TDD-UNet:Transformer with double decoder UNet for COVID-19 lesions segmentation

Article Computer Science, Artificial Intelligence

Crosslink-Net: Double-Branch Encoder Network via Fusing Vertical and Horizontal Convolutions for Medical Image Segmentation

Qian Yu et al.

Summary: Accurate image segmentation plays a crucial role in medical image analysis but faces challenges due to various shapes, sizes, and blurry boundaries. To address these challenges, a novel double-branch encoder architecture is proposed and validated for medical image segmentation.

IEEE TRANSACTIONS ON IMAGE PROCESSING (2022)

Add to Collection

Article Computer Science, Interdisciplinary Applications

CA-Net: Comprehensive Attention Convolutional Neural Networks for Explainable Medical Image Segmentation

Ran Gu et al.

Summary: The study presented a comprehensive attention-based CNN (CA-Net) for medical image segmentation, which significantly improved accuracy and explainability. The CA-Net achieved better segmentation results for skin lesions, placenta, and fetal brain compared to the U-Net model, while reducing the model size by approximately 15 times.

IEEE TRANSACTIONS ON MEDICAL IMAGING (2021)

Add to Collection

Article Radiology, Nuclear Medicine & Medical Imaging

Automatic segmentation of hippocampus in hippocampal sparing whole brain radiotherapy: A multitask edge-aware learning

Qingtao Qiu et al.

Summary: The study aimed to improve hippocampus segmentation accuracy through multitask edge-aware learning, achieving significant results compared to baseline segmentation. The proposed framework may have applications in other low-contrast small organ segmentations in medical imaging modalities.

MEDICAL PHYSICS (2021)

Add to Collection

Article Computer Science, Artificial Intelligence

Federated semi-supervised learning for COVID region segmentation in chest CT using multi-national data from China, Italy, Japan

Dong Yang et al.

Summary: This study explores the use of federated and semi-supervised learning techniques to address the variability in data and annotations in COVID-19 detection. Through experiments on a multinational database, it is found that federated learning protects data privacy and semi-supervised learning helps reduce annotation burden. This novel approach shows promising results in transfer learning.

MEDICAL IMAGE ANALYSIS (2021)

Add to Collection

Article Radiology, Nuclear Medicine & Medical Imaging