1
|
Demirbaş AA, Üzen H, Fırat H. Spatial-attention ConvMixer architecture for classification and detection of gastrointestinal diseases using the Kvasir dataset. Health Inf Sci Syst 2024; 12:32. [PMID: 38685985 PMCID: PMC11056348 DOI: 10.1007/s13755-024-00290-x] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Grants] [Track Full Text] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 12/08/2023] [Accepted: 04/12/2024] [Indexed: 05/02/2024] Open
Abstract
Gastrointestinal (GI) disorders, encompassing conditions like cancer and Crohn's disease, pose a significant threat to public health. Endoscopic examinations have become crucial for diagnosing and treating these disorders efficiently. However, the subjective nature of manual evaluations by gastroenterologists can lead to potential errors in disease classification. In addition, the difficulty of diagnosing diseased tissues in GI and the high similarity between classes made the subject a difficult area. Automated classification systems that use artificial intelligence to solve these problems have gained traction. Automatic detection of diseases in medical images greatly benefits in the diagnosis of diseases and reduces the time of disease detection. In this study, we suggested a new architecture to enable research on computer-assisted diagnosis and automated disease detection in GI diseases. This architecture, called Spatial-Attention ConvMixer (SAC), further developed the patch extraction technique used as the basis of the ConvMixer architecture with a spatial attention mechanism (SAM). The SAM enables the network to concentrate selectively on the most informative areas, assigning importance to each spatial location within the feature maps. We employ the Kvasir dataset to assess the accuracy of classifying GI illnesses using the SAC architecture. We compare our architecture's results with Vanilla ViT, Swin Transformer, ConvMixer, MLPMixer, ResNet50, and SqueezeNet models. Our SAC method gets 93.37% accuracy, while the other architectures get respectively 79.52%, 74.52%, 92.48%, 63.04%, 87.44%, and 85.59%. The proposed spatial attention block improves the accuracy of the ConvMixer architecture on the Kvasir, outperforming the state-of-the-art methods with an accuracy rate of 93.37%.
Collapse
Affiliation(s)
| | - Hüseyin Üzen
- Department of Computer Engineering, Faculty of Engineering, Bingol University, Bingol, Turkey
| | - Hüseyin Fırat
- Department of Computer Engineering, Faculty of Engineering, Dicle University, Diyarbakır, Turkey
| |
Collapse
|
2
|
Saleem S, Amin J, Sharif M, Mallah GA, Kadry S, Gandomi AH. Leukemia segmentation and classification: A comprehensive survey. Comput Biol Med 2022; 150:106028. [PMID: 36126356 DOI: 10.1016/j.compbiomed.2022.106028] [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: 04/30/2022] [Revised: 07/11/2022] [Accepted: 08/20/2022] [Indexed: 11/30/2022]
Abstract
Blood is made up of leukocytes (WBCs), erythrocytes (RBCs), and thrombocytes. The ratio of blood cancer diseases is increasing rapidly, among which leukemia is one of the famous cancer which may lead to death. Leukemia cancer is initiated by the unnecessary growth of immature WBCs present in the sponge tissues of bone marrow. It is generally analyzed by etiologists by perceiving slides of blood smear images under a microscope. The morphological features and blood cells count facilitated the etiologists to detect leukemia. Due to the late detection and expensive instruments used for leukemia analysis, the death rate has risen significantly. The fluorescence-based cell sorting technique and manual recounts using a hemocytometer are error-prone and imprecise. Leukemia detection methods consist of pre-processing, segmentation, features extraction, and classification. In this article, recent deep learning methodologies and challenges for leukemia detection are discussed. These methods are helpful to examine the microscopic blood smears images and for the detection of leukemia more accurately.
Collapse
Affiliation(s)
- Saba Saleem
- Department of Computer Science, COMSATS University Islamabad, Wah Campus, Pakistan
| | - Javaria Amin
- Department of Computer Science, University of Wah, Wah Cantt, Pakistan
| | - Muhammad Sharif
- Department of Computer Science, COMSATS University Islamabad, Wah Campus, Pakistan
| | | | - Seifedine Kadry
- Department of Applied Data Science, Noroff University College, Kristiansand, Norway; Department of Electrical and Computer Engineering, Lebanese American University, Byblos, Lebanon
| | - Amir H Gandomi
- Faculty of Engineering & Information Technology, University of Technology Sydney, Ultimo, NSW, 2007, Australia.
| |
Collapse
|
3
|
Umer MJ, Amin J, Sharif M, Anjum MA, Azam F, Shah JH. An integrated framework for COVID-19 classification based on classical and quantum transfer learning from a chest radiograph. CONCURRENCY AND COMPUTATION : PRACTICE & EXPERIENCE 2022; 34:e6434. [PMID: 34512201 PMCID: PMC8420477 DOI: 10.1002/cpe.6434] [Citation(s) in RCA: 8] [Impact Index Per Article: 4.0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Subscribe] [Scholar Register] [Received: 02/22/2021] [Revised: 04/21/2021] [Accepted: 05/13/2021] [Indexed: 05/07/2023]
Abstract
COVID-19 is a quickly spreading over 10 million persons globally. The overall number of infected patients worldwide is estimated to be around 133,381,413 people. Infection rate is being increased on daily basis. It has also caused a devastating effect on the world economy and public health. Early stage detection of this disease is mandatory to reduce the mortality rate. Artificial intelligence performs a vital role for COVID-19 detection at an initial stage using chest radiographs. The proposed methods comprise of the two phases. Deep features (DFs) are derived from its last fully connected layers of pre-trained models like AlexNet and MobileNet in phase-I. Later these feature vectors are fused serially. Best features are selected through feature selection method of PCA and passed to the SVM and KNN for classification. In phase-II, quantum transfer learning model is utilized, in which a pre-trained ResNet-18 model is applied for DF collection and then these features are supplied as an input to the 4-qubit quantum circuit for model training with the tuned hyperparameters. The proposed technique is evaluated on two publicly available x-ray imaging datasets. The proposed methodology achieved an accuracy index of 99.0% with three classes including corona virus-positive images, normal images, and pneumonia radiographs. In comparison to other recently published work, the experimental findings show that the proposed approach outperforms it.
Collapse
Affiliation(s)
- Muhammad Junaid Umer
- Department of Computer ScienceComsats University Islamabad, Wah CampusRawalpindiPakistan
| | - Javeria Amin
- Department of Computer ScienceUniversity of WahRawalpindiPakistan
| | - Muhammad Sharif
- Department of Computer ScienceComsats University Islamabad, Wah CampusRawalpindiPakistan
| | | | - Faisal Azam
- Department of Computer ScienceComsats University Islamabad, Wah CampusRawalpindiPakistan
| | - Jamal Hussain Shah
- Department of Computer ScienceComsats University Islamabad, Wah CampusRawalpindiPakistan
| |
Collapse
|
4
|
Yunus U, Amin J, Sharif M, Yasmin M, Kadry S, Krishnamoorthy S. Recognition of Knee Osteoarthritis (KOA) Using YOLOv2 and Classification Based on Convolutional Neural Network. Life (Basel) 2022; 12:1126. [PMID: 36013305 PMCID: PMC9410095 DOI: 10.3390/life12081126] [Citation(s) in RCA: 8] [Impact Index Per Article: 4.0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 05/26/2022] [Revised: 07/02/2022] [Accepted: 07/05/2022] [Indexed: 12/23/2022] Open
Abstract
Knee osteoarthritis (KOA) is one of the deadliest forms of arthritis. If not treated at an early stage, it may lead to knee replacement. That is why early diagnosis of KOA is necessary for better treatment. Manually KOA detection is a time-consuming and error-prone task. Computerized methods play a vital role in accurate and speedy detection. Therefore, the classification and localization of the KOA method are proposed in this work using radiographic images. The two-dimensional radiograph images are converted into three-dimensional and LBP features are extracted having the dimension of N × 59 out of which the best features of N × 55 are selected using PCA. The deep features are also extracted using Alex-Net and Dark-net-53 with the dimensions of N × 1024 and N × 4096, respectively, where N represents the number of images. Then, N × 1000 features are selected individually from both models using PCA. Finally, the extracted features are fused serially with the dimension of N × 2055 and passed to the classifiers on a 10-fold cross-validation that provides an accuracy of 90.6% for the classification of KOA grades. The localization model is proposed with the combination of an open exchange neural network (ONNX) and YOLOv2 that is trained on the selected hyper-parameters. The proposed model provides 0.98 mAP for the localization of classified images. The experimental analysis proves that the presented framework provides better results as compared to existing works.
Collapse
Affiliation(s)
- Usman Yunus
- Department of Computer Science, COMSATS University Islamabad, Wah Campus, Wah Cantt 47010, Pakistan; (U.Y.); (M.S.); (M.Y.)
| | - Javeria Amin
- Department of Computer Science, University of Wah, Wah Cantt 47010, Pakistan;
| | - Muhammad Sharif
- Department of Computer Science, COMSATS University Islamabad, Wah Campus, Wah Cantt 47010, Pakistan; (U.Y.); (M.S.); (M.Y.)
| | - Mussarat Yasmin
- Department of Computer Science, COMSATS University Islamabad, Wah Campus, Wah Cantt 47010, Pakistan; (U.Y.); (M.S.); (M.Y.)
| | - Seifedine Kadry
- Department of Applied Data Science, Noroff University College, 4612 Kristiansand, Norway;
| | - Sujatha Krishnamoorthy
- Zhejiang Bioinformatics International Science and Technology Cooperation Center, Wenzhou-Kean University, Wenzhou 325060, China
- Wenzhou Municipal Key Lab of Applied Biomedical and Biopharmaceutical Informatics, Wenzhou-Kean University, Wenzhou 325060, China
| |
Collapse
|
5
|
Amin J, Anjum MA, Sharif M, Kadry S, Nadeem A, Ahmad SF. Liver Tumor Localization Based on YOLOv3 and 3D-Semantic Segmentation Using Deep Neural Networks. Diagnostics (Basel) 2022; 12:diagnostics12040823. [PMID: 35453870 PMCID: PMC9025116 DOI: 10.3390/diagnostics12040823] [Citation(s) in RCA: 5] [Impact Index Per Article: 2.5] [Reference Citation Analysis] [Abstract] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 02/02/2022] [Revised: 03/18/2022] [Accepted: 03/22/2022] [Indexed: 12/17/2022] Open
Abstract
Worldwide, more than 1.5 million deaths are occur due to liver cancer every year. The use of computed tomography (CT) for early detection of liver cancer could save millions of lives per year. There is also an urgent need for a computerized method to interpret, detect and analyze CT scans reliably, easily, and correctly. However, precise segmentation of minute tumors is a difficult task because of variation in the shape, intensity, size, low contrast of the tumor, and the adjacent tissues of the liver. To address these concerns, a model comprised of three parts: synthetic image generation, localization, and segmentation, is proposed. An optimized generative adversarial network (GAN) is utilized for generation of synthetic images. The generated images are localized by using the improved localization model, in which deep features are extracted from pre-trained Resnet-50 models and fed into a YOLOv3 detector as an input. The proposed modified model localizes and classifies the minute liver tumor with 0.99 mean average precision (mAp). The third part is segmentation, in which pre-trained Inceptionresnetv2 employed as a base-Network of Deeplabv3 and subsequently is trained on fine-tuned parameters with annotated ground masks. The experiments reflect that the proposed approach has achieved greater than 95% accuracy in the testing phase and it is proven that, in comparison to the recently published work in this domain, this research has localized and segmented the liver and minute liver tumor with more accuracy.
Collapse
Affiliation(s)
- Javaria Amin
- Department of Computer Science, University of Wah, Wah Cantt 47040, Pakistan;
| | | | - Muhammad Sharif
- Department of Computer Science, Comsats University Islamabad, Wah Campus, Wah Cantt 47040, Pakistan;
| | - Seifedine Kadry
- Department of Applied Data Science, Noroff University College, 4609 Kristiansand, Norway
- Correspondence:
| | - Ahmed Nadeem
- Department of Pharmacology & Toxicology, College of Pharmacy, King Saud University, P.O. Box 2455, Riyadh 11451, Saudi Arabia; (A.N.); (S.F.A.)
| | - Sheikh F. Ahmad
- Department of Pharmacology & Toxicology, College of Pharmacy, King Saud University, P.O. Box 2455, Riyadh 11451, Saudi Arabia; (A.N.); (S.F.A.)
| |
Collapse
|
6
|
|
7
|
Saleem S, Amin J, Sharif M, Anjum MA, Iqbal M, Wang SH. A deep network designed for segmentation and classification of leukemia using fusion of the transfer learning models. COMPLEX INTELL SYST 2021. [DOI: 10.1007/s40747-021-00473-z] [Citation(s) in RCA: 16] [Impact Index Per Article: 5.3] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 12/11/2022]
Abstract
AbstractWhite blood cells (WBCs) are a portion of the immune system which fights against germs. Leukemia is the most common blood cancer which may lead to death. It occurs due to the production of a large number of immature WBCs in the bone marrow that destroy healthy cells. To overcome the severity of this disease, it is necessary to diagnose the shapes of immature cells at an early stage that ultimately reduces the modality rate of the patients. Recently different types of segmentation and classification methods are presented based upon deep-learning (DL) models but still have some limitations. This research aims to propose a modified DL approach for the accurate segmentation of leukocytes and their classification. The proposed technique includes two core steps: preprocessing-based classification and segmentation. In preprocessing, synthetic images are generated using a generative adversarial network (GAN) and normalized by color transformation. The optimal deep features are extracted from each blood smear image using pretrained deep models i.e., DarkNet-53 and ShuffleNet. More informative features are selected by principal component analysis (PCA) and fused serially for classification. The morphological operations based on color thresholding with the deep semantic method are utilized for leukemia segmentation of classified cells. The classification accuracy achieved with ALL-IDB and LISC dataset is 100% and 99.70% for the classification of leukocytes i.e., blast, no blast, basophils, neutrophils, eosinophils, lymphocytes, and monocytes, respectively. Whereas semantic segmentation achieved 99.10% and 98.60% for average and global accuracy, respectively. The proposed method achieved outstanding outcomes as compared to the latest existing research works.
Collapse
|