Article
Geosciences, Multidisciplinary
Christophe Ferrante, Lionel Cavin, Torsten Vennemann, Rossana Martini
Summary: A study was conducted on the histological and geochemical features of two Allosaurus bones found in Utah, United States. The results showed that the bones exhibited different growth characteristics with no direct correlation, potentially influenced by different environmental factors.
FRONTIERS IN EARTH SCIENCE
(2021)
Article
Geology
John I. Ejembi, Sally L. Potter-McIntyre, Anna Paltseva, Eric C. Ferre
Summary: The Wanakah Formation and Tidwell Member of the Morrison Formation in western Colorado during the Middle-Late Jurassic exhibit different types of paleosols and pedogenic carbonates, indicating variable redox conditions and a sub-humid to humid paleoclimate with seasonal precipitations. Clay mineralogy analysis suggests episodic wetting and drying, leading to the formation of illite from smectite alteration. The study provides new insights into the local and regional paleoclimatic and paleoenvironmental conditions during the Middle Jurassic in western Colorado.
INTERNATIONAL GEOLOGY REVIEW
(2023)
Article
Computer Science, Artificial Intelligence
Joris Fournel, Axel Bartoli, David Bendahan, Maxime Guye, Monique Bernard, Elisa Rauseo, Mohammed Y. Khanji, Steffen E. Petersen, Alexis Jacquier, Badih Ghattas
Summary: In this study, a 2D deep learning method was used for segmentation of cardiovascular MR images, with simultaneous quality control at 2D and 3D levels, outperforming 3D methods. Additionally, it was shown that this approach can enhance the accuracy of clinical measurements derived from image segmentations.
MEDICAL IMAGE ANALYSIS
(2021)
Article
Paleontology
Jorge Herrera-Flores, Thomas Stubbs, Francisco Sour-Tovar
Summary: This study reexamines and redescribes the type specimens and associated material of the Late Jurassic rhynchocephalian Opisthias rarus, and presents a fragment of the holotype for the first time. It also comments on undescribed material from the same quarry which may belong to an unnamed species of Opisthias. Additionally, a new specimen of Theretairus antiquus is described, rejecting its proposed status as a junior synonym of O. rarus, and contributing to the knowledge of the Late Jurassic microvertebrate fauna of North America.
ACTA PALAEONTOLOGICA POLONICA
(2022)
Article
Energy & Fuels
Tingyu Du, Dongxing Wang, Xu Qian
Summary: This paper briefly describes the image analysis technology needed in the construction of intelligent coal mine, with a focus on the reconstruction process of 3-dimensional image model in coal mine. The proposed method involves target detection or image segmentation, image registration, diffusion and fusion of point matrices, leading to the generation of a 3D model. The improved registration algorithm using swarm intelligence significantly reduces the registration region and improves the accuracy of depth data obtained in 3D reconstruction, which is crucial for subsequent scene analysis.
Article
Engineering, Biomedical
Yanlin Wu, Guanglei Wang, Zhongyang Wang, Hongrui Wang, Yan Li
Summary: In recent years, the Unet network based on convolution has become a popular structure for medical image segmentation tasks. However, its limited receptive field hinders effective modeling of long-distance feature dependence. To address this issue, this study proposes DI-Unet, which incorporates Dimensional Interactive (DI) self-attention for capturing cross-dimensional information while reducing computation.
BIOMEDICAL SIGNAL PROCESSING AND CONTROL
(2022)
Article
Computer Science, Artificial Intelligence
Feng Chen, Zhiheng Chen, Yuxiao Du, Zhuocheng Wu, Yuxing Li, Qi Hu
Summary: This study proposes a method of image deformation by body part size to reduce the cost and increase the personalized effect of virtual try-on. The method employs image segmentation algorithm to snap out the garment and adjusts its size and position according to the standard mannequin image. Experimental results demonstrate that this method can exhibit good performance in personalized try-on.
INTERNATIONAL JOURNAL OF INTELLIGENT SYSTEMS
(2022)
Article
Radiology, Nuclear Medicine & Medical Imaging
Ahmed S. Fahmy, Ethan J. Rowin, Raymond H. Chan, Warren J. Manning, Martin S. Maron, Reza Nezafat
Summary: The deep learning model combining LGE and cine images improves the accuracy of myocardium scarring quantification in cardiac magnetic resonance imaging, compared to using LGE images alone.
JOURNAL OF MAGNETIC RESONANCE IMAGING
(2021)
Article
Biology
Yasmina Al Khalil, Sina Amirrajab, Cristian Lorenz, Juergen Weese, Josien Pluim, Marcel Breeuwer
Summary: Cardiac magnetic resonance (CMR) image segmentation is crucial for analyzing cardiac function and diagnosing heart-related diseases. Existing deep learning-based automatic segmentation methods are not always applicable to realistic clinical scenarios, due to homogeneous training datasets and lack of variation. This work presents a model for segmenting all three cardiac structures in a multi-center, multi-disease, and multi-view scenario. The proposed approach effectively addresses the challenges of such heterogeneous data through heart region detection, image synthesis augmentation, and late-fusion segmentation. Extensive experiments demonstrate the ability of the model to handle outlier cases and achieve better consistency in clinical parameter estimation.
COMPUTERS IN BIOLOGY AND MEDICINE
(2023)
Article
Biology
Jiaxin Liu, Xiang Zhou, Wenxue Guan, Shenghua Gong, Jun Liu
Summary: This study utilizes parametric color imaging processing on DSA images to assist doctors in accurately and quickly diagnosing vascular bleeding points.
COMPUTERS IN BIOLOGY AND MEDICINE
(2022)
Article
Mathematical & Computational Biology
Meera Srikrishna, Rolf A. Heckemann, Joana B. Pereira, Giovanni Volpe, Anna Zettergren, Silke Kern, Eric Westman, Ingmar Skoog, Michael Scholl
Summary: Brain tissue segmentation is crucial for analyzing brain scans, with CT being a more accessible modality compared to MRI. The study developed and compared 2D and 3D deep learning models for brain tissue classification in CT scans, finding that 2D models performed better than 3D models.
FRONTIERS IN COMPUTATIONAL NEUROSCIENCE
(2022)
Article
Computer Science, Artificial Intelligence
Hicham Messaoudi, Ahror Belaid, Douraied Ben Salem, Pierre-Henri Conze
Summary: Convolutional neural networks (CNNs) have significantly advanced image analysis and computer vision applications in the past decade. However, progress in medical image analysis has been slowed down by limited annotated data and acquisition constraints. This paper proposes an efficient way to transfer the efficiency of a 2D classification network to 2D and 3D medical image segmentation applications, achieving promising results in various benchmarks.
MEDICAL IMAGE ANALYSIS
(2023)
Article
Engineering, Electrical & Electronic
Lixin He, Wangwei Liu, Yiming Li, Handong Wang, Shenjie Cao, Chengying Zhou
Summary: In this work, we propose a new method for crack image detection and segmentation that addresses the issues of poor crack structure detection in complex background conditions and loss of details in segmentation. The method consists of two phases: the coding phase, which uses the channel attention mechanism and crack characteristics to enhance feature extraction, and the decoding phase, which uses the spatial attention mechanism to capture crack edge information and achieve accurate crack positioning through image information fusion. Experimental results show that our method outperforms existing methods in terms of crack segmentation accuracy, with mean intersection over the union ratios of 87.2% and 83.9% on public and self-built datasets, respectively.
Article
Engineering, Biomedical
Weijin Xu, Huihua Yang, Mingying Zhang, Zhiwei Cao, Xipeng Pan, Wentao Liu
Summary: Accurate brain tumor segmentation is a challenging task due to the diversity and complexity of tumors. This article proposes a method based on a corner attention module and high-dimensional perceptual loss to improve the accuracy of brain tumor segmentation. Experimental results show that the method achieves competitive performance on various benchmark tests.
BIOMEDICAL SIGNAL PROCESSING AND CONTROL
(2022)
Article
Radiology, Nuclear Medicine & Medical Imaging
Hyun Woo Goo
Summary: The study demonstrates that quantifying ventricular volume percentage using 3D brain CT data is highly accurate and reproducible for interpreting serial changes in hydrocephalus.
KOREAN JOURNAL OF RADIOLOGY
(2021)