检索结果-内蒙古大学图书馆

Position: Bayesian Deep Learning is Needed in the Age of Large-Scale AI

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Papamarkou, Theodore Skoularidou, Maria Palla, Konstantina Aitchison, Laurence Arbel, Julyan Dunson, David Filippone, Maurizio Fortuin, Vincent Hennig, Philipp Hernández-Lobato, José Miguel Hubin, Aliaksandr Immer, Alexander Karaletsos, Theofanis Khan, Mohammad Emtiyaz Kristiadi, Agustinus Li, Yingzhen Mandt, Stephan Nemeth, Christopher Osborne, Michael A. Rudner, Tim G.J. Rügamer, David Teh, Yee Whye Welling, Max Wilson, Andrew Gordon Zhang, Ruqi Department of Mathematics The University of Manchester Manchester United Kingdom Eric and Wendy Schmidt Center Broad Institute of MIT and Harvard Cambridge United States Spotify London United Kingdom Computational Neuroscience Unit University of Bristol Bristol United Kingdom Centre Inria de l'Université Grenoble Alpes Grenoble France Department of Statistical Science Duke University United States Statistics Program KAUST Saudi Arabia Helmholtz AI Munich Germany Department of Computer Science Technical University of Munich Munich Germany Munich Center for Machine Learning Munich Germany Tübingen AI Center University of Tübingen Tübingen Germany Department of Engineering University of Cambridge Cambridge United Kingdom Department of Mathematics University of Oslo Oslo Norway Bioinformatics and Applied Statistics Norwegian University of Life Sciences Ås Norway Department of Computer Science ETH Zurich Switzerland Chan Zuckerberg Initiative CA United States Center for Advanced Intelligence Project RIKEN Tokyo Japan Vector Institute Toronto Canada Department of Computing Imperial College London London United Kingdom Department of Computer Science UC Irvine Irvine United States Department of Mathematics and Statistics Lancaster University Lancaster United Kingdom Department of Engineering Science University of Oxford Oxford United Kingdom Center for Data Science New York University New York United States Department of Statistics LMU Munich Munich Germany DeepMind London United Kingdom Department of Statistics University of Oxford Oxford United Kingdom Informatics Institute University of Amsterdam Amsterdam Netherlands Courant Institute of Mathematical Sciences Center for Data Science Computer Science Department New York University New York United States Department of Computer Science Purdue University West Lafayette United States

In the current landscape of deep learning research, there is a predominant emphasis on achieving high predictive accuracy in supervised tasks involving large image and language datasets. However, a broader perspective reveals a multitude of overlooked metrics, tasks, and data types, such as uncertainty, active and continual learning, and scientific data, that demand attention. Bayesian deep learning (BDL) constitutes a promising avenue, offering advantages across these diverse settings. This paper posits that BDL can elevate the capabilities of deep learning. It revisits the strengths of BDL, acknowledges existing challenges, and highlights some exciting research avenues aimed at addressing these obstacles. Looking ahead, the discussion focuses on possible ways to combine large-scale foundation models with BDL to unlock their full potential. © 2024, CC BY-NC-ND.

关键词： Large datasets

The role of input imaging combination and ADC threshold on segmentation of acute ischemic stroke lesion using U-Net

学校读者我要写书评

暂无评论

European Radiology 2023年第9期33卷 6157-6167页

作者： Li, Ya-Hui Lin, Shao-Chieh Chung, Hsiao-Wen Chang, Chia-Ching Peng, Hsu-Hsia Huang, Teng-Yi Shen, Wu-Chung Tsai, Chon-Haw Lo, Yu-Chien Lee, Tung-Yang Juan, Cheng-Hsuan Juan, Cheng-En Chang, Hing-Chiu Liu, Yi-Jui Juan, Chun-Jung Graduate Institute of Biomedical Electronics and Bioinformatics National Taiwan University Taipei Taiwan Department of Medical Imaging China Medical University Hsinchu Hospital No. 199 Sec. 1 Xinglong Rd. Zhubei City Hsinchu County 302 Hsinchu Taiwan Ph.D. Program in Electrical and Communication Engineering Feng Chia University Taichung Taiwan Department of Electrical Engineering National Taiwan University Taipei Taiwan Department of Management Science National Yang Ming Chiao Tung University Hsinchu Taiwan Department of Biomedical Engineering and Environmental Sciences National Tsing Hua University Hsinchu Taiwan Department of Electrical Engineering National Taiwan University of Science and Technology Taipei Taiwan Department of Radiology School of Medicine College of Medicine China Medical University Taichung Taiwan Department of Medical Imaging Medical University Hospital Taichung Taiwan Department of Neurology China Medical University Hospital Taichung Taiwan Cheng Ching Hospital Taichung Taiwan Master’s Program of Biomedical Informatics and Biomedical Engineering Feng Chia University Taichung Taiwan Department of Automatic Control Engineering Feng Chia University No. 100 Wenhwa Rd. Seatwen Taichung 40724 Taiwan Department of Biomedical Engineering The Chinese University of Hong Kong ERB1112 11/F William M.W. Mong Engineering Building N.T Shatin Hong Kong Multi-Scale Medical Robotics Center The Chinese University of Hong Kong N.T Shatin Hong Kong Department of Biomedical Engineering National Defense Medical Center Taipei Taiwan Department of Computer Science and Information Engineering National Taiwan University Taipei Taiwan

Background: To evaluate the effect of the weighting of input imaging combo and ADC threshold on the performance of the U-Net and to find an optimized input imaging combo and ADC threshold in segmenting acute ischemic stroke (AIS) lesion. Methods: This study retrospectively enrolled a total of 212 patients having AIS. Four combos, including ADC-ADC-ADC (AAA), DWI-ADC-ADC (DAA), DWI-DWI-ADC (DDA), and DWI-DWI-DWI (DDD), were used as input images, respectively. Three ADC thresholds including 0.6, 0.8 and 1.8 × 10^–3 mm²/s were applied. Dice similarity coefficient (DSC) was used to evaluate the segmentation performance of U-Nets. Nonparametric Kruskal–Wallis test with Tukey–Kramer post-hoc tests were used for comparison. A p <.05 was considered statistically significant. Results: The DSC significantly varied among different combos of images and different ADC thresholds. Hybrid U-Nets outperformed uniform U-Nets at ADC thresholds of 0.6 × 10^–3 mm²/s and 0.8 × 10^–3 mm²/s (p <.001). The U-Net with imaging combo of DDD had segmentation performance similar to hybrid U-Nets at an ADC threshold of 1.8 × 10^–3 mm²/s (p =.062 to 1). The U-Net using the imaging combo of DAA at the ADC threshold of 0.6 × 10^–3 mm²/s achieved the highest DSC in the segmentation of AIS lesion. Conclusions: The segmentation performance of U-Net for AIS varies among the input imaging combos and ADC thresholds. The U-Net is optimized by choosing the imaging combo of DAA at an ADC threshold of 0.6 × 10^–3 mm²/s in segmentating AIS lesion with highest DSC. Key Points: • Segmentation performance of U-Net for AIS differs among input imaging combos. • Segmentation performance of U-Net for AIS differs among ADC thresholds. • U-Net is optimized using DAA with ADC = 0.6 × 10^–3mm²/s. © 2023, The Author(s), under exclusive licence to European Society of Radiology.

关键词： Deep Learning Diffusion Magnetic Resonance Imaging Ischemic Stroke Neural Networks, computer Retrospective Study

Count time series modelling of Twitter data topic modelling: A case of Indonesia flood events

学校读者我要写书评

暂无评论

AIP Conference Proceedings 2023年第1期2594卷

作者： A. A. Hidayat B Pardamean Bioinformatics & Data Science Research Center Bina Nusantara University Jakarta Indonesia 11480 Computer Science Department BINUS Graduate Program - Master of Computer Science Bina Nusantara University Jakarta Indonesia 11480

Twitter data provide rich and powerful information to leverage the dynamics of public perception to establish situational awareness and disaster mitigation strategies during critical times. In this paper, we perform topic modeling via Latent Dirichlet Allocation to extract topics from a collection of tweets related to Indonesia flood events in February 2021 with the query: “banjir”. The extracted topics are used as one of the features to build a generalized linear count time series model with Negative Binomial distribution. We find seven major topics from the model in which tweets containing a topic about the government’s role in handling the situation dominate the conversation. Taking into account a simple intervention analysis, we demonstrate a statistically significant change in the users’ behavior before and after the severe Jakarta flood on 20 February 2021. Moreover, a metric evaluation demonstrates that a covariate that describes the turning point of the Jakarta flood event is convenient to build a more accurate count time series model of the tweets.

关键词：

Estimating Enzyme Participation in Metabolic Pathways for Microbial Communities from RNA-seq Data 16th

学校读者我要写书评

暂无评论

Estimating Enzyme Participation in Metabolic Pathways for Mi...

16th International Symposium on bioinformatics Research and Applications, ISBRA 2020

作者： Rondel, F. Hosseini, R. Sahoo, B. Knyazev, S. Mandric, I. Stewart, Frank Măndoiu, I.I. Pasaniuc, B. Zelikovsky, A. Department of Computer Science Georgia State University Atlanta United States Department of Computer Science University of California Los Angeles Los AngelesCA United States Department of Microbiology and Immunology Montana State University BozemanMT United States Computer Science and Engineering Department University of Connecticut StorrsCT United States Bioinformatics Interdepartmental Program University of California Los AngelesCA United States

ISBN: (纸本)9783030578206

Metatranscriptome sequence data analysis is necessary for understanding biochemical changes in the microbial community and their effects. In this paper, we propose a methodology to estimate activities of individual metabolic pathways to better understand the activity of the entire metabolic network. Our novel pipeline includes an expectation-maximization based estimation of enzyme expression and simultaneous estimation of pathway activity level and enzyme participation level in each pathway. We applied our novel pipeline to metatranscriptome data generated from surface water planktonic communities sampled over a day-night cycle in the Northern Gulf of Mexico (Louisiana Shelf). Our results show the estimated enzyme expression, pathway activity levels as well as enzyme participation levels in each pathway are robust and stable across all data points. In contrast to expression of enzymes, the estimated activity levels of significant number of metabolic pathways strongly correlate with the environmental parameters. © 2020, Springer Nature Switzerland AG.

关键词： Pipelines

Building an Ethical and Trustworthy Biomedical AI Ecosystem for the Translational and Clinical Integration of Foundational Models

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Sankar, Baradwaj Simha Gilliland, Destiny Rincon, Jack Hermjakob, Henning Yan, Yu Adam, Irsyad Lemaster, Gwyneth Wang, Dean Watson, Karol Bui, Alex Wang, Wei Ping, Peipei Department of Physiology University of California Los AngelesCA United States NIH BRIDGE2AI Center UCLA NHLBI Integrated Cardiovascular Data Science Training Program at UCLA Los AngelesCA United States Cambridge United Kingdom Bioinformatics IDP University of California Los AngelesCA United States University of California Los AngelesCA United States Medical Informatics Home Area University of California Los AngelesCA United States Department of Computer Science University of California Los AngelesCA United States

Foundational Models (FMs) are gaining increasing attention in the biomedical AI ecosystem due to their ability to represent and contextualize multimodal biomedical data. These capabilities make FMs a valuable tool for a variety of tasks, including biomedical reasoning, hypothesis generation, and interpreting complex imaging data. In this review paper, we address the unique challenges associated with establishing an ethical and trustworthy biomedical AI ecosystem, with a particular focus on the development of FMs and their downstream applications. We explore strategies that can be implemented throughout the biomedical AI pipeline to effectively tackle these challenges, ensuring that these FMs are translated responsibly into clinical and translational settings. Additionally, we emphasize the importance of key stewardship and co-design principles that not only ensure robust regulation but also guarantee that the interests of all stakeholders-especially those involved in or affected by these clinical and translational applications-are adequately represented. We aim to empower the biomedical AI community to harness these models responsibly and effectively. As we navigate this exciting frontier, our collective commitment to ethical stewardship, co-design, and responsible translation will be instrumental in ensuring that the evolution of FMs truly enhances patient care and medical decision-making, ultimately leading to a more equitable and trustworthy biomedical AI ecosystem. © 2024, CC BY.

关键词：

Evaluation of GPT and BERT-based models on identifying protein-protein interactions in biomedical text

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Rehana, Hasin Çam, Nur Bengisu Basmaci, Mert Zheng, Jie Jemiyo, Christianah He, Yongqun Özgür, Arzucan Hur, Junguk Computer Science Graduate Program University of North Dakota Grand ForksND58202 United States Department of Computer Engineering Bogazici University Istanbul34342 Turkey Unit for Laboratory Animal Medicine Department of Microbiology and Immunology University of Michigan Ann ArborMI48109 United States Department of Biomedical Sciences University of North Dakota School of Medicine and Health Sciences Grand ForksND58202 United States Center for Computational Medicine and Bioinformatics University of Michigan Ann ArborMI48109 United States

Detecting protein-protein interactions (PPIs) is crucial for understanding genetic mechanisms, disease pathogenesis, and drug design. However, with the fast-paced growth of biomedical literature, there is a growing need for automated and accurate extraction of PPIs to facilitate scientific knowledge discovery. Pre-trained language models, such as generative pre-trained transformers (GPT) and bidirectional encoder representations from transformers (BERT), have shown promising results in natural language processing (NLP) tasks. We evaluated the performance of PPI identification of multiple GPT and BERT models using three manually curated gold-standard corpora: Learning Language in Logic (LLL) with 164 PPIs in 77 sentences, Human Protein Reference Database with 163 PPIs in 145 sentences, and Interaction Extraction Performance Assessment with 335 PPIs in 486 sentences. BERT-based models achieved the best overall performance, with BioBERT achieving the highest recall (91.95%) and F1-score (86.84%) and PubMedBERT achieving the highest precision (85.25%). Interestingly, despite not being explicitly trained for biomedical texts, GPT-4 achieved commendable performance, comparable to the top-performing BERT models. It achieved a precision of 88.37%, a recall of 85.14%, and an F1-score of 86.49% on the LLL dataset. These results suggest that GPT models can effectively detect PPIs from text data, offering promising avenues for application in biomedical literature mining. Further research could explore how these models might be fine-tuned for even more specialized tasks within the biomedical domain. © 2023, CC BY.

关键词： Extraction

Systematic literature review: Accelerate the rice production for global food security

学校读者我要写书评

暂无评论

AIP Conference Proceedings 2023年第1期2594卷

作者： N. Dominic T. W. Cenggoro B. Pardamean Computer Science Department BINUS Graduate Program - Master of Computer Science Bina Nusantara University Jakarta Indonesia 11480 Computer Science Department School of Computer Science Bina Nusantara University Jakarta Indonesia 11480 Bioinformatics & Data Science Research Center Bina Nusantara University Jakarta Indonesia 11480

By 2050, about 70% of extra food production is needed to feed the world's population. As a majority staple food, the expansion of rice production should be accelerated. This systematic literature review aims to discover the factors that influence rice yields through a methodology that is partitioned into four main stages, i.e., query entry into multiple database sources, article title filtering, abstract filtering, and final article selection. The results show that genetics, irrigation system management, and farmers’ long-evolved local knowledge or experiences are the three big factors that elevate the rice production rates. Nevertheless, global warming is a serious challenge that should be deeply considered due to its great impact on reducing rice yields by up to 14.5%. Thus, the development of genetically modified rice varieties needs to be escalated while still maintaining other exogenous factors. This study contributes to providing a broader perspective for decision-makers and other relevant experts to accelerate rice production to achieve global food security.

关键词：

Flexible dual-branched message passing neural network for quantum mechanical property prediction with molecular conformation

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Jo, Jeonghee Kwak, Bumju Lee, Byunghan Yoon, Sungroh Interdisciplinary Program in Bioinformatics Seoul National University Recommendation Team Kakao Corporation Department of Electronic and IT Media Engineering Seoul National University of Science and Technology Department of Electrical and Computer Engineering Seoul National University

A molecule is a complex of heterogeneous components, and the spatial arrangements of these components determine the whole molecular properties and characteristics. With the advent of deep learning in computational chemistry, several studies have focused on how to predict molecular properties based on molecular configurations. Message passing neural network provides an effective framework for capturing molecular geometric features with the perspective of a molecule as a graph. However, most of these studies assumed that all heterogeneous molecular features, such as atomic charge, bond length, or other geometric features always contribute equivalently to the target prediction, regardless of the task type. In this study, we propose a dual-branched neural network for molecular property prediction based on message-passing framework. Our model learns heterogeneous molecular features with different scales, which are trained flexibly according to each prediction target. In addition, we introduce a discrete branch to learn single atom features without local aggregation, apart from message-passing steps. We verify that this novel structure can improve the model performance with faster convergence in most targets. The proposed model outperforms other recent models with sparser representations. Our experimental results indicate that in the chemical property prediction tasks, the diverse chemical nature of targets should be carefully considered for both model performance and generalizability. © 2021, CC BY.

关键词： Deep learning

Application for sugarcane plantation: Manpower and transportation planning

学校读者我要写书评

暂无评论

AIP Conference Proceedings 2023年第1期2594卷

作者： H. G. Mawandha D. Sudigyo T. Suparyanto B. Pardamean Department of Agrotechnology Institute of Agriculture STIPER Yogyakarta Indonesia 55281 Bioinformatics & Data Science Research Center Bina Nusantara University Jakarta Indonesia 11480 Computer Science Department BINUS Graduate Program - Master of Computer Science Program Bina Nusantara University Jakarta Indonesia 11480

Sugarcane plantations in Indonesia are still managed in a labor-intensive manner, especially plantation-owned by farmers. Labor intensive means that every stage of work in sugarcane plantation management still uses manpower. One of the jobs in sugarcane plantations that requires a lot of labor is sugar cane harvesting. Trend changing in the workforce gender would affect the quantity and quality of jobs, which is determined by the workforce's capacity and the workload of the work performed. The result of this research was a sugarcane agricultural calculator that made it simple for farmers to measure manpower and transportation requirements in estimating harvest costs. Manpower and transportation of harvested sugarcane commodities were estimated through the workload approach, namely, crop production and manpower's ability to carry a load. These two components helped sugarcane farmers to determine manpower and transportation during the harvest. The final yield produced by the projected harvest labor cost would be known immediately.

关键词：