Article
Computer Science, Artificial Intelligence
Tianyu Ma, Alan Q. Wang, Adrian V. Dalca, Mert R. Sabuncu
Summary: The convolutional neural network (CNN) is a commonly used architecture for computer vision tasks. A new building block called hyper-convolution is presented in this paper, which encodes the convolutional kernel using spatial coordinates and enables a more flexible architecture design. Experimental results showed that replacing regular convolutions with hyper-convolutions improved performance with fewer parameters and increased robustness against noise.
MEDICAL IMAGE ANALYSIS
(2023)
Article
Computer Science, Information Systems
Khaled Almezhghwi, Sertan Serte, Fadi Al-Turjman
Summary: The study proposes two artificial intelligence approaches utilizing deep learning for the classification of chest X-ray images. These methods, based on the AlexNet model and VGGNet16 method, can accurately identify lung diseases.
MULTIMEDIA TOOLS AND APPLICATIONS
(2021)
Article
Computer Science, Interdisciplinary Applications
Vishnu Preetham Revelli, Gauri Sharma, S. Kiruthika Devi
Summary: This project aims to extract text from braille text images and provide translated English text and audio output using a customized CNN model. The CNN model demonstrates robustness in image recognition and classification tasks, making it valuable for addressing challenges faced by visually impaired individuals.
ADVANCES IN ENGINEERING SOFTWARE
(2022)
Article
Computer Science, Artificial Intelligence
Hejun Jiang, Xiaolan Tang, Kai Jin, Wenlong Chen, Juhua Pu
Summary: In this paper, a solution combining maximum flow and deep neural networks, called CAF-Net, is proposed for the content delivery problem in vehicular networks. Experimental results show that ResNet 50 has the smallest error and significantly reduces the computation time for delivery ratio, demonstrating the feasibility of applying deep learning to vehicular networks.
Article
Computer Science, Information Systems
Tomasz Szandala
Summary: Technology has advanced rapidly in recent years, leading to the introduction of new solutions based on Machine Learning and Artificial Intelligence on a daily basis. However, understanding how these models make decisions has become challenging due to their complex black box decision-making process. Therefore, explainable artificial intelligence methods are crucial for further development. This paper discusses the need to revise existing state-of-the-art techniques in order to fully comprehend the prediction-generating process, and compares them with a new method called PRISM, which utilizes Principal Component Analysis for visualizing important features recognized by a given Convolutional Neural Network. The main objective of this paper is to examine how PRISM enhances the understanding of the decision-making process and introduce a tool called TorchPRISM for analyzing the output.
INFORMATION SCIENCES
(2023)
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
Engineering, Aerospace
Jiawei Hu, Weiwei Zhang
Summary: A method is proposed for flow field modeling using Convolutional Neural Networks (CNNs) to address the limitations of non-orthogonal and non-uniform meshes commonly used in numerical simulation. By transforming the flow field from the non-uniform physical plane to the uniform computational plane, the method achieves high accuracy and interpretability, providing an efficient solution for parameter space research.
AEROSPACE SCIENCE AND TECHNOLOGY
(2023)
Article
Computer Science, Artificial Intelligence
Umar Asif, Deval Mehta, Stefan Von Cavallar, Jianbin Tang, Stefan Harrer
Summary: This paper presents a holistic framework for video-based action recognition by combining spatial and motion features from the body, face, and hands. The proposed Deep Actions Stamps (DeepActs) encode effective spatio-temporal features and improve action recognition accuracy compared to methods based on limited body joints. The DeepActsNet, a deep learning based ensemble model, achieves highly accurate action recognition with less computational cost.
PATTERN RECOGNITION
(2023)
Article
Automation & Control Systems
Wen Xin Cheng, Ruobin Gao, P. N. Suganthan, Kum Fai Yuen
Summary: Emotion recognition based on EEG signals is crucial in medical healthcare, helping diagnose emotional disorders in patients. A randomized CNN model is proposed to improve emotion recognition performance without the need for backpropagation.
ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE
(2022)
Article
Biology
Guy M. Hagen, Justin Bendesky, Rosa Machado, Tram-Anh Nguyen, Tanmay Kumar, Jonathan Ventura
Summary: Fluorescence microscopy is important in biological research, but photobleaching and phototoxicity are limiting factors. Machine learning methods can improve signal-to-noise ratio and reduce phototoxicity. High-quality data is essential for training deep learning methods.
Article
Radiology, Nuclear Medicine & Medical Imaging
Hamidreza Bolhasani, Somayyeh Jafarali Jassbi, Arash Sharifi
Summary: Deep learning has become an intriguing topic in various fields for more than a decade, especially in the area of image classification. However, the heavy computational operations in deep neural networks pose challenges in terms of power consumption and runtime. This paper proposes a deep learning accelerator and its data flow for histopathologic image classification, which exhibits significant performance improvement compared to current general-purpose accelerators and data flows, as demonstrated by simulation results.
JOURNAL OF DIGITAL IMAGING
(2023)
Article
Biochemistry & Molecular Biology
Iman Hamid, Katharine L. Korunes, Daniel R. Schrider, Amy Goldberg
Summary: Gene flow can introduce adaptive alleles into new populations, and ancestry patterns have been used to identify post-admixture positive selection. However, current methods have limitations in accurately identifying selection regions. To address this, we propose a deep learning object detection method applied to local ancestry-painted genomes. This method shows robustness to various demographic scenarios and successfully localizes known adaptive loci to narrow regions.
MOLECULAR BIOLOGY AND EVOLUTION
(2023)
Article
Mathematics
Freddy Gabbay, Gil Shomron
Summary: The study introduces a value-locality-based compression algorithm called VELCRO for neural networks, which efficiently compresses networks deployed for specialized tasks to improve computational efficiency. VELCRO consists of two stages - preprocessing and compression, saving computation and avoiding processing of output feature map elements through replacing activation function values with average arithmetic values.
Article
Computer Science, Artificial Intelligence
Francesco Ponzio, Enrico Macii, Elisa Ficarra, Santa Di Cataldo
Summary: In real-world scenarios, training Convolutional Neural Networks (CNNs) with high quality images and correct labels is difficult. This affects the performance of CNNs during both training and inference. To tackle this issue, we propose a new two-module CNN called Wise2WipedNet (W2WNet), which uses Bayesian inference to identify and discard spurious images during training and provides prediction confidence during inference. Our experiments on various image classification tasks and histological image analysis demonstrate that W2WNet can effectively identify image degradation and mislabelling issues, resulting in improved classification accuracy.
EXPERT SYSTEMS WITH APPLICATIONS
(2023)
Review
Biochemical Research Methods
Yurui Chen, Louxin Zhang
Summary: This article introduces the application of deep learning in drug response prediction and summarizes the latest deep learning methods. Although deep learning methods have shown some limitations in certain cases, combining them with established bioinformatics analyses can help overcome some of these challenges.
BRIEFINGS IN BIOINFORMATICS
(2022)
Review
Plant Sciences
Gane Ka-Shu Wong, Douglas E. Soltis, Jim Leebens-Mack, Norman J. Wickett, Michael S. Barker, Yves Van de Peer, Sean W. Graham, Michael Melkonian
ANNUAL REVIEW OF PLANT BIOLOGY, VOL 71, 2020
(2020)
Article
Plant Sciences
Qundan Lv, Jie Qiu, Jie Liu, Zheng Li, Wenting Zhang, Qin Wang, Jie Fang, Junjie Pan, Zhengdao Chen, Wenliang Cheng, Michael S. Barker, Xuehui Huang, Xin Wei, Kejun Cheng
Article
Plant Sciences
Xinshuai Qi, Hong An, Tara E. Hall, Chenlu Di, Paul D. Blischak, Michael T. W. McKibben, Yue Hao, Gavin C. Conant, J. Chris Pires, Michael S. Barker
Summary: The study found a close relationship between domestication and polyploidy in Brassica rapa crops, with genetic diversity derived from ancient polyploidy playing a key role in the domestication of B. rapa and supporting its importance in the success of modern agriculture.
Review
Plant Sciences
Zheng Li, Michael T. W. McKibben, Geoffrey S. Finch, Paul D. Blischak, Brittany L. Sutherland, Michael S. Barker
Summary: This review discusses the impact of polyploidy on chromosome pairing behavior in land plants, as well as the two major processes of diploidization: cytological diploidization and genic diploidization/fractionation. It also compares gene fractionation variation across land plants and highlights differences in diploidization between plants and animals.
ANNUAL REVIEW OF PLANT BIOLOGY, VOL 72, 2021
(2021)
Article
Plant Sciences
Hannah E. Marx, Stacy A. Jorgensen, Eldridge Wisely, Zheng Li, Katrina M. Dlugosch, Michael S. Barker
Summary: This study involved generating and analyzing RNA-seq data for 24 vascular plant species, highlighting the challenges of collecting RNA data from diverse plant communities and revealing no significant differences in transcriptome quality between diploid and polyploid species. The findings provide opportunities for future large-scale studies at the intersection of ecology and genomics.
APPLICATIONS IN PLANT SCIENCES
(2021)
Article
Biochemistry & Molecular Biology
Yue Hao, Makenzie E. Mabry, Patrick P. Edger, Michael Freeling, Chunfang Zheng, Lingling Jin, Robert VanBuren, Marivi Colle, Hong An, R. Shawn Abrahams, Jacob D. Washburn, Xinshuai Qi, Kerrie Barry, Christopher Daum, Shengqiang Shu, Jeremy Schmutz, David Sankoff, Michael S. Barker, Eric Lyons, J. Chris Pires, Gavin C. Conant
Summary: The study investigates the gene loss history after whole-genome triplication (WGT) in Brassiceae tribe members, confirming a two-step formation model with significant temporal gaps. It highlights distinguishable homoeolog loss rates among subgenomes and proposes a mix and match model of allopolyploidy where genes from different subgenomes function together without difficulty.
Article
Ecology
Cristian Roman-Palacios, Cesar A. Medina, Shing H. Zhan, Michael S. Barker
Summary: Understanding the mechanisms behind chromosome evolution can provide insights into lineage origin, persistence, and evolutionary tempo. A database of chromosome counts for animals was presented, showing similarities in distribution with flowering plants, though driven by different factors. Animals and plants exhibit similar frequencies of speciation-related changes in chromosome number, but plant speciation is more often associated with changes in ploidy.
JOURNAL OF EVOLUTIONARY BIOLOGY
(2021)
Article
Biotechnology & Applied Microbiology
Philipp E. Bayer, Armin Scheben, Agnieszka A. Golicz, Yuxuan Yuan, Sebastien Faure, HueyTyng Lee, Harmeet Singh Chawla, Robyn Anderson, Ian Bancroft, Harsh Raman, Yong Pyo Lim, Steven Robbens, Lixi Jiang, Shengyi Liu, Michael S. Barker, M. Eric Schranz, Xiaowu Wang, Graham J. King, J. Chris Pires, Boulos Chalhoub, Rod J. Snowdon, Jacqueline Batley, David Edwards
Summary: Plant genomes show significant presence/absence variation (PAV) within a species, with different causes of gene loss between diploids and polyploids. In diploids, gene loss propensity is primarily associated with transposable elements, while in polyploids like B. napus, gene loss propensity is linked to homoeologous recombination. These findings provide insights into the underlying biological and physical factors of gene presence/absence, paving the way for the application of machine learning methods in the field.
PLANT BIOTECHNOLOGY JOURNAL
(2021)
Article
Multidisciplinary Sciences
Yifei Liu, Bo Wang, Shaohua Shu, Zheng Li, Chi Song, Di Liu, Yan Niu, Jinxin Liu, Jingjing Zhang, Heping Liu, Zhigang Hu, Bisheng Huang, Xiuyu Liu, Wei Liu, Liping Jiang, Mohammad Murtaza Alami, Yuxin Zhou, Yutao Ma, Xiangxiang He, Yicheng Yang, Tianyuan Zhang, Hui Hu, Michael S. Barker, Shilin Chen, Xuekui Wang, Jing Nie
Summary: Chinese goldthread (Coptis chinensis) is an early-diverging eudicot plant with diverse medicinal applications. The high-quality genome assembly and annotation of C. chinensis revealed a single ancient whole-genome duplication event shared by the Ranunculaceae family. The study also highlighted the functional importance of CYP719 gene in diversifying protoberberine-type alkaloids.
NATURE COMMUNICATIONS
(2021)
Article
Evolutionary Biology
David E. Jarvis, Peter J. Maughan, Joseph DeTemple, Veronica Mosquera, Zheng Li, Michael S. Barker, Leigh A. Johnson, Clinton J. Whipple
Summary: This study used the chromosome-scale reference genome of Gilia yorkii to investigate genome evolution in the Polemoniaceae and identified important genes related to inflorescence architecture and flower color variation through quantitative trait loci mapping. The results demonstrate that Gilia can serve as a genetic model for studying the evolution of development in plants.
GENOME BIOLOGY AND EVOLUTION
(2022)
Article
Multidisciplinary Sciences
David Wickell, Li-Yaung Kuo, Hsiao-Pei Yang, Amra Dhabalia Ashok, Iker Irisarri, Armin Dadras, Sophie de Vries, Jan de Vries, Yao-Moan Huang, Zheng Li, Michael S. Barker, Nolan T. Hartwick, Todd P. Michael, Fay-Wei Li
Summary: Despite the extensive characterization of crassulacean acid metabolism (CAM) in terrestrial angiosperms, little attention has been given to aquatics and early diverging land plants. Here, the authors assemble the genome of Isoetes taiwanensis and investigate the genetic factors driving CAM in this aquatic lycophyte. Despite broad similarities between CAM in Isoetes and terrestrial angiosperms, several key differences are identified, including the recruitment of 'bacterial-type' PEPC and diverged circadian control of key CAM pathway genes in Isoetes.
NATURE COMMUNICATIONS
(2021)
Article
Biology
Jacqueline Heckenhauer, Paul B. Frandsen, John S. Sproul, Zheng Li, Juraj Paule, Amanda M. Larracuente, Peter J. Maughan, Michael S. Barker, Julio Schneider, Russell J. Stewart, Steffen U. Pauls
Summary: The size of genomes in caddisflies varies greatly, and the expansion of repetitive elements, particularly transposable elements, is identified as a major driver of larger genome sizes. The association between transposable elements and genome size shows a linear relationship. Moreover, expanded genomes are more likely to occur in caddisfly lineages with higher ecological diversity.