hannover, Germany
| | - Christian Visscher
- Institute for Animal Nutrition, University of Veterinary Medicine Hannover, Hannover, Germany
| | - Klaus Jung
- Institute of Animal Genomics, University of Veterinary Medicine Hannover, Hannover, Germany.
| |
Collapse
|
16
|
Huang Y, Yang X, Liu L, Zhou H, Chang A, Zhou X, Chen R, Yu J, Chen J, Chen C, Liu S, Chi H, Hu X, Yue K, Li L, Grau V, Fan DP, Dong F, Ni D. Segment anything model for medical images? Med Image Anal 2024; 92:103061. [PMID: 38086235 DOI: 10.1016/j.media.2023.103061] [Citation(s) in RCA: 3] [Impact Index Per Article: 3.0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 04/29/2023] [Revised: 09/28/2023] [Accepted: 12/05/2023] [Indexed: 01/12/2024]
Abstract
The Segment Anything Model (SAM) is the first foundation model for general image segmentation. It has achieved impressive results on various natural image segmentation tasks. However, medical image segmentation (MIS) is more challenging because of the complex modalities, fine anatomical structures, uncertain and complex object boundaries, and wide-range object scales. To fully validate SAM's performance on medical data, we collected and sorted 53 open-source datasets and built a large medical segmentation dataset with 18 modalities, 84 objects, 125 object-modality paired targets, 1050K 2D images, and 6033K masks. We comprehensively analyzed different models and strategies on the so-called COSMOS 1050K dataset. Our findings mainly include the following: (1) SAM showed remarkable performance in some specific objects but was unstable, imperfect, or even totally failed in other situations. (2) SAM with the large ViT-H showed better overall performance than that with the small ViT-B. (3) SAM performed better with manual hints, especially box, than the Everything mode. (4) SAM could help human annotation with high labeling quality and less time. (5) SAM was sensitive to the randomness in the center point and tight box prompts, and may suffer from a serious performance drop. (6) SAM performed better than interactive methods with one or a few points, but will be outpaced as the number of points increases. (7) SAM's performance correlated to different factors, including boundary complexity, intensity differences, etc. (8) Finetuning the SAM on specific medical tasks could improve its average DICE performance by 4.39% and 6.68% for ViT-B and ViT-H, respectively. Codes and models are available at: https://github.com/yuhoo0302/Segment-Anything-Model-for-Medical-Images. We hope that this comprehensive report can help researchers explore the potential of SAM applications in MIS, and guide how to appropriately use and develop SAM.
Collapse
Affiliation(s)
- Yuhao Huang
- National-Regional Key Technology Engineering Laboratory for Medical Ultrasound, School of Biomedical Engineering, Shenzhen University Medical School, Shenzhen University, Shenzhen, China; Medical UltraSound Image Computing (MUSIC) Lab, Shenzhen University, Shenzhen, China; Marshall Laboratory of Biomedical Engineering, Shenzhen University, Shenzhen, China
| | - Xin Yang
- National-Regional Key Technology Engineering Laboratory for Medical Ultrasound, School of Biomedical Engineering, Shenzhen University Medical School, Shenzhen University, Shenzhen, China; Medical UltraSound Image Computing (MUSIC) Lab, Shenzhen University, Shenzhen, China; Marshall Laboratory of Biomedical Engineering, Shenzhen University, Shenzhen, China
| | - Lian Liu
- National-Regional Key Technology Engineering Laboratory for Medical Ultrasound, School of Biomedical Engineering, Shenzhen University Medical School, Shenzhen University, Shenzhen, China; Medical UltraSound Image Computing (MUSIC) Lab, Shenzhen University, Shenzhen, China; Marshall Laboratory of Biomedical Engineering, Shenzhen University, Shenzhen, China
| | - Han Zhou
- National-Regional Key Technology Engineering Laboratory for Medical Ultrasound, School of Biomedical Engineering, Shenzhen University Medical School, Shenzhen University, Shenzhen, China; Medical UltraSound Image Computing (MUSIC) Lab, Shenzhen University, Shenzhen, China; Marshall Laboratory of Biomedical Engineering, Shenzhen University, Shenzhen, China
| | - Ao Chang
- National-Regional Key Technology Engineering Laboratory for Medical Ultrasound, School of Biomedical Engineering, Shenzhen University Medical School, Shenzhen University, Shenzhen, China; Medical UltraSound Image Computing (MUSIC) Lab, Shenzhen University, Shenzhen, China; Marshall Laboratory of Biomedical Engineering, Shenzhen University, Shenzhen, China
| | - Xinrui Zhou
- National-Regional Key Technology Engineering Laboratory for Medical Ultrasound, School of Biomedical Engineering, Shenzhen University Medical School, Shenzhen University, Shenzhen, China; Medical UltraSound Image Computing (MUSIC) Lab, Shenzhen University, Shenzhen, China; Marshall Laboratory of Biomedical Engineering, Shenzhen University, Shenzhen, China
| | - Rusi Chen
- National-Regional Key Technology Engineering Laboratory for Medical Ultrasound, School of Biomedical Engineering, Shenzhen University Medical School, Shenzhen University, Shenzhen, China; Medical UltraSound Image Computing (MUSIC) Lab, Shenzhen University, Shenzhen, China; Marshall Laboratory of Biomedical Engineering, Shenzhen University, Shenzhen, China
| | - Junxuan Yu
- National-Regional Key Technology Engineering Laboratory for Medical Ultrasound, School of Biomedical Engineering, Shenzhen University Medical School, Shenzhen University, Shenzhen, China; Medical UltraSound Image Computing (MUSIC) Lab, Shenzhen University, Shenzhen, China; Marshall Laboratory of Biomedical Engineering, Shenzhen University, Shenzhen, China
| | - Jiongquan Chen
- National-Regional Key Technology Engineering Laboratory for Medical Ultrasound, School of Biomedical Engineering, Shenzhen University Medical School, Shenzhen University, Shenzhen, China; Medical UltraSound Image Computing (MUSIC) Lab, Shenzhen University, Shenzhen, China; Marshall Laboratory of Biomedical Engineering, Shenzhen University, Shenzhen, China
| | - Chaoyu Chen
- National-Regional Key Technology Engineering Laboratory for Medical Ultrasound, School of Biomedical Engineering, Shenzhen University Medical School, Shenzhen University, Shenzhen, China; Medical UltraSound Image Computing (MUSIC) Lab, Shenzhen University, Shenzhen, China; Marshall Laboratory of Biomedical Engineering, Shenzhen University, Shenzhen, China
| | - Sijing Liu
- National-Regional Key Technology Engineering Laboratory for Medical Ultrasound, School of Biomedical Engineering, Shenzhen University Medical School, Shenzhen University, Shenzhen, China; Medical UltraSound Image Computing (MUSIC) Lab, Shenzhen University, Shenzhen, China; Marshall Laboratory of Biomedical Engineering, Shenzhen University, Shenzhen, China
| | | | - Xindi Hu
- Shenzhen RayShape Medical Technology Co., Ltd, Shenzhen, China
| | - Kejuan Yue
- Hunan First Normal University, Changsha, China
| | - Lei Li
- Department of Engineering Science, University of Oxford, Oxford, UK
| | - Vicente Grau
- Department of Engineering Science, University of Oxford, Oxford, UK
| | - Deng-Ping Fan
- Computer Vision Lab (CVL), ETH Zurich, Zurich, Switzerland
| | - Fajin Dong
- Ultrasound Department, the Second Clinical Medical College, Jinan University, China; First Affiliated Hospital, Southern University of Science and Technology, Shenzhen People's Hospital, Shenzhen, China.
| | - Dong Ni
- National-Regional Key Technology Engineering Laboratory for Medical Ultrasound, School of Biomedical Engineering, Shenzhen University Medical School, Shenzhen University, Shenzhen, China; Medical UltraSound Image Computing (MUSIC) Lab, Shenzhen University, Shenzhen, China; Marshall Laboratory of Biomedical Engineering, Shenzhen University, Shenzhen, China.
| |
Collapse
|
17
|
Li J, Li H, Zhang Y, Wang Z, Zhu S, Li X, Hu K, Gao X. MCNet: A multi-level context-aware network for the segmentation of adrenal gland in CT images. Neural Netw 2024; 170:136-148. [PMID: 37979222 DOI: 10.1016/j.neunet.2023.11.028] [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: 06/13/2023] [Revised: 10/14/2023] [Accepted: 11/11/2023] [Indexed: 11/20/2023]
Abstract
Accurate segmentation of the adrenal gland from abdominal computed tomography (CT) scans is a crucial step towards facilitating the computer-aided diagnosis of adrenal-related diseases such as essential hypertension and adrenal tumors. However, the small size of the adrenal gland, which occupies less than 1% of the abdominal CT slice, poses a significant challenge to accurate segmentation. To address this problem, we propose a novel multi-level context-aware network (MCNet) to segment adrenal glands in CT images. Our MCNet mainly consists of two components, i.e., the multi-level context aggregation (MCA) module and multi-level context guidance (MCG) module. Specifically, the MCA module employs multi-branch dilated convolutional layers to capture geometric information, which enables handling of changes in complex scenarios such as variations in the size and shape of objects. The MCG module, on the other hand, gathers valuable features from the shallow layer and leverages the complete utilization of feature information at different resolutions in various codec stages. Finally, we evaluate the performance of the MCNet on two CT datasets, including our clinical dataset (Ad-Seg) and a publicly available dataset known as Distorted Golden Standards (DGS), from different perspectives. Compared to ten other state-of-the-art segmentation methods, our MCNet achieves 71.34% and 75.29% of the best Dice similarity coefficient on the two datasets, respectively, which is at least 2.46% and 1.19% higher than other segmentation methods.
Collapse
Affiliation(s)
- Jinhao Li
- Key Laboratory of Intelligent Computing and Information Processing of Ministry of Education, Xiangtan University, Xiangtan 411105, China
| | - Huying Li
- Key Laboratory of Intelligent Computing and Information Processing of Ministry of Education, Xiangtan University, Xiangtan 411105, China
| | - Yuan Zhang
- Key Laboratory of Intelligent Computing and Information Processing of Ministry of Education, Xiangtan University, Xiangtan 411105, China
| | - Zhiqiang Wang
- Key Laboratory of Medical Imaging and Artificial Intelligence of Hunan Province, Xiangnan University, Chenzhou 423000, China; College of Medical Imaging Laboratory and Rehabilitation, Xiangnan University, Chenzhou 423000, China.
| | - Sheng Zhu
- Department of Nuclear Medicine, Affiliated Hospital of Xiangnan University, Chenzhou 423000, China
| | | | - Kai Hu
- Key Laboratory of Intelligent Computing and Information Processing of Ministry of Education, Xiangtan University, Xiangtan 411105, China; Key Laboratory of Medical Imaging and Artificial Intelligence of Hunan Province, Xiangnan University, Chenzhou 423000, China.
| | - Xieping Gao
- Hunan Provincial Key Laboratory of Intelligent Computing and Language Information Processing, Hunan Normal University, Changsha 410081, China
| |
Collapse
|
18
|
Wang C, Wang L, Wang N, Wei X, Feng T, Wu M, Yao Q, Zhang R. CFATransUnet: Channel-wise cross fusion attention and transformer for 2D medical image segmentation. Comput Biol Med 2024; 168:107803. [PMID: 38064854 DOI: 10.1016/j.compbiomed.2023.107803] [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: 07/29/2023] [Revised: 11/23/2023] [Accepted: 11/29/2023] [Indexed: 01/10/2024]
Abstract
Medical image segmentation faces current challenges in effectively extracting and fusing long-distance and local semantic information, as well as mitigating or eliminating semantic gaps during the encoding and decoding process. To alleviate the above two problems, we propose a new U-shaped network structure, called CFATransUnet, with Transformer and CNN blocks as the backbone network, equipped with Channel-wise Cross Fusion Attention and Transformer (CCFAT) module, containing Channel-wise Cross Fusion Transformer (CCFT) and Channel-wise Cross Fusion Attention (CCFA). Specifically, we use a Transformer and CNN blocks to construct the encoder and decoder for adequate extraction and fusion of long-range and local semantic features. The CCFT module utilizes the self-attention mechanism to reintegrate semantic information from different stages into cross-level global features to reduce the semantic asymmetry between features at different levels. The CCFA module adaptively acquires the importance of each feature channel based on a global perspective in a network learning manner, enhancing effective information grasping and suppressing non-important features to mitigate semantic gaps. The combination of CCFT and CCFA can guide the effective fusion of different levels of features more powerfully with a global perspective. The consistent architecture of the encoder and decoder also alleviates the semantic gap. Experimental results suggest that the proposed CFATransUnet achieves state-of-the-art performance on four datasets. The code is available at https://github.com/CPU0808066/CFATransUnet.
Collapse
Affiliation(s)
- Cheng Wang
- Department of Optical Science and Engineering, Fudan University, Shanghai 200433, China
| | - Le Wang
- Academy for Engineering and Technology, Fudan University, Shanghai 200433, China
| | - Nuoqi Wang
- Department of Optical Science and Engineering, Fudan University, Shanghai 200433, China
| | - Xiaoling Wei
- Department of Endodontics, Shanghai Stomatological Hospital, Fudan University, Shanghai 200001, China
| | - Ting Feng
- Academy for Engineering and Technology, Fudan University, Shanghai 200433, China
| | - Minfeng Wu
- Department of Dermatology, Huadong Hospital Affiliated to Fudan University, Shanghai, 200040, China.
| | - Qi Yao
- Academy for Engineering and Technology, Fudan University, Shanghai 200433, China.
| | - Rongjun Zhang
- Department of Optical Science and Engineering, Fudan University, Shanghai 200433, China; Academy for Engineering and Technology, Fudan University, Shanghai 200433, China; Zhuhai Fudan Innovation Institute, Zhuhai 519031, China.
| |
Collapse
|
19
|
Liao W, Luo X, He Y, Dong Y, Li C, Li K, Zhang S, Zhang S, Wang G, Xiao J. Comprehensive Evaluation of a Deep Learning Model for Automatic Organs-at-Risk Segmentation on Heterogeneous Computed Tomography Images for Abdominal Radiation Therapy. Int J Radiat Oncol Biol Phys 2023; 117:994-1006. [PMID: 37244625 DOI: 10.1016/j.ijrobp.2023.05.034] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 11/08/2022] [Revised: 03/13/2023] [Accepted: 05/18/2023] [Indexed: 05/29/2023]
Abstract
PURPOSE Our purpose was to develop a deep learning model (AbsegNet) that produces accurate contours of 16 organs at risk (OARs) for abdominal malignancies as an essential part of fully automated radiation treatment planning. METHODS AND MATERIALS Three data sets with 544 computed tomography scans were retrospectively collected. Data set 1 was split into 300 training cases and 128 test cases (cohort 1) for AbsegNet. Data set 2, including cohort 2 (n = 24) and cohort 3 (n = 20), were used to validate AbsegNet externally. Data set 3, including cohort 4 (n = 40) and cohort 5 (n = 32), were used to clinically assess the accuracy of AbsegNet-generated contours. Each cohort was from a different center. The Dice similarity coefficient and 95th-percentile Hausdorff distance were calculated to evaluate the delineation quality for each OAR. Clinical accuracy evaluation was classified into 4 levels: no revision, minor revisions (0% < volumetric revision degrees [VRD] ≤ 10%), moderate revisions (10% ≤ VRD < 20%), and major revisions (VRD ≥20%). RESULTS For all OARs, AbsegNet achieved a mean Dice similarity coefficient of 86.73%, 85.65%, and 88.04% in cohorts 1, 2, and 3, respectively, and a mean 95th-percentile Hausdorff distance of 8.92, 10.18, and 12.40 mm, respectively. The performance of AbsegNet outperformed SwinUNETR, DeepLabV3+, Attention-UNet, UNet, and 3D-UNet. When experts evaluated contours from cohorts 4 and 5, 4 OARs (liver, kidney_L, kidney_R, and spleen) of all patients were scored as having no revision, and over 87.5% of patients with contours of the stomach, esophagus, adrenals, or rectum were considered as having no or minor revisions. Only 15.0% of patients with colon and small bowel contours required major revisions. CONCLUSIONS We propose a novel deep-learning model to delineate OARs on diverse data sets. Most contours produced by AbsegNet are accurate and robust and are, therefore, clinically applicable and helpful to facilitate radiation therapy workflow.
Collapse
Affiliation(s)
- Wenjun Liao
- Department of Radiation Oncology, Radiation Oncology Key Laboratory of Sichuan Province, Sichuan Clinical Research Center for Cancer, Sichuan Cancer Hospital & Institute, Sichuan Cancer Center, Affiliated Cancer Hospital of University of Electronic Science and Technology of China, Chengdu, China
| | - Xiangde Luo
- School of Mechanical and Electrical Engineering, University of Electronic Science and Technology of China, Chengdu, China; Shanghai AI Laboratory, Shanghai, China
| | - Yuan He
- Department of Radiation Oncology, The First Affiliated Hospital of USTC, Division of Life Sciences and Medicine, University of Science and Technology of China, Hefei, Anhui, China
| | - Ye Dong
- Department of NanFang PET Center, Nanfang Hospital, Southern Medical University, Guangzhou, China
| | - Churong Li
- Department of Radiation Oncology, Radiation Oncology Key Laboratory of Sichuan Province, Sichuan Clinical Research Center for Cancer, Sichuan Cancer Hospital & Institute, Sichuan Cancer Center, Affiliated Cancer Hospital of University of Electronic Science and Technology of China, Chengdu, China
| | - Kang Li
- West China Biomedical Big Data Center
| | - Shichuan Zhang
- Department of Radiation Oncology, Radiation Oncology Key Laboratory of Sichuan Province, Sichuan Clinical Research Center for Cancer, Sichuan Cancer Hospital & Institute, Sichuan Cancer Center, Affiliated Cancer Hospital of University of Electronic Science and Technology of China, Chengdu, China
| | - Shaoting Zhang
- School of Mechanical and Electrical Engineering, University of Electronic Science and Technology of China, Chengdu, China; Shanghai AI Laboratory, Shanghai, China
| | - Guotai Wang
- School of Mechanical and Electrical Engineering, University of Electronic Science and Technology of China, Chengdu, China; Shanghai AI Laboratory, Shanghai, China
| | - Jianghong Xiao
- Radiotherapy Physics & Technology Center, Department of Radiation Oncology, Cancer Center, West China Hospital, Sichuan University, Chengdu, China.
| |
Collapse
|
20
|
Bongratz F, Rickmann AM, Wachinger C. Abdominal organ segmentation via deep diffeomorphic mesh deformations. Sci Rep 2023; 13:18270. [PMID: 37880251 PMCID: PMC10600339 DOI: 10.1038/s41598-023-45435-2] [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: 08/02/2023] [Accepted: 10/19/2023] [Indexed: 10/27/2023] Open
Abstract
Abdominal organ segmentation from CT and MRI is an essential prerequisite for surgical planning and computer-aided navigation systems. It is challenging due to the high variability in the shape, size, and position of abdominal organs. Three-dimensional numeric representations of abdominal shapes with point-wise correspondence to a template are further important for quantitative and statistical analyses thereof. Recently, template-based surface extraction methods have shown promising advances for direct mesh reconstruction from volumetric scans. However, the generalization of these deep learning-based approaches to different organs and datasets, a crucial property for deployment in clinical environments, has not yet been assessed. We close this gap and employ template-based mesh reconstruction methods for joint liver, kidney, pancreas, and spleen segmentation. Our experiments on manually annotated CT and MRI data reveal limited generalization capabilities of previous methods to organs of different geometry and weak performance on small datasets. We alleviate these issues with a novel deep diffeomorphic mesh-deformation architecture and an improved training scheme. The resulting method, UNetFlow, generalizes well to all four organs and can be easily fine-tuned on new data. Moreover, we propose a simple registration-based post-processing that aligns voxel and mesh outputs to boost segmentation accuracy.
Collapse
Affiliation(s)
- Fabian Bongratz
- Department of Radiology, Technical University of Munich, Munich, 81675, Germany.
- Munich Center for Machine Learning, Munich, Germany.
| | - Anne-Marie Rickmann
- Department of Radiology, Technical University of Munich, Munich, 81675, Germany
- Department of Child and Adolescent Psychiatry, Ludwig-Maximilians-University, Munich, 80336, Germany
| | - Christian Wachinger
- Department of Radiology, Technical University of Munich, Munich, 81675, Germany
- Department of Child and Adolescent Psychiatry, Ludwig-Maximilians-University, Munich, 80336, Germany
- Munich Center for Machine Learning, Munich, Germany
| |
Collapse
|
21
|
Zou L, Cai Z, Qiu Y, Gui L, Mao L, Yang X. CTG-Net: an efficient cascaded framework driven by terminal guidance mechanism for dilated pancreatic duct segmentation. Phys Med Biol 2023; 68:215006. [PMID: 37586389 DOI: 10.1088/1361-6560/acf110] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 05/24/2023] [Accepted: 08/16/2023] [Indexed: 08/18/2023]
Abstract
Pancreatic duct dilation indicates a high risk of various pancreatic diseases. Segmentation for dilated pancreatic duct (DPD) on computed tomography (CT) image shows the potential to assist the early diagnosis, surgical planning and prognosis. Because of the DPD's tiny size, slender tubular structure and the surrounding distractions, most current researches on DPD segmentation achieve low accuracy and always have segmentation errors on the terminal DPD regions. To address these problems, we propose a cascaded terminal guidance network to efficiently improve the DPD segmentation performance. Firstly, a basic cascaded segmentation architecture is established to get the pancreas and coarse DPD segmentation, a DPD graph structure is build on the coarse DPD segmentation to locate the terminal DPD regions. Then, a terminal anatomy attention module is introduced for jointly learning the local intensity from the CT images, feature cues from the coarse DPD segmentation and global anatomy information from the designed pancreas anatomy-aware maps. Finally, a terminal distraction attention module which explicitly learns the distribution of the terminal distraction regions is proposed to reduce the false positive and false negative predictions. We also propose a new metric called tDice to measure the terminal segmentation accuracy for targets with tubular structures and two other metrics for segmentation error evaluation. We collect our dilated pancreatic duct segmentation dataset with 150 CT scans from patients with five types of pancreatic tumors. Experimental results on our dataset show that our proposed approach boosts DPD segmentation accuracy by nearly 20% compared with the existing results, and achieves more than 9% improvement for the terminal segmentation accuracy compared with the state-of-the-art methods.
Collapse
Affiliation(s)
- Liwen Zou
- Department of Mathematics, Nanjing University, Nanjing, 210093, People's Republic of China
| | - Zhenghua Cai
- Medical School, Nanjing University, Nanjing, 210007, People's Republic of China
| | - Yudong Qiu
- Department of General Surgery, Nanjing Drum Tower Hospital, Nanjing, 210008, People's Republic of China
| | - Luying Gui
- School of Mathematics and Statistics, Nanjing University of Science and Technology, Nanjing, 210094, People's Republic of China
| | - Liang Mao
- Department of General Surgery, Nanjing Drum Tower Hospital, Nanjing, 210008, People's Republic of China
| | - Xiaoping Yang
- Department of Mathematics, Nanjing University, Nanjing, 210093, People's Republic of China
| |
Collapse
|
22
|
Wang T, Liu X, Zhang C, He Y, Chan Y, Xie Y, Liang X. Ring artifacts correction for computed tomography image using unsupervised contrastive learning. Phys Med Biol 2023; 68:205008. [PMID: 37714184 DOI: 10.1088/1361-6560/acfa60] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 04/17/2023] [Accepted: 09/15/2023] [Indexed: 09/17/2023]
Abstract
Objective.Computed tomography (CT) is a widely employed imaging technology for disease detection. However, CT images often suffer from ring artifacts, which may result from hardware defects and other factors. These artifacts compromise image quality and impede diagnosis. To address this challenge, we propose a novel method based on dual contrast learning image style transformation network model (DCLGAN) that effectively eliminates ring artifacts from CT images while preserving texture details.Approach. Our method involves simulating ring artifacts on real CT data to generate the uncorrected CT (uCT) data and transforming them into strip artifacts. Subsequently, the DCLGAN synthetic network is applied in the polar coordinate system to remove the strip artifacts and generate a synthetic CT (sCT). We compare the uCT and sCT images to obtain a residual image, which is then filtered to extract the strip artifacts. An inverse polar transformation is performed to obtain the ring artifacts, which are subtracted from the original CT image to produce a corrected image.Main results.To validate the effectiveness of our approach, we tested it using real CT data, simulated data, and cone beam computed tomography images of the patient's brain. The corrected CT images showed a reduction in mean absolute error by 12.36 Hounsfield units (HU), a decrease in root mean square error by 18.94 HU, an increase in peak signal-to-noise ratio by 3.53 decibels (dB), and an improvement in structural similarity index by 9.24%.Significance.These results demonstrate the efficacy of our method in eliminating ring artifacts and preserving image details, making it a valuable tool for CT imaging.
Collapse
Affiliation(s)
- Tangsheng Wang
- Shenzhen Institute of Advanced Technology, Chinese Academy of Sciences, Shenzhen, Guangdong 518055, People's Republic of China
- University of Chinese Academy of Sciences, Beijing 100190, People's Republic of China
| | - Xuan Liu
- Shenzhen Institute of Advanced Technology, Chinese Academy of Sciences, Shenzhen, Guangdong 518055, People's Republic of China
| | - Chulong Zhang
- Shenzhen Institute of Advanced Technology, Chinese Academy of Sciences, Shenzhen, Guangdong 518055, People's Republic of China
| | - Yutong He
- Shenzhen Institute of Advanced Technology, Chinese Academy of Sciences, Shenzhen, Guangdong 518055, People's Republic of China
| | - Yinping Chan
- Shenzhen Institute of Advanced Technology, Chinese Academy of Sciences, Shenzhen, Guangdong 518055, People's Republic of China
| | - Yaoqin Xie
- Shenzhen Institute of Advanced Technology, Chinese Academy of Sciences, Shenzhen, Guangdong 518055, People's Republic of China
| | - Xiaokun Liang
- Shenzhen Institute of Advanced Technology, Chinese Academy of Sciences, Shenzhen, Guangdong 518055, People's Republic of China
| |
Collapse
|
23
|
Lee HH, Tang Y, Yang Q, Yu X, Cai LY, Remedios LW, Bao S, Landman BA, Huo Y. Semantic-Aware Contrastive Learning for Multi-Object Medical Image Segmentation. IEEE J Biomed Health Inform 2023; 27:4444-4453. [PMID: 37310834 PMCID: PMC10524443 DOI: 10.1109/jbhi.2023.3285230] [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] [Indexed: 06/15/2023]
Abstract
Medical image segmentation, or computing voxel-wise semantic masks, is a fundamental yet challenging task in medical imaging domain. To increase the ability of encoder-decoder neural networks to perform this task across large clinical cohorts, contrastive learning provides an opportunity to stabilize model initialization and enhances downstream tasks performance without ground-truth voxel-wise labels. However, multiple target objects with different semantic meanings and contrast level may exist in a single image, which poses a problem for adapting traditional contrastive learning methods from prevalent "image-level classification" to "pixel-level segmentation". In this article, we propose a simple semantic-aware contrastive learning approach leveraging attention masks and image-wise labels to advance multi-object semantic segmentation. Briefly, we embed different semantic objects to different clusters rather than the traditional image-level embeddings. We evaluate our proposed method on a multi-organ medical image segmentation task with both in-house data and MICCAI Challenge 2015 BTCV datasets. Compared with current state-of-the-art training strategies, our proposed pipeline yields a substantial improvement of 5.53% and 6.09% on Dice score for both medical image segmentation cohorts respectively (p-value 0.01). The performance of the proposed method is further assessed on external medical image cohort via MICCAI Challenge FLARE 2021 dataset, and achieves a substantial improvement from Dice 0.922 to 0.933 (p-value 0.01).
Collapse
|
24
|
Zhao Q, Zhong L, Xiao J, Zhang J, Chen Y, Liao W, Zhang S, Wang G. Efficient Multi-Organ Segmentation From 3D Abdominal CT Images With Lightweight Network and Knowledge Distillation. IEEE TRANSACTIONS ON MEDICAL IMAGING 2023; 42:2513-2523. [PMID: 37030798 DOI: 10.1109/tmi.2023.3262680] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [MESH Headings] [Track Full Text] [Subscribe] [Scholar Register] [Indexed: 06/19/2023]
Abstract
Accurate segmentation of multiple abdominal organs from Computed Tomography (CT) images plays an important role in computer-aided diagnosis, treatment planning and follow-up. Currently, 3D Convolution Neural Networks (CNN) have achieved promising performance for automatic medical image segmentation tasks. However, most existing 3D CNNs have a large set of parameters and huge floating point operations (FLOPs), and 3D CT volumes have a large size, leading to high computational cost, which limits their clinical application. To tackle this issue, we propose a novel framework based on lightweight network and Knowledge Distillation (KD) for delineating multiple organs from 3D CT volumes. We first propose a novel lightweight medical image segmentation network named LCOV-Net for reducing the model size and then introduce two knowledge distillation modules (i.e., Class-Affinity KD and Multi-Scale KD) to effectively distill the knowledge from a heavy-weight teacher model to improve LCOV-Net's segmentation accuracy. Experiments on two public abdominal CT datasets for multiple organ segmentation showed that: 1) Our LCOV-Net outperformed existing lightweight 3D segmentation models in both computational cost and accuracy; 2) The proposed KD strategy effectively improved the performance of the lightweight network, and it outperformed existing KD methods; 3) Combining the proposed LCOV-Net and KD strategy, our framework achieved better performance than the state-of-the-art 3D nnU-Net with only one-fifth parameters. The code is available at https://github.com/HiLab-git/LCOVNet-and-KD.
Collapse
|
25
|
Azuri I, Wattad A, Peri-Hanania K, Kashti T, Rosen R, Caspi Y, Istaiti M, Wattad M, Applbaum Y, Zimran A, Revel-Vilk S, C. Eldar Y. A Deep-Learning Approach to Spleen Volume Estimation in Patients with Gaucher Disease. J Clin Med 2023; 12:5361. [PMID: 37629403 PMCID: PMC10455264 DOI: 10.3390/jcm12165361] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 07/09/2023] [Revised: 08/04/2023] [Accepted: 08/10/2023] [Indexed: 08/27/2023] Open
Abstract
The enlargement of the liver and spleen (hepatosplenomegaly) is a common manifestation of Gaucher disease (GD). An accurate estimation of the liver and spleen volumes in patients with GD, using imaging tools such as magnetic resonance imaging (MRI), is crucial for the baseline assessment and monitoring of the response to treatment. A commonly used method in clinical practice to estimate the spleen volume is the employment of a formula that uses the measurements of the craniocaudal length, diameter, and thickness of the spleen in MRI. However, the inaccuracy of this formula is significant, which, in turn, emphasizes the need for a more precise and reliable alternative. To this end, we employed deep-learning techniques, to achieve a more accurate spleen segmentation and, subsequently, calculate the resulting spleen volume with higher accuracy on a testing set cohort of 20 patients with GD. Our results indicate that the mean error obtained using the deep-learning approach to spleen volume estimation is 3.6 ± 2.7%, which is significantly lower than the common formula approach, which resulted in a mean error of 13.9 ± 9.6%. These findings suggest that the integration of deep-learning methods into the clinical routine practice for spleen volume calculation could lead to improved diagnostic and monitoring outcomes.
Collapse
Affiliation(s)
- Ido Azuri
- Bioinformatics Unit, Department of Life Sciences Core Facilities, Weizmann Institute of Science, Rehovot 7610001, Israel
| | - Ameer Wattad
- Department of Radiology, Shaare Zedek Medical Center, Jerusalem 9103102, Israel
| | - Keren Peri-Hanania
- Department of Computer Science and Applied Mathematics, Weizmann Institute of Science, Rehovot 7610001, Israel
| | - Tamar Kashti
- Department of Computer Science and Applied Mathematics, Weizmann Institute of Science, Rehovot 7610001, Israel
| | - Ronnie Rosen
- Department of Computer Science and Applied Mathematics, Weizmann Institute of Science, Rehovot 7610001, Israel
| | - Yaron Caspi
- Department of Computer Science and Applied Mathematics, Weizmann Institute of Science, Rehovot 7610001, Israel
| | - Majdolen Istaiti
- Gaucher Unit, Shaare Zedek Medical Center, Jerusalem 9103102, Israel
| | - Makram Wattad
- Department of Radiology, Shaare Zedek Medical Center, Jerusalem 9103102, Israel
| | - Yaakov Applbaum
- Department of Radiology, Shaare Zedek Medical Center, Jerusalem 9103102, Israel
- Faculty of Medicine, Hebrew University, Jerusalem 9112102, Israel
| | - Ari Zimran
- Gaucher Unit, Shaare Zedek Medical Center, Jerusalem 9103102, Israel
- Faculty of Medicine, Hebrew University, Jerusalem 9112102, Israel
| | - Shoshana Revel-Vilk
- Gaucher Unit, Shaare Zedek Medical Center, Jerusalem 9103102, Israel
- Faculty of Medicine, Hebrew University, Jerusalem 9112102, Israel
| | - Yonina C. Eldar
- Department of Computer Science and Applied Mathematics, Weizmann Institute of Science, Rehovot 7610001, Israel
| |
Collapse
|
26
|
Dan Y, Jin W, Wang Z, Sun C. Optimization of U-shaped pure transformer medical image segmentation network. PeerJ Comput Sci 2023; 9:e1515. [PMID: 37705654 PMCID: PMC10495965 DOI: 10.7717/peerj-cs.1515] [Citation(s) in RCA: 1] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 04/17/2023] [Accepted: 07/13/2023] [Indexed: 09/15/2023]
Abstract
In recent years, neural networks have made pioneering achievements in the field of medical imaging. In particular, deep neural networks based on U-shaped structures are widely used in different medical image segmentation tasks. In order to improve the early diagnosis and clinical decision-making system of lung diseases, it has become a key step to use the neural network for lung segmentation to assist in positioning and observing the shape. There is still the problem of low precision. For the sake of achieving better segmentation accuracy, an optimized pure Transformer U-shaped segmentation is proposed in this article. The optimization segmentation network adopts the method of adding skip connections and performing special splicing processing, which reduces the information loss in the encoding process and increases the information in the decoding process, so as to achieve the purpose of improving the segmentation accuracy. The final experiment shows that our improved network achieves 97.86% accuracy in segmentation of the "Chest Xray Masks and Labels" dataset, which is better than the full convolutional network or the combination of Transformer and convolution.
Collapse
Affiliation(s)
- Yongping Dan
- School of Electronic and Information, Zhongyuan University of Technology, Zhengzhou, Henan, China
| | - Weishou Jin
- School of Electronic and Information, Zhongyuan University of Technology, Zhengzhou, Henan, China
| | - Zhida Wang
- School of Electronic and Information, Zhongyuan University of Technology, Zhengzhou, Henan, China
| | - Changhao Sun
- School of Electronic and Information, Zhongyuan University of Technology, Zhengzhou, Henan, China
| |
Collapse
|
27
|
Gottlich HC, Gregory AV, Sharma V, Khanna A, Moustafa AU, Lohse CM, Potretzke TA, Korfiatis P, Potretzke AM, Denic A, Rule AD, Takahashi N, Erickson BJ, Leibovich BC, Kline TL. Effect of Dataset Size and Medical Image Modality on Convolutional Neural Network Model Performance for Automated Segmentation: A CT and MR Renal Tumor Imaging Study. J Digit Imaging 2023; 36:1770-1781. [PMID: 36932251 PMCID: PMC10406754 DOI: 10.1007/s10278-023-00804-1] [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: 09/11/2022] [Revised: 02/28/2023] [Accepted: 03/01/2023] [Indexed: 03/19/2023] Open
Abstract
The aim of this study is to investigate the use of an exponential-plateau model to determine the required training dataset size that yields the maximum medical image segmentation performance. CT and MR images of patients with renal tumors acquired between 1997 and 2017 were retrospectively collected from our nephrectomy registry. Modality-based datasets of 50, 100, 150, 200, 250, and 300 images were assembled to train models with an 80-20 training-validation split evaluated against 50 randomly held out test set images. A third experiment using the KiTS21 dataset was also used to explore the effects of different model architectures. Exponential-plateau models were used to establish the relationship of dataset size to model generalizability performance. For segmenting non-neoplastic kidney regions on CT and MR imaging, our model yielded test Dice score plateaus of [Formula: see text] and [Formula: see text] with the number of training-validation images needed to reach the plateaus of 54 and 122, respectively. For segmenting CT and MR tumor regions, we modeled a test Dice score plateau of [Formula: see text] and [Formula: see text], with 125 and 389 training-validation images needed to reach the plateaus. For the KiTS21 dataset, the best Dice score plateaus for nn-UNet 2D and 3D architectures were [Formula: see text] and [Formula: see text] with number to reach performance plateau of 177 and 440. Our research validates that differing imaging modalities, target structures, and model architectures all affect the amount of training images required to reach a performance plateau. The modeling approach we developed will help future researchers determine for their experiments when additional training-validation images will likely not further improve model performance.
Collapse
Affiliation(s)
| | - Adriana V Gregory
- Division of Nephrology and Hypertension, Mayo Clinic, Rochester, MN, USA
| | - Vidit Sharma
- Department of Urology, Mayo Clinic, Rochester, MN, USA
| | | | - Amr U Moustafa
- Division of Nephrology and Hypertension, Mayo Clinic, Rochester, MN, USA
| | - Christine M Lohse
- Division of Biomedical Statistics and Informatics, Mayo Clinic, Rochester, MN, USA
| | | | | | | | - Aleksandar Denic
- Division of Nephrology and Hypertension, Mayo Clinic, Rochester, MN, USA
| | - Andrew D Rule
- Division of Nephrology and Hypertension, Mayo Clinic, Rochester, MN, USA
| | | | | | | | | |
Collapse
|
28
|
Automatic abdominal segmentation using novel 3D self-adjustable organ aware deep network in CT images. Biomed Signal Process Control 2023. [DOI: 10.1016/j.bspc.2023.104691] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 02/27/2023]
|
29
|
Fogarollo S, Bale R, Harders M. Towards liver segmentation in the wild via contrastive distillation. Int J Comput Assist Radiol Surg 2023; 18:1143-1149. [PMID: 37145251 PMCID: PMC10329587 DOI: 10.1007/s11548-023-02912-3] [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: 03/09/2023] [Accepted: 04/06/2023] [Indexed: 05/06/2023]
Abstract
PURPOSE Automatic liver segmentation is a key component for performing computer-assisted hepatic procedures. The task is challenging due to the high variability in organ appearance, numerous imaging modalities, and limited availability of labels. Moreover, strong generalization performance is required in real-world scenarios. However, existing supervised methods cannot be applied to data not seen during training (i.e. in the wild) because they generalize poorly. METHODS We propose to distill knowledge from a powerful model with our novel contrastive distillation scheme. We use a pre-trained large neural network to train our smaller model. A key novelty is to map neighboring slices close together in the latent representation, while mapping distant slices far away. Then, we use ground-truth labels to learn a U-Net style upsampling path and recover the segmentation map. RESULTS The pipeline is proven to be robust enough to perform state-of-the-art inference on target unseen domains. We carried out an extensive experimental validation using six common abdominal datasets, covering multiple modalities, as well as 18 patient datasets from the Innsbruck University Hospital. A sub-second inference time and a data-efficient training pipeline make it possible to scale our method to real-world conditions. CONCLUSION We propose a novel contrastive distillation scheme for automatic liver segmentation. A limited set of assumptions and superior performance to state-of-the-art techniques make our method a candidate for application to real-world scenarios.
Collapse
Affiliation(s)
- Stefano Fogarollo
- Department of Computer Science Interactive Graphics and Simulation Group (IGS), University of Innsbruck, Innsbruck, Austria.
| | - Reto Bale
- Interventional Oncology-Microinvasive Therapy (SIP), Department of Radiology, Medical University Innsbruck, Innsbruck, Austria
| | - Matthias Harders
- Department of Computer Science Interactive Graphics and Simulation Group (IGS), University of Innsbruck, Innsbruck, Austria
| |
Collapse
|
30
|
Tong N, Xu Y, Zhang J, Gou S, Li M. Robust and efficient abdominal CT segmentation using shape constrained multi-scale attention network. Phys Med 2023; 110:102595. [PMID: 37178624 DOI: 10.1016/j.ejmp.2023.102595] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 09/22/2022] [Revised: 03/02/2023] [Accepted: 04/17/2023] [Indexed: 05/15/2023] Open
Abstract
PURPOSE Although many deep learning-based abdominal multi-organ segmentation networks have been proposed, the various intensity distributions and organ shapes of the CT images from multi-center, multi-phase with various diseases introduce new challenges for robust abdominal CT segmentation. To achieve robust and efficient abdominal multi-organ segmentation, a new two-stage method is presented in this study. METHODS A binary segmentation network is used for coarse localization, followed by a multi-scale attention network for the fine segmentation of liver, kidney, spleen, and pancreas. To constrain the organ shapes produced by the fine segmentation network, an additional network is pre-trained to learn the shape features of the organs with serious diseases and then employed to constrain the training of the fine segmentation network. RESULTS The performance of the presented segmentation method was extensively evaluated on the multi-center data set from the Fast and Low GPU Memory Abdominal oRgan sEgmentation (FLARE) challenge, which was held in conjunction with International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI) 2021. Dice Similarity Coefficient (DSC) and Normalized Surface Dice (NSD) were calculated to quantitatively evaluate the segmentation accuracy and efficiency. An average DSC and NSD of 83.7% and 64.4% were achieved, and our method finally won the second place among more than 90 participating teams. CONCLUSIONS The evaluation results on the public challenge demonstrate that our method shows promising performance in robustness and efficiency, which may promote the clinical application of the automatic abdominal multi-organ segmentation.
Collapse
Affiliation(s)
- Nuo Tong
- AI-based Big Medical Imaging Data Frontier Research Center, Academy of Advanced Interdisciplinary Research, Xidian University, Xi'an, Shaanxi 710071, China
| | - Yinan Xu
- Key Lab of Intelligent Perception and Image Understanding of Ministry of Education, Xidian University, Xi'an, Shaanxi, 710071, China
| | - Jinsong Zhang
- Xijing Hospital of Air Force Military Medical University, Xian, Shaanxi 710032, China
| | - Shuiping Gou
- AI-based Big Medical Imaging Data Frontier Research Center, Academy of Advanced Interdisciplinary Research, Xidian University, Xi'an, Shaanxi 710071, China; Key Lab of Intelligent Perception and Image Understanding of Ministry of Education, Xidian University, Xi'an, Shaanxi, 710071, China.
| | - Mengbin Li
- Xijing Hospital of Air Force Military Medical University, Xian, Shaanxi 710032, China.
| |
Collapse
|
31
|
Song Y, Yu L, Lei B, Choi KS, Qin J. Data Discernment for Affordable Training in Medical Image Segmentation. IEEE TRANSACTIONS ON MEDICAL IMAGING 2023; 42:1431-1445. [PMID: 37015694 DOI: 10.1109/tmi.2022.3228316] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Track Full Text] [Subscribe] [Scholar Register] [Indexed: 05/04/2023]
Abstract
Collecting sufficient high-quality training data for deep neural networks is often expensive or even unaffordable in medical image segmentation tasks. We thus propose to train the network by using external data that can be collected in a cheaper way, e.g., crowd-sourcing. We show that by data discernment, the network is able to mine valuable knowledge from external data, even though the data distribution is very different from that of the original (internal) data. We discern the external data by learning an importance weight for each of them, with the goal to enhance the contribution of informative external data to network updating, while suppressing the data that are 'useless' or even 'harmful'. An iterative algorithm that alternatively estimates the importance weight and updates the network is developed by formulating the data discernment as a constrained nonlinear programming problem. It estimates the importance weight according to the distribution discrepancy between the external data and the internal dataset, and imposes a constraint to drive the network to learn more effectively, compared with the network without using the external data. We evaluate the proposed algorithm on two tasks: abdominal CT image and cervical smear image segmentation, using totally 6 publicly available datasets. The effectiveness of the algorithm is demonstrated by extensive experiments. Source codes are available at: https://github.com/YouyiSong/Data-Discernment.
Collapse
|
32
|
Uncertainty-guided mutual consistency learning for semi-supervised medical image segmentation. Artif Intell Med 2022; 138:102476. [PMID: 36990583 DOI: 10.1016/j.artmed.2022.102476] [Citation(s) in RCA: 5] [Impact Index Per Article: 2.5] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 12/20/2021] [Revised: 11/20/2022] [Accepted: 12/13/2022] [Indexed: 12/23/2022]
Abstract
Medical image segmentation is a fundamental and critical step in many clinical approaches. Semi-supervised learning has been widely applied to medical image segmentation tasks since it alleviates the heavy burden of acquiring expert-examined annotations and takes the advantage of unlabeled data which is much easier to acquire. Although consistency learning has been proven to be an effective approach by enforcing an invariance of predictions under different distributions, existing approaches cannot make full use of region-level shape constraint and boundary-level distance information from unlabeled data. In this paper, we propose a novel uncertainty-guided mutual consistency learning framework to effectively exploit unlabeled data by integrating intra-task consistency learning from up-to-date predictions for self-ensembling and cross-task consistency learning from task-level regularization to exploit geometric shape information. The framework is guided by the estimated segmentation uncertainty of models to select out relatively certain predictions for consistency learning, so as to effectively exploit more reliable information from unlabeled data. Experiments on two publicly available benchmark datasets showed that: (1) Our proposed method can achieve significant performance improvement by leveraging unlabeled data, with up to 4.13% and 9.82% in Dice coefficient compared to supervised baseline on left atrium segmentation and brain tumor segmentation, respectively. (2) Compared with other semi-supervised segmentation methods, our proposed method achieve better segmentation performance under the same backbone network and task settings on both datasets, demonstrating the effectiveness and robustness of our method and potential transferability for other medical image segmentation tasks.
Collapse
|
33
|
Huang SY, Hsu WL, Hsu RJ, Liu DW. Fully Convolutional Network for the Semantic Segmentation of Medical Images: A Survey. Diagnostics (Basel) 2022; 12:diagnostics12112765. [PMID: 36428824 PMCID: PMC9689961 DOI: 10.3390/diagnostics12112765] [Citation(s) in RCA: 3] [Impact Index Per Article: 1.5] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 09/23/2022] [Revised: 10/19/2022] [Accepted: 11/04/2022] [Indexed: 11/16/2022] Open
Abstract
There have been major developments in deep learning in computer vision since the 2010s. Deep learning has contributed to a wealth of data in medical image processing, and semantic segmentation is a salient technique in this field. This study retrospectively reviews recent studies on the application of deep learning for segmentation tasks in medical imaging and proposes potential directions for future development, including model development, data augmentation processing, and dataset creation. The strengths and deficiencies of studies on models and data augmentation, as well as their application to medical image segmentation, were analyzed. Fully convolutional network developments have led to the creation of the U-Net and its derivatives. Another noteworthy image segmentation model is DeepLab. Regarding data augmentation, due to the low data volume of medical images, most studies focus on means to increase the wealth of medical image data. Generative adversarial networks (GAN) increase data volume via deep learning. Despite the increasing types of medical image datasets, there is still a deficiency of datasets on specific problems, which should be improved moving forward. Considering the wealth of ongoing research on the application of deep learning processing to medical image segmentation, the data volume and practical clinical application problems must be addressed to ensure that the results are properly applied.
Collapse
Affiliation(s)
- Sheng-Yao Huang
- Institute of Medical Science, Tzu Chi University, Hualien 97071, Taiwan
- Department of Radiation Oncology, Hualien Tzu Chi General Hospital, Buddhist Tzu Chi Medical Foundation, Hualien 97071, Taiwan
| | - Wen-Lin Hsu
- Department of Radiation Oncology, Hualien Tzu Chi General Hospital, Buddhist Tzu Chi Medical Foundation, Hualien 97071, Taiwan
- Cancer Center, Hualien Tzu Chi Hospital, Buddhist Tzu Chi Medical Foundation, Hualien 97071, Taiwan
- School of Medicine, Tzu Chi University, Hualien 97071, Taiwan
| | - Ren-Jun Hsu
- Institute of Medical Science, Tzu Chi University, Hualien 97071, Taiwan
- Cancer Center, Hualien Tzu Chi Hospital, Buddhist Tzu Chi Medical Foundation, Hualien 97071, Taiwan
- School of Medicine, Tzu Chi University, Hualien 97071, Taiwan
- Correspondence: (R.-J.H.); (D.-W.L.); Tel. & Fax: +886-3-8561825 (R.-J.H. & D.-W.L.)
| | - Dai-Wei Liu
- Institute of Medical Science, Tzu Chi University, Hualien 97071, Taiwan
- Department of Radiation Oncology, Hualien Tzu Chi General Hospital, Buddhist Tzu Chi Medical Foundation, Hualien 97071, Taiwan
- Cancer Center, Hualien Tzu Chi Hospital, Buddhist Tzu Chi Medical Foundation, Hualien 97071, Taiwan
- School of Medicine, Tzu Chi University, Hualien 97071, Taiwan
- Correspondence: (R.-J.H.); (D.-W.L.); Tel. & Fax: +886-3-8561825 (R.-J.H. & D.-W.L.)
| |
Collapse
|
34
|
Hu W, Jiang H, Wang M. Flexible needle puncture path planning for liver tumors based on deep reinforcement learning. Phys Med Biol 2022; 67. [DOI: 10.1088/1361-6560/ac8fdd] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 02/19/2022] [Accepted: 09/06/2022] [Indexed: 11/12/2022]
Abstract
Abstract
Objective. Minimally invasive surgery has been widely adopted in the treatment of patients with liver tumors. In liver tumor puncture surgery, an image-guided ablation needle for puncture surgery, which first reaches a target tumor along a predetermined path, and then ablates the tumor or injects drugs near the tumor, is often used to reduce patient trauma, improving the safety of surgery operations and avoiding possible damage to large blood vessels and key organs. In this paper, a path planning method for computer tomography (CT) guided ablation needle in liver tumor puncture surgery is proposed. Approach. Given a CT volume containing abdominal organs, we first classify voxels and optimize the number of voxels to reduce volume rendering pressure, then we reconstruct a multi-scale 3D model of the liver and hepatic vessels. Secondly, multiple entry points of the surgical path are selected based on the strong and weak constraints of clinical puncture surgery through multi-agent reinforcement learning. We select the optimal needle entry point based on the length measurement. Then, through the incremental training of the double deep Q-learning network (DDQN), the transmission of network parameters from the small-scale environment to the larger-scale environment is accomplished, and the optimal surgical path with more optimized details is obtained. Main results. To avoid falling into local optimum in network training, improve both the convergence speed and performance of the network, and maximize the cumulative reward, we train the path planning network on different scales 3D reconstructed organ models, and validate our method on tumor samples from public datasets. The scores of human surgeons verified the clinical relevance of the proposed method. Significance. Our method can robustly provide the optimal puncture path of flexible needle for liver tumors, which is expected to provide a reference for surgeons’ preoperative planning.
Collapse
|
35
|
Ma J, Zhang Y, Gu S, An X, Wang Z, Ge C, Wang C, Zhang F, Wang Y, Xu Y, Gou S, Thaler F, Payer C, Štern D, Henderson EGA, McSweeney DM, Green A, Jackson P, McIntosh L, Nguyen QC, Qayyum A, Conze PH, Huang Z, Zhou Z, Fan DP, Xiong H, Dong G, Zhu Q, He J, Yang X. Fast and Low-GPU-memory abdomen CT organ segmentation: The FLARE challenge. Med Image Anal 2022; 82:102616. [PMID: 36179380 DOI: 10.1016/j.media.2022.102616] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 01/13/2022] [Revised: 06/26/2022] [Accepted: 09/02/2022] [Indexed: 11/27/2022]
Abstract
Automatic segmentation of abdominal organs in CT scans plays an important role in clinical practice. However, most existing benchmarks and datasets only focus on segmentation accuracy, while the model efficiency and its accuracy on the testing cases from different medical centers have not been evaluated. To comprehensively benchmark abdominal organ segmentation methods, we organized the first Fast and Low GPU memory Abdominal oRgan sEgmentation (FLARE) challenge, where the segmentation methods were encouraged to achieve high accuracy on the testing cases from different medical centers, fast inference speed, and low GPU memory consumption, simultaneously. The winning method surpassed the existing state-of-the-art method, achieving a 19× faster inference speed and reducing the GPU memory consumption by 60% with comparable accuracy. We provide a summary of the top methods, make their code and Docker containers publicly available, and give practical suggestions on building accurate and efficient abdominal organ segmentation models. The FLARE challenge remains open for future submissions through a live platform for benchmarking further methodology developments at https://flare.grand-challenge.org/.
Collapse
Affiliation(s)
- Jun Ma
- Department of Mathematics, Nanjing University of Science and Technology, 210094, Nanjing, China
| | - Yao Zhang
- Institute of Computing Technology, Chinese Academy of Sciences and the University of Chinese Academy of Sciences, 100019, Beijing, China
| | - Song Gu
- Department of Image Reconstruction, Nanjing Anke Medical Technology Co., Ltd., 211113, Nanjing, China
| | - Xingle An
- Infervision Technology Co. Ltd., 100020, Beijing, China
| | - Zhihe Wang
- Shenzhen Haichuang Medical Co., Ltd., 518049, Shenzhen, China
| | - Cheng Ge
- Institute of Bioinformatics and Medical Engineering, Jiangsu University of Technology, 213001, Changzhou, China
| | - Congcong Wang
- School of Computer Science and Engineering, Tianjin University of Technology, 300384, Tianjin, China; Engineering Research Center of Learning-Based Intelligent System, Ministry of Education, 300384, Tianjin, China
| | - Fan Zhang
- Radiological Algorithm, Fosun Aitrox Information Technology Co., Ltd., 200033, Shanghai, China
| | - Yu Wang
- Radiological Algorithm, Fosun Aitrox Information Technology Co., Ltd., 200033, Shanghai, China
| | - Yinan Xu
- Key Lab of Intelligent Perception and Image Understanding of Ministry of Education, Xidian University, 710071, Shaanxi, China
| | - Shuiping Gou
- Key Lab of Intelligent Perception and Image Understanding of Ministry of Education, Xidian University, 710071, Shaanxi, China
| | - Franz Thaler
- Gottfried Schatz Research Center: Biophysics, Medical University of Graz, 8010, Graz, Austria; Institute of Computer Graphics and Vision, Graz University of Technology, 8010, Graz, Austria
| | - Christian Payer
- Institute of Computer Graphics and Vision, Graz University of Technology, 8010, Graz, Austria
| | - Darko Štern
- Gottfried Schatz Research Center: Biophysics, Medical University of Graz, 8010, Graz, Austria
| | - Edward G A Henderson
- Division of Cancer Sciences, The University of Manchester, M139PL, Manchester, UK; Radiotherapy Related Research, The Christie NHS Foundation Trust, M139PL, Manchester, UK
| | - Dónal M McSweeney
- Division of Cancer Sciences, The University of Manchester, M139PL, Manchester, UK; Radiotherapy Related Research, The Christie NHS Foundation Trust, M139PL, Manchester, UK
| | - Andrew Green
- Division of Cancer Sciences, The University of Manchester, M139PL, Manchester, UK; Radiotherapy Related Research, The Christie NHS Foundation Trust, M139PL, Manchester, UK
| | - Price Jackson
- Peter MacCallum Cancer Centre, 3000, Melbourne, Australia
| | | | - Quoc-Cuong Nguyen
- University of Information Technology, VNU-HCM, 700000, Ho Chi Minh City, Viet Nam
| | - Abdul Qayyum
- Brest National School of Engineering, UMR CNRS 6285 LabSTICC, 29280, Brest, France
| | | | - Ziyan Huang
- Institute of Medical Robotics, Shanghai Jiao Tong University, 200240, Shanghai, China
| | - Ziqi Zhou
- Guangdong Key Laboratory of Intelligent Information Processing, Shenzhen University, 518000, Shenzhen, China
| | - Deng-Ping Fan
- College of Computer Science, Nankai University, 300071, Tianjin, China; Inception Institute of Artificial Intelligence, Abu Dhabi, United Arab Emirates
| | - Huan Xiong
- Mohamed bin Zayed University of Artificial Intelligence, Abu Dhabi, United Arab Emirates; Harbin Institute of Technology, 150001, Harbin, China
| | - Guoqiang Dong
- Department of Nuclear Medicine, Nanjing Drum Tower Hospital, the Affiliated Hospital of Nanjing University Medical School, 210008, Nanjing, China; Department of Interventional Radiology, The Second Affiliated Hospital of Bengbu Medical College, 233017, Bengbu, China
| | - Qiongjie Zhu
- Department of Nuclear Medicine, Nanjing Drum Tower Hospital, the Affiliated Hospital of Nanjing University Medical School, 210008, Nanjing, China; Department of Radiology, Shidong Hospital, 200438, Shanghai, China
| | - Jian He
- Department of Nuclear Medicine, Nanjing Drum Tower Hospital, the Affiliated Hospital of Nanjing University Medical School, 210008, Nanjing, China
| | - Xiaoping Yang
- Department of Mathematics, Nanjing University, 210093, Nanjing, China.
| |
Collapse
|
36
|
Advancing Brain Metastases Detection in T1-Weighted Contrast-Enhanced 3D MRI Using Noisy Student-Based Training. Diagnostics (Basel) 2022; 12:diagnostics12082023. [PMID: 36010373 PMCID: PMC9407228 DOI: 10.3390/diagnostics12082023] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 07/23/2022] [Revised: 08/17/2022] [Accepted: 08/19/2022] [Indexed: 11/17/2022] Open
Abstract
The detection of brain metastases (BM) in their early stages could have a positive impact on the outcome of cancer patients. The authors previously developed a framework for detecting small BM (with diameters of <15 mm) in T1-weighted contrast-enhanced 3D magnetic resonance images (T1c). This study aimed to advance the framework with a noisy-student-based self-training strategy to use a large corpus of unlabeled T1c data. Accordingly, a sensitivity-based noisy-student learning approach was formulated to provide high BM detection sensitivity with a reduced count of false positives. This paper (1) proposes student/teacher convolutional neural network architectures, (2) presents data and model noising mechanisms, and (3) introduces a novel pseudo-labeling strategy factoring in the sensitivity constraint. The evaluation was performed using 217 labeled and 1247 unlabeled exams via two-fold cross-validation. The framework utilizing only the labeled exams produced 9.23 false positives for 90% BM detection sensitivity, whereas the one using the introduced learning strategy led to ~9% reduction in false detections (i.e., 8.44). Significant reductions in false positives (>10%) were also observed in reduced labeled data scenarios (using 50% and 75% of labeled data). The results suggest that the introduced strategy could be utilized in existing medical detection applications with access to unlabeled datasets to elevate their performances.
Collapse
|
37
|
Gou S, Xu Y, Yang H, Tong N, Zhang X, Wei L, Zhao L, Zheng M, Liu W. Automated cervical tumor segmentation on MR images using multi-view feature attention network. Biomed Signal Process Control 2022. [DOI: 10.1016/j.bspc.2022.103832] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/02/2022]
|
38
|
Zhang Y, Liao Q, Ding L, Zhang J. Bridging 2D and 3D segmentation networks for computation-efficient volumetric medical image segmentation: An empirical study of 2.5D solutions. Comput Med Imaging Graph 2022; 99:102088. [DOI: 10.1016/j.compmedimag.2022.102088] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 10/25/2020] [Revised: 05/24/2022] [Accepted: 05/26/2022] [Indexed: 11/28/2022]
|
39
|
Liu J, Fan H, Wang Q, Li W, Tang Y, Wang D, Zhou M, Chen L. Local Label Point Correction for Edge Detection of Overlapping Cervical Cells. Front Neuroinform 2022; 16:895290. [PMID: 35645753 PMCID: PMC9133536 DOI: 10.3389/fninf.2022.895290] [Citation(s) in RCA: 2] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [Abstract] [Grants] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 03/13/2022] [Accepted: 04/20/2022] [Indexed: 11/18/2022] Open
Abstract
Accurate labeling is essential for supervised deep learning methods. However, it is almost impossible to accurately and manually annotate thousands of images, which results in many labeling errors for most datasets. We proposes a local label point correction (LLPC) method to improve annotation quality for edge detection and image segmentation tasks. Our algorithm contains three steps: gradient-guided point correction, point interpolation, and local point smoothing. We correct the labels of object contours by moving the annotated points to the pixel gradient peaks. This can improve the edge localization accuracy, but it also causes unsmooth contours due to the interference of image noise. Therefore, we design a point smoothing method based on local linear fitting to smooth the corrected edge. To verify the effectiveness of our LLPC, we construct a largest overlapping cervical cell edge detection dataset (CCEDD) with higher precision label corrected by our label correction method. Our LLPC only needs to set three parameters, but yields 30–40% average precision improvement on multiple networks. The qualitative and quantitative experimental results show that our LLPC can improve the quality of manual labels and the accuracy of overlapping cell edge detection. We hope that our study will give a strong boost to the development of the label correction for edge detection and image segmentation. We will release the dataset and code at: https://github.com/nachifur/LLPC.
Collapse
Affiliation(s)
- Jiawei Liu
- State Key Laboratory of Robotics, Shenyang Institute of Automation, Chinese Academy of Sciences, Shenyang, China
- Institutes for Robotics and Intelligent Manufacturing, Chinese Academy of Sciences, Shenyang, China
- University of Chinese Academy of Sciences, Beijing, China
| | - Huijie Fan
- State Key Laboratory of Robotics, Shenyang Institute of Automation, Chinese Academy of Sciences, Shenyang, China
- Institutes for Robotics and Intelligent Manufacturing, Chinese Academy of Sciences, Shenyang, China
- *Correspondence: Huijie Fan
| | - Qiang Wang
- State Key Laboratory of Robotics, Shenyang Institute of Automation, Chinese Academy of Sciences, Shenyang, China
- Key Laboratory of Manufacturing Industrial Integrated, Shenyang University, Shenyang, China
| | - Wentao Li
- State Key Laboratory of Robotics, Shenyang Institute of Automation, Chinese Academy of Sciences, Shenyang, China
- Institutes for Robotics and Intelligent Manufacturing, Chinese Academy of Sciences, Shenyang, China
| | - Yandong Tang
- State Key Laboratory of Robotics, Shenyang Institute of Automation, Chinese Academy of Sciences, Shenyang, China
- Institutes for Robotics and Intelligent Manufacturing, Chinese Academy of Sciences, Shenyang, China
| | - Danbo Wang
- Department of Gynecology, Cancer Hospital of China Medical University, Liaoning Cancer Hospital & Institute, Shenyang, China
- Danbo Wang
| | - Mingyi Zhou
- Department of Gynecology, Cancer Hospital of China Medical University, Liaoning Cancer Hospital & Institute, Shenyang, China
| | - Li Chen
- Department of Pathology, Cancer Hospital of China Medical University, Liaoning Cancer Hospital & Institute, Shenyang, China
| |
Collapse
|
40
|
Amjad A, Xu J, Thill D, Lawton C, Hall W, Awan MJ, Shukla M, Erickson BA, Li XA. General and custom deep learning autosegmentation models for organs in head and neck, abdomen, and male pelvis. Med Phys 2022; 49:1686-1700. [PMID: 35094390 PMCID: PMC8917093 DOI: 10.1002/mp.15507] [Citation(s) in RCA: 16] [Impact Index Per Article: 8.0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 07/13/2021] [Revised: 01/19/2022] [Accepted: 01/21/2022] [Indexed: 11/11/2022] Open
Abstract
PURPOSE To reduce workload and inconsistencies in organ segmentation for radiation treatment planning, we developed and evaluated general and custom autosegmentation models on computed tomography (CT) for three major tumor sites using a well-established deep convolutional neural network (DCNN). METHODS Five CT-based autosegmentation models for 42 organs at risk (OARs) in head and neck (HN), abdomen (ABD), and male pelvis (MP) were developed using a full three-dimensional (3D) DCNN architecture. Two types of deep learning (DL) models were separately trained using either general diversified multi-institutional datasets or custom well-controlled single-institution datasets. To improve segmentation accuracy, an adaptive spatial resolution approach for small and/or narrow OARs and a pseudo scan extension approach, when CT scan length is too short to cover entire organs, were implemented. The performance of the obtained models was evaluated based on accuracy and clinical applicability of the autosegmented contours using qualitative visual inspection and quantitative calculation of dice similarity coefficient (DSC), mean distance to agreement (MDA), and time efficiency. RESULTS The five DL autosegmentation models developed for the three anatomical sites were found to have high accuracy (DSC ranging from 0.8 to 0.98) for 74% OARs and marginally acceptable for 26% OARs. The custom models performed slightly better than the general models, even with smaller custom datasets used for the custom model training. The organ-based approaches improved autosegmentation accuracy for small or complex organs (e.g., eye lens, optic nerves, inner ears, and bowels). Compared with traditional manual contouring times, the autosegmentation times, including subsequent manual editing, if necessary, were substantially reduced by 88% for MP, 80% for HN, and 65% for ABD models. CONCLUSIONS The obtained autosegmentation models, incorporating organ-based approaches, were found to be effective and accurate for most OARs in the male pelvis, head and neck, and abdomen. We have demonstrated that our multianatomical DL autosegmentation models are clinically useful for radiation treatment planning.
Collapse
Affiliation(s)
- Asma Amjad
- Department of Radiation Oncology, Medical College of Wisconsin, WI, USA
| | | | | | - Colleen Lawton
- Department of Radiation Oncology, Medical College of Wisconsin, WI, USA
| | - William Hall
- Department of Radiation Oncology, Medical College of Wisconsin, WI, USA
| | - Musaddiq J. Awan
- Department of Radiation Oncology, Medical College of Wisconsin, WI, USA
| | - Monica Shukla
- Department of Radiation Oncology, Medical College of Wisconsin, WI, USA
| | - Beth A. Erickson
- Department of Radiation Oncology, Medical College of Wisconsin, WI, USA
| | - X. Allen Li
- Department of Radiation Oncology, Medical College of Wisconsin, WI, USA
| |
Collapse
|
41
|
Omari EA, Zhang Y, Ahunbay E, Paulson E, Amjad A, Chen X, Liang Y, Li XA. Multi parametric magnetic resonance imaging for radiation treatment planning. Med Phys 2022; 49:2836-2845. [PMID: 35170769 DOI: 10.1002/mp.15534] [Citation(s) in RCA: 1] [Impact Index Per Article: 0.5] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 05/21/2021] [Revised: 10/05/2021] [Accepted: 01/03/2022] [Indexed: 11/09/2022] Open
Abstract
In recent years, multi-parametric magnetic resonance imaging (MpMRI) has played a major role in radiation therapy treatment planning. The superior soft tissue contrast, functional or physiological imaging capabilities and the flexibility of site-specific image sequence development has placed MpMRI at the forefront. In this article, the present status of MpMRI for external beam radiation therapy planning is reviewed. Common MpMRI sequences, preprocessing and QA strategies are briefly discussed, and various image registration techniques and strategies are addressed. Image segmentation methods including automatic segmentation and deep learning techniques for organs at risk and target delineation are reviewed. Due to the advancement in MRI guided online adaptive radiotherapy, treatment planning considerations addressing MRI only planning are also discussed. This article is protected by copyright. All rights reserved.
Collapse
Affiliation(s)
- Eenas A Omari
- Department of Radiation Oncology, Medical College of Wisconsin, Milwaukee, WI, 53226, USA
| | - Ying Zhang
- Department of Radiation Oncology, Medical College of Wisconsin, Milwaukee, WI, 53226, USA
| | - Ergun Ahunbay
- Department of Radiation Oncology, Medical College of Wisconsin, Milwaukee, WI, 53226, USA
| | - Eric Paulson
- Department of Radiation Oncology, Medical College of Wisconsin, Milwaukee, WI, 53226, USA
| | - Asma Amjad
- Department of Radiation Oncology, Medical College of Wisconsin, Milwaukee, WI, 53226, USA
| | - Xinfeng Chen
- Department of Radiation Oncology, Medical College of Wisconsin, Milwaukee, WI, 53226, USA
| | - Ying Liang
- Department of Radiation Oncology, Medical College of Wisconsin, Milwaukee, WI, 53226, USA
| | - X Allen Li
- Department of Radiation Oncology, Medical College of Wisconsin, Milwaukee, WI, 53226, USA
| |
Collapse
|
42
|
Zhang Y, Liao Q, Yuan L, Zhu H, Xing J, Zhang J. Exploiting Shared Knowledge From Non-COVID Lesions for Annotation-Efficient COVID-19 CT Lung Infection Segmentation. IEEE J Biomed Health Inform 2021; 25:4152-4162. [PMID: 34415840 PMCID: PMC8843066 DOI: 10.1109/jbhi.2021.3106341] [Citation(s) in RCA: 12] [Impact Index Per Article: 4.0] [Reference Citation Analysis] [Abstract] [MESH Headings] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/25/2022]
Abstract
The novel Coronavirus disease (COVID-19) is a highly contagious virus and has spread all over the world, posing an extremely serious threat to all countries. Automatic lung infection segmentation from computed tomography (CT) plays an important role in the quantitative analysis of COVID-19. However, the major challenge lies in the inadequacy of annotated COVID-19 datasets. Currently, there are several public non-COVID lung lesion segmentation datasets, providing the potential for generalizing useful information to the related COVID-19 segmentation task. In this paper, we propose a novel relation-driven collaborative learning model to exploit shared knowledge from non-COVID lesions for annotation-efficient COVID-19 CT lung infection segmentation. The model consists of a general encoder to capture general lung lesion features based on multiple non-COVID lesions, and a target encoder to focus on task-specific features based on COVID-19 infections. We develop a collaborative learning scheme to regularize feature-level relation consistency of given input and encourage the model to learn more general and discriminative representation of COVID-19 infections. Extensive experiments demonstrate that trained with limited COVID-19 data, exploiting shared knowledge from non-COVID lesions can further improve state-of-the-art performance with up to 3.0% in dice similarity coefficient and 4.2% in normalized surface dice. In addition, experimental results on large scale 2D dataset with CT slices show that our method significantly outperforms cutting-edge segmentation methods metrics. Our method promotes new insights into annotation-efficient deep learning and illustrates strong potential for real-world applications in the global fight against COVID-19 in the absence of sufficient high-quality annotations.
Collapse
|
43
|
Artificial Intelligence Model to Detect Real Contact Relationship between Mandibular Third Molars and Inferior Alveolar Nerve Based on Panoramic Radiographs. Diagnostics (Basel) 2021; 11:diagnostics11091664. [PMID: 34574005 PMCID: PMC8465495 DOI: 10.3390/diagnostics11091664] [Citation(s) in RCA: 9] [Impact Index Per Article: 3.0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 07/29/2021] [Revised: 09/01/2021] [Accepted: 09/08/2021] [Indexed: 11/17/2022] Open
Abstract
This study aimed to develop a novel detection model for automatically assessing the real contact relationship between mandibular third molars (MM3s) and the inferior alveolar nerve (IAN) based on panoramic radiographs processed with deep learning networks, minimizing pseudo-contact interference and reducing the frequency of cone beam computed tomography (CBCT) use. A deep-learning network approach based on YOLOv4, named as MM3-IANnet, was applied to oral panoramic radiographs for the first time. The relationship between MM3s and the IAN in CBCT was considered the real contact relationship. Accuracy metrics were calculated to evaluate and compare the performance of the MM3-IANnet, dentists and a cooperative approach with dentists and the MM3-IANnet. Our results showed that in comparison with detection by dentists (AP = 76.45%) or the MM3-IANnet (AP = 83.02%), the cooperative dentist-MM3-IANnet approach yielded the highest average precision (AP = 88.06%). In conclusion, the MM3-IANnet detection model is an encouraging artificial intelligence approach that might assist dentists in detecting the real contact relationship between MM3s and IANs based on panoramic radiographs.
Collapse
|
44
|
Santhosh Reddy D, Rajalakshmi P, Mateen M. A deep learning based approach for classification of abdominal organs using ultrasound images. Biocybern Biomed Eng 2021. [DOI: 10.1016/j.bbe.2021.05.004] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/24/2022]
|