Article
Computer Science, Interdisciplinary Applications
Fei Long, Jing-Jie Peng, Weitao Song, Xiaobo Xia, Jun Sang
Summary: In this study, a capsule-based model, BloodCaps, was proposed for accurate multiclassification of blood cells, achieving superior accuracy on a large-scale dataset. Compared to other convolutional neural networks, BloodCaps showed the best performance across multiple datasets.
COMPUTER METHODS AND PROGRAMS IN BIOMEDICINE
(2021)
Article
Computer Science, Artificial Intelligence
Khondaker Tasrif Noor, Antonio Robles-Kelly
Summary: In this paper, the authors propose H-CapsNet, a capsule network designed for hierarchical image classification. The network effectively captures hierarchical relationships using dedicated capsules for each class hierarchy. A modified hinge loss is utilized to enforce consistency among the involved hierarchies. Additionally, a strategy for dynamically adjusting training parameters is presented to achieve better balance between the class hierarchies. Experimental results demonstrate that H-CapsNet outperforms competing hierarchical classification networks.
PATTERN RECOGNITION
(2024)
Article
Computer Science, Artificial Intelligence
Yaw Afriyie, Benjamin A. Weyori, Alex A. Opoku
Summary: Blood cell classification from medical images is crucial in clinical applications. Manual classification is time-consuming and subjective. This study proposes a capsule network-based model that achieves better performance and serves as an effective decision-making tool for medical diagnostics.
NEURAL PROCESSING LETTERS
(2022)
Article
Biotechnology & Applied Microbiology
Xufeng Yao, Kai Sun, Xixi Bu, Congyi Zhao, Yu Jin
Summary: The study proposed a TWO-DCNN method for WBC classification, achieving the best performance with high accuracy and robustness in low-resolution and noisy data sets. It can be used as an alternative method for clinical applications.
ARTIFICIAL CELLS NANOMEDICINE AND BIOTECHNOLOGY
(2021)
Article
Computer Science, Artificial Intelligence
Benteng Ma, Yu Feng, Geng Chen, Changyang Li, Yong Xia
Summary: Medical data sharing is crucial but suffers from privacy issues. This paper proposes a novel federated learning algorithm, FedAR, which addresses data heterogeneity by employing a flexible re-weighting scheme and achieves superior performance.
PATTERN RECOGNITION
(2023)
Article
Computer Science, Artificial Intelligence
G. Kalyani, B. Janakiramaiah, A. Karuna, L. V. Narasimha Prasad
Summary: An improved capsule network is developed for the detection and classification of diabetic retinopathy, achieving high accuracy on different stages of fundus images through feature extraction and probability estimation.
COMPLEX & INTELLIGENT SYSTEMS
(2023)
Article
Computer Science, Information Systems
Arman Haghanifar, Mahdiyar Molahasani Majdabadi, Sina Haghanifar, Younhee Choi, Seok-Bum Ko
Summary: Dental caries is a chronic disease that affects a large portion of the population throughout their lives. Detecting caries using dental x-rays can be challenging due to low image quality. In this study, we propose an automatic diagnosis system using deep pretrained models and a capsule network to detect dental caries in Panoramic images. Our model achieved an accuracy of 86.05% on a dataset of 470 Panoramic images, demonstrating its effectiveness in caries detection. The system has potential for assisting domain experts in a fully automated manner.
MULTIMEDIA TOOLS AND APPLICATIONS
(2023)
Article
Computer Science, Information Systems
Neeraj Baghel, Upendra Verma, Kapil Kumar Nagwanshi
Summary: White blood cells play a significant role in monitoring health conditions. Blood cell defects are responsible for numerous health conditions. The proposed automated method with machine learning improves accuracy and efficiency in classifying blood cells.
MULTIMEDIA TOOLS AND APPLICATIONS
(2022)
Article
Computer Science, Artificial Intelligence
Zakaria Senousy, Mohamed Medhat Gaber, Mohammed M. Abdelsamea
Summary: Deep learning algorithms can automate the examination of medical images in clinical practice. This paper proposes a new model called AUQantO, which improves the performance of deep learning architectures for medical image classification by optimizing uncertainty quantification techniques and excluding images with high levels of uncertainty.
APPLIED SOFT COMPUTING
(2023)
Article
Chemistry, Multidisciplinary
Yavuz Kahraman, Alptekin Durmusoglu
Summary: Fabric quality is crucial in the textile industry, and fabric defect detection plays a vital role in ensuring high-quality fabric. Researchers have developed computer vision-based methods to overcome the limitations of human-based defect detection. Capsule Networks, a new generation method, have been proposed as an alternative to Convolutional Neural Networks for deep learning tasks. Experimental results show that this method achieves high performance in fabric defect detection with minimal information loss.
APPLIED SCIENCES-BASEL
(2022)
Review
Environmental Sciences
Leiyu Chen, Shaobo Li, Qiang Bai, Jing Yang, Sanlong Jiang, Yanming Miao
Summary: This article summarizes the application of deep learning in image classification, covering the development of CNNs from their predecessors to the latest network architectures, as well as a comprehensive comparison and analysis of various image classification methods.
Article
Computer Science, Information Systems
Wenna Wu, Shengwu Liao, Zhentai Lu
Summary: This research proposes a white blood cell classification model based on radiomics and deep learning, which combines radiomic features and deep features to improve classification accuracy significantly.
Review
Biology
Huiyan Jiang, Zhaoshuo Diao, Tianyu Shi, Yang Zhou, Feiyu Wang, Wenrui Hu, Xiaolin Zhu, Shijie Luo, Guoyu Tong, Yu-Dong Yao
Summary: With the advancement of deep learning in natural image classification, detection, and segmentation, deep learning-based methods have become dominant in medical image processing. They have shown great effectiveness in single lesion recognition and segmentation. However, multiple-lesion recognition is more challenging due to the little variation or wide range of lesions involved. Recent studies have explored deep learning-based algorithms to tackle this challenge.
COMPUTERS IN BIOLOGY AND MEDICINE
(2023)
Article
Computer Science, Artificial Intelligence
Jianxiong Tang, Jian-Huang Lai, Xiaohua Xie, Lingxiao Yang, Wei-Shi Zheng
Summary: This paper proposes a fast and memory-efficient Activation Consistency Coupled ANN-SNN (AC2AS) framework for training SNN in low-power environments. The framework utilizes a weight-shared architecture between ANN and SNN, as well as spiking mapping units, to achieve fast training and ensure activation consistency for SNN. Experimental results show that AC2AS-based models perform well on benchmark datasets and achieve comparable accuracy with reduced time steps, training time, GPU memory costs, and spike activities compared to the Spike-based BP model.
PATTERN RECOGNITION
(2023)
Review
Biology
Yizhou Chen, Xu-Hua Yang, Zihan Wei, Ali Asghar Heidari, Nenggan Zheng, Zhicheng Li, Huiling Chen, Haigen Hu, Qianwei Zhou, Qiu Guan
Summary: This paper provides a comprehensive review and analysis of medical image augmentation, focusing on the advantages of different augmentation models, loss functions, and evaluation metrics. It explores the potential role of augmented models in limited training set scenarios and discusses the limitations and research directions in this field. The research shows that this field is still actively developing.
COMPUTERS IN BIOLOGY AND MEDICINE
(2022)
Article
Computer Science, Information Systems
Umit Atila, Yasin Ortakci, Kasim Ozacar, Emrullah Demiral, Ismail Rakip Karas
ISPRS INTERNATIONAL JOURNAL OF GEO-INFORMATION
(2018)
Article
Physics, Applied
Yusuf Yargi Baydilli, Ilker Turker
INTERNATIONAL JOURNAL OF MODERN PHYSICS B
(2019)
Article
Engineering, Multidisciplinary
Umit Atila, Murat Dorteder, Rafet Durgut, Ismail Sahin
ENGINEERING OPTIMIZATION
(2020)
Article
Computer Science, Interdisciplinary Applications
Umit Atila, Yusuf Yargi Baydilli, Eftal Sehirli, Muhammed Kamil Turan
COMPUTER METHODS AND PROGRAMS IN BIOMEDICINE
(2020)
Article
Computer Science, Artificial Intelligence
Farag Hamed Kuwil, Umit Atila, Radwan Abu-Issa, Fionn Murtagh
EXPERT SYSTEMS WITH APPLICATIONS
(2020)
Article
Engineering, Biomedical
M. Ucar, K. Akyol, U. Atila, E. Ucar
Summary: This study proposed a hybrid deep learning approach for automatic recognition of different tympanic membrane conditions from otoscopy images. The experimental results showed that the proposed model achieved high performance in classifying middle ear diseases.
Article
Ecology
Umit Atila, Murat Ucar, Kemal Akyol, Emine Ucar
Summary: The study introduces the use of EfficientNet deep learning architecture for classifying plant leaf diseases and compares its performance with other state-of-the-art deep learning models. Trained using the PlantVillage dataset, the results demonstrate that the B5 and B4 models of EfficientNet architecture outperformed other deep learning models in terms of accuracy and precision.
ECOLOGICAL INFORMATICS
(2021)
Article
Engineering, Biomedical
Emine Ucar, Umit Atila, Murat Ucar, Kemal Akyol
Summary: The study proposes a deep learning approach for rapid and accurate detection of Covid-19 on X-ray images, extracting deep features using pre-trained architectures and employing a two-stage classifier method for binary classification. The Bi-LSTM network showed superior performance with 92.489% accuracy compared to other classifiers, including well-known ensemble approaches.
BIOMEDICAL SIGNAL PROCESSING AND CONTROL
(2021)
Article
Engineering, Multidisciplinary
Uemit Atila, Furkan Sabaz
Summary: Lip-reading has become increasingly important in recent years, especially with the rise of deep learning applications. Researchers have developed automatic lip-reading systems for different languages, but there is still no study or dataset available for Turkish. This study aims to investigate the performance of state-of-the-art deep learning models on Turkish lip-reading, using two new datasets created with image processing techniques. The results show that the ResNet-18 and Bi-LSTM pair achieves the best accuracy in both word and sentence datasets.
ENGINEERING SCIENCE AND TECHNOLOGY-AN INTERNATIONAL JOURNAL-JESTECH
(2022)
Article
Computer Science, Artificial Intelligence
Murat Dorterler, Umit Atila, Rafet Durgut, Ismail Sahin
Summary: This study applied four different multiobjective metaheuristic algorithms to robot gripper design problems, achieving optimal results and examining the relationship between design variables and objective functions.
TURKISH JOURNAL OF ELECTRICAL ENGINEERING AND COMPUTER SCIENCES
(2021)
Article
Computer Science, Interdisciplinary Applications
Yusuf Yargi Baydilli, Umit Atila, Abdullah Elen
COMPUTER METHODS AND PROGRAMS IN BIOMEDICINE
(2020)