1
|
Farooq MU, Ullah Z, Khan A, Gwak J. DC-AAE: Dual channel adversarial autoencoder with multitask learning for KL-grade classification in knee radiographs. Comput Biol Med 2023; 167:107570. [PMID: 37897960 DOI: 10.1016/j.compbiomed.2023.107570] [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: 05/04/2022] [Revised: 08/25/2023] [Accepted: 10/10/2023] [Indexed: 10/30/2023]
Abstract
Knee osteoarthritis (OA) is a frequent musculoskeletal disorder that leads to physical disability in older adults. Manual OA assessment is performed via visual inspection, which is highly subjective as it suffers from moderate to high inter-observer variability. Many deep learning-based techniques have been proposed to address this issue. However, owing to the limited amount of labelled data, all existing solutions have limitations in terms of performance or the number of classes. This paper proposes a novel fully automatic Kellgren and Lawrence (KL) grade classification scheme in knee radiographs. We developed a semi-supervised multi-task learning-based approach that enables the exploitation of additional unlabelled data in an unsupervised as well as supervised manner. Specifically, we propose a dual-channel adversarial autoencoder, which is first trained in an unsupervised manner for reconstruction tasks only. To exploit the additional data in a supervised way, we propose a multi-task learning framework by introducing an auxiliary task. In particular, we use leg side identification as an auxiliary task, which allows the use of more datasets, e.g., CHECK dataset. The work demonstrates that the utilization of additional data can improve the primary task of KL-grade classification for which only limited labelled data is available. This semi-supervised learning essentially helps to improve the feature learning ability of our framework, which leads to improved performance for KL-grade classification. We rigorously evaluated our proposed model on the two largest publicly available datasets for various aspects, i.e., overall performance, the effect of additional unlabelled samples and auxiliary tasks, robustness analysis, and ablation study. The proposed model achieved the accuracy, precision, recall, and F1 score of 75.53%, 74.1%, 78.51%, and 75.34%, respectively. Furthermore, the experimental results show that the suggested model not only achieves state-of-the-art performance on two publicly available datasets but also exhibits remarkable robustness.
Collapse
Affiliation(s)
- Muhammad Umar Farooq
- Department of IT, Energy Convergence (BK21 FOUR), Korea National University of Transportation, Chungju 27469, South Korea
| | - Zahid Ullah
- Department of Software, Korea National University of Transportation, Chungju 27469, South Korea
| | - Asifullah Khan
- Pattern Recognition Lab, DCIS, PIEAS, Nilore, Islamabad 45650, Pakistan
| | - Jeonghwan Gwak
- Department of IT, Energy Convergence (BK21 FOUR), Korea National University of Transportation, Chungju 27469, South Korea; Department of Software, Korea National University of Transportation, Chungju 27469, South Korea; Department of Biomedical Engineering, Korea National University of Transportation, Chungju 27469, South Korea; Department of AI Robotics Engineering, Korea National University of Transportation, Chungju 27469, South Korea.
| |
Collapse
|
2
|
Qi W, Wu HC, Chan SC. MDF-Net: A Multi-Scale Dynamic Fusion Network for Breast Tumor Segmentation of Ultrasound Images. IEEE TRANSACTIONS ON IMAGE PROCESSING : A PUBLICATION OF THE IEEE SIGNAL PROCESSING SOCIETY 2023; 32:4842-4855. [PMID: 37639409 DOI: 10.1109/tip.2023.3304518] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [MESH Headings] [Track Full Text] [Subscribe] [Scholar Register] [Indexed: 08/31/2023]
Abstract
Breast tumor segmentation of ultrasound images provides valuable information of tumors for early detection and diagnosis. Accurate segmentation is challenging due to low image contrast between areas of interest; speckle noises, and large inter-subject variations in tumor shape and size. This paper proposes a novel Multi-scale Dynamic Fusion Network (MDF-Net) for breast ultrasound tumor segmentation. It employs a two-stage end-to-end architecture with a trunk sub-network for multiscale feature selection and a structurally optimized refinement sub-network for mitigating impairments such as noise and inter-subject variation via better feature exploration and fusion. The trunk network is extended from UNet++ with a simplified skip pathway structure to connect the features between adjacent scales. Moreover, deep supervision at all scales, instead of at the finest scale in UNet++, is proposed to extract more discriminative features and mitigate errors from speckle noise via a hybrid loss function. Unlike previous works, the first stage is linked to a loss function of the second stage so that both the preliminary segmentations and refinement subnetworks can be refined together at training. The refinement sub-network utilizes a structurally optimized MDF mechanism to integrate preliminary segmentation information (capturing general tumor shape and size) at coarse scales and explores inter-subject variation information at finer scales. Experimental results from two public datasets show that the proposed method achieves better Dice and other scores over state-of-the-art methods. Qualitative analysis also indicates that our proposed network is more robust to tumor size/shapes, speckle noise and heavy posterior shadows along tumor boundaries. An optional post-processing step is also proposed to facilitate users in mitigating segmentation artifacts. The efficiency of the proposed network is also illustrated on the "Electron Microscopy neural structures segmentation dataset". It outperforms a state-of-the-art algorithm based on UNet-2022 with simpler settings. This indicates the advantages of our MDF-Nets in other challenging image segmentation tasks with small to medium data sizes.
Collapse
|
3
|
Liao X, Li J, Li L, Shangguan C, Huang S. RGBD Salient Object Detection, Based on Specific Object Imaging. SENSORS (BASEL, SWITZERLAND) 2022; 22:8973. [PMID: 36433571 PMCID: PMC9696882 DOI: 10.3390/s22228973] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Figures] [Subscribe] [Scholar Register] [Received: 09/03/2022] [Revised: 11/16/2022] [Accepted: 11/16/2022] [Indexed: 06/16/2023]
Abstract
RGBD salient object detection, based on the convolutional neural network, has achieved rapid development in recent years. However, existing models often focus on detecting salient object edges, instead of objects. Importantly, detecting objects can more intuitively display the complete information of the detection target. To take care of this issue, we propose a RGBD salient object detection method, based on specific object imaging, which can quickly capture and process important information on object features, and effectively screen out the salient objects in the scene. The screened target objects include not only the edge of the object, but also the complete feature information of the object, which realizes the detection and imaging of the salient objects. We conduct experiments on benchmark datasets and validate with two common metrics, and the results show that our method reduces the error by 0.003 and 0.201 (MAE) on D3Net and JLDCF, respectively. In addition, our method can still achieve a very good detection and imaging performance in the case of the greatly reduced training data.
Collapse
Affiliation(s)
- Xiaolian Liao
- School of Physics and Telecommunications Engineering, South China Normal University, Guangzhou 510006, China
| | - Jun Li
- School of Physics and Telecommunications Engineering, South China Normal University, Guangzhou 510006, China
- School of Electronics and Information Engineering, South China Normal University, Foshan 528225, China
| | - Leyi Li
- School of Physics and Telecommunications Engineering, South China Normal University, Guangzhou 510006, China
| | - Caoxi Shangguan
- School of Physics and Telecommunications Engineering, South China Normal University, Guangzhou 510006, China
| | - Shaoyan Huang
- School of Physics and Telecommunications Engineering, South China Normal University, Guangzhou 510006, China
| |
Collapse
|
4
|
Ilesanmi AE, Chaumrattanakul U, Makhanov SS. Methods for the segmentation and classification of breast ultrasound images: a review. J Ultrasound 2021; 24:367-382. [PMID: 33428123 PMCID: PMC8572242 DOI: 10.1007/s40477-020-00557-5] [Citation(s) in RCA: 3] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 11/11/2020] [Accepted: 12/21/2020] [Indexed: 02/07/2023] Open
Abstract
PURPOSE Breast ultrasound (BUS) is one of the imaging modalities for the diagnosis and treatment of breast cancer. However, the segmentation and classification of BUS images is a challenging task. In recent years, several methods for segmenting and classifying BUS images have been studied. These methods use BUS datasets for evaluation. In addition, semantic segmentation algorithms have gained prominence for segmenting medical images. METHODS In this paper, we examined different methods for segmenting and classifying BUS images. Popular datasets used to evaluate BUS images and semantic segmentation algorithms were examined. Several segmentation and classification papers were selected for analysis and review. Both conventional and semantic methods for BUS segmentation were reviewed. RESULTS Commonly used methods for BUS segmentation were depicted in a graphical representation, while other conventional methods for segmentation were equally elucidated. CONCLUSIONS We presented a review of the segmentation and classification methods for tumours detected in BUS images. This review paper selected old and recent studies on segmenting and classifying tumours in BUS images.
Collapse
Affiliation(s)
- Ademola E. Ilesanmi
- School of ICT, Sirindhorn International Institute of Technology, Thammasat University, Pathum Thani, 12000 Thailand
| | | | - Stanislav S. Makhanov
- School of ICT, Sirindhorn International Institute of Technology, Thammasat University, Pathum Thani, 12000 Thailand
| |
Collapse
|
5
|
Webb JM, Adusei SA, Wang Y, Samreen N, Adler K, Meixner DD, Fazzio RT, Fatemi M, Alizad A. Comparing deep learning-based automatic segmentation of breast masses to expert interobserver variability in ultrasound imaging. Comput Biol Med 2021; 139:104966. [PMID: 34715553 PMCID: PMC8642313 DOI: 10.1016/j.compbiomed.2021.104966] [Citation(s) in RCA: 5] [Impact Index Per Article: 1.7] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 06/18/2021] [Revised: 10/18/2021] [Accepted: 10/19/2021] [Indexed: 12/14/2022]
Abstract
Deep learning is a powerful tool that became practical in 2008, harnessing the power of Graphic Processing Unites, and has developed rapidly in image, video, and natural language processing. There are ongoing developments in the application of deep learning to medical data for a variety of tasks across multiple imaging modalities. The reliability and repeatability of deep learning techniques are of utmost importance if deep learning can be considered a tool for assisting experts, including physicians, radiologists, and sonographers. Owing to the high costs of labeling data, deep learning models are often evaluated against one expert, and it is unknown if any errors fall within a clinically acceptable range. Ultrasound is a commonly used imaging modality for breast cancer screening processes and for visually estimating risk using the Breast Imaging Reporting and Data System score. This process is highly dependent on the skills and experience of the sonographers and radiologists, thereby leading to interobserver variability and interpretation. For these reasons, we propose an interobserver reliability study comparing the performance of a current top-performing deep learning segmentation model against three experts who manually segmented suspicious breast lesions in clinical ultrasound (US) images. We pretrained the model using a US thyroid segmentation dataset with 455 patients and 50,993 images, and trained the model using a US breast segmentation dataset with 733 patients and 29,884 images. We found a mean Fleiss kappa value of 0.78 for the performance of three experts in breast mass segmentation compared to a mean Fleiss kappa value of 0.79 for the performance of experts and the optimized deep learning model.
Collapse
Affiliation(s)
- Jeremy M Webb
- Department of Radiology, Mayo Clinic College of Medicine and Science, Rochester, MN
| | - Shaheeda A Adusei
- Department of Physiology and Biomedical Engineering, Mayo Clinic College of Medicine and Science, Rochester, MN
| | - Yinong Wang
- Department of Radiology, Mayo Clinic College of Medicine and Science, Rochester, MN
| | - Naziya Samreen
- Department of Radiology, Mayo Clinic College of Medicine and Science, Rochester, MN
| | - Kalie Adler
- Department of Radiology, Mayo Clinic College of Medicine and Science, Rochester, MN
| | - Duane D Meixner
- Department of Radiology, Mayo Clinic College of Medicine and Science, Rochester, MN
| | - Robert T Fazzio
- Department of Radiology, Mayo Clinic College of Medicine and Science, Rochester, MN
| | - Mostafa Fatemi
- Department of Physiology and Biomedical Engineering, Mayo Clinic College of Medicine and Science, Rochester, MN
| | - Azra Alizad
- Department of Radiology, Mayo Clinic College of Medicine and Science, Rochester, MN,Department of Physiology and Biomedical Engineering, Mayo Clinic College of Medicine and Science, Rochester, MN,Corresponding author: Azra Alizad, 200 1 St. SW, Rochester, MN 55 902,
| |
Collapse
|
6
|
Fang L, Wang X, Wang M. Superpixel/voxel medical image segmentation algorithm based on the regional interlinked value. Pattern Anal Appl 2021. [DOI: 10.1007/s10044-021-01021-8] [Citation(s) in RCA: 2] [Impact Index Per Article: 0.7] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/24/2022]
|
7
|
A method for segmentation of tumors in breast ultrasound images using the variant enhanced deep learning. Biocybern Biomed Eng 2021. [DOI: 10.1016/j.bbe.2021.05.007] [Citation(s) in RCA: 14] [Impact Index Per Article: 4.7] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/18/2022]
|
8
|
Glass-cutting medical images via a mechanical image segmentation method based on crack propagation. Nat Commun 2020; 11:5669. [PMID: 33168802 PMCID: PMC7652839 DOI: 10.1038/s41467-020-19392-7] [Citation(s) in RCA: 5] [Impact Index Per Article: 1.3] [Reference Citation Analysis] [Abstract] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 01/29/2020] [Accepted: 10/07/2020] [Indexed: 11/23/2022] Open
Abstract
Medical image segmentation is crucial in diagnosing and treating diseases, but automatic segmentation of complex images is very challenging. Here we present a method, called the crack propagation method (CPM), based on the principles of fracture mechanics. This unique method converts the image segmentation problem into a mechanical one, extracting the boundary information of the target area by tracing the crack propagation on a thin plate with grooves corresponding to the area edge. The greatest advantage of CPM is in segmenting images involving blurred or even discontinuous boundaries, a task difficult to achieve by existing auto-segmentation methods. The segmentation results for synthesized images and real medical images show that CPM has high accuracy in segmenting complex boundaries. With increasing demand for medical imaging in clinical practice and research, this method will show its unique potential. Automatic segmentation of complex medical images is challenging. Here, the authors present a crack propagation method based on the principles of fracture mechanics: extracting the boundary information of the target area by tracing the crack propagation on a thin plate with grooves corresponding to the area edge.
Collapse
|