Related references
Note: Only part of the references are listed.
Review
Medicine, Research & Experimental
Jana L. Jacobs et al.
Summary: The COVID-19 pandemic has led to unexpected evolution and emergence of SARS-CoV-2 viral variants. Five major variants (Alpha, Beta, Gamma, Delta, and Omicron) have emerged, each with unique and overlapping amino acid substitutions that impact transmissibility, disease severity, and susceptibility to immune responses and monoclonal antibodies. Some recent variants exhibit immune evasion properties, particularly in cases of prolonged infection. Tracking existing variants and surveilling new ones are crucial for effective pandemic response.
ANNUAL REVIEW OF MEDICINE
(2023)
Article
Engineering, Electrical & Electronic
Hassan Mkindu et al.
Summary: Lung cancer is the leading cause of cancer-related death, and radiologists use computed tomography (CT) to diagnose lung nodules. Manual analysis of hundreds of CT images by radiologists is burdensome and sometimes inaccurate. This study proposes a CAD scheme based on 3D multi-scale vision transformer (3D-MSViT) to enhance feature extraction and improve lung nodule prediction efficiency.
SIGNAL IMAGE AND VIDEO PROCESSING
(2023)
Article
Multidisciplinary Sciences
Tao Qi et al.
Summary: Extracting useful knowledge from big data is crucial for machine learning. However, when the data is privacy-sensitive, federated learning provides a promising option to extract knowledge from decentralized data by exchanging model parameters. To address the privacy concerns, the authors propose PrivateKT, a method that uses selected small public data to transfer high-quality knowledge in federated learning with privacy guarantees. The results demonstrate that PrivateKT significantly closes the performance gap under strict differential privacy restrictions, offering a potential direction for privacy-preserving knowledge transfer in machine learning systems.
NATURE COMMUNICATIONS
(2023)
Article
Engineering, Biomedical
Yuan Yang et al.
Summary: The study proposes a lightweight segmentation network called SuperMini-Seg for automatically segmenting lung lesions from COVID-19 CT images, which is helpful in establishing a quantitative model for diagnosing and treating COVID-19. The network introduces a new module called the transformer parallel convolution module (TPCB) that combines transformer and convolution operations. It adopts a double-branch parallel structure to downsample the image and designs a gated attention mechanism in between. The model also incorporates the attentive hierarchical spatial pyramid (AHSP) module and criss-cross attention module, with over 100K parameters. Compared to other advanced methods, it achieves almost state-of-the-art segmentation accuracy and high calculation efficiency, making it suitable for practical deployment.
BIOMEDICAL SIGNAL PROCESSING AND CONTROL
(2023)
Article
Computer Science, Information Systems
Shangwang Liu et al.
Summary: This study proposes a multi-scale representation learning network (MRL-Net) that integrates CNN and Transformer to achieve accurate segmentation of COVID-19 lesions in lung CT images. Experimental results show that this method outperforms current state-of-the-art methods and achieves better performance in COVID-19 image segmentation.
IEEE JOURNAL OF BIOMEDICAL AND HEALTH INFORMATICS
(2023)
Article
Mathematical & Computational Biology
Haoran Lu et al.
Summary: This article investigates the effects of different parts of U-Net on segmentation ability in medical image segmentation and proposes a more efficient architecture, Half-UNet. Experiments show that Half-UNet achieves comparable segmentation accuracy to U-Net while significantly reducing parameters and floating-point operations.
FRONTIERS IN NEUROINFORMATICS
(2022)
Review
Critical Care Medicine
Stephen Machnicki et al.
Summary: The COVID-19 pandemic has had devastating medical and economic consequences globally. Chest CT imaging plays a crucial role in the management of patients with COVID-19, aiding in diagnosis, predicting disease progression, and guiding treatment. Summarizing the findings of CT scans can enhance understanding of the disease and improve patient care.
Article
Radiology, Nuclear Medicine & Medical Imaging
Michael Osadebey et al.
Summary: This research aims to improve the diagnostic accuracy of respiratory diseases by implementing each of the three stages of traditional methods using deep learning methods on five different configurations of training data. By utilizing ground truths from 3DIRCAD and ILD databases, along with data augmentation from LIDC-IDRI image collection and a realistic phantom, the proposed method achieved Dice-scores ranging from 0.76 to 0.95. The study emphasizes the importance of efficient and clinically relevant automated segmentation methods for the diagnosis of respiratory diseases, highlighting the potential benefits of deep learning models in improving segmentation accuracy and clinical relevance.
BMC MEDICAL IMAGING
(2021)
Article
Computer Science, Artificial Intelligence
Jesper E. Van Engelen et al.
Article
Computer Science, Interdisciplinary Applications
Deng-Ping Fan et al.
IEEE TRANSACTIONS ON MEDICAL IMAGING
(2020)
Article
Computer Science, Artificial Intelligence
Vijay Badrinarayanan et al.
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE
(2017)
Article
Computer Science, Artificial Intelligence
Evan Shelhamer et al.
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE
(2017)
Article
Computer Science, Artificial Intelligence
Sinno Jialin Pan et al.
IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING
(2010)
Article
Computer Science, Artificial Intelligence
JH Yao
INTERNATIONAL JOURNAL OF PATTERN RECOGNITION AND ARTIFICIAL INTELLIGENCE
(2003)