检索结果-内蒙古大学图书馆

machine learning-based extrachromosomal DNA identification in large-scale cohorts reveals its clinical implications in cancer

引用

NATURE COMMUNICATIONS 2024年第1期15卷 1515页

作者： Wang, Shixiang Wu, Chen-Yi He, Ming-Ming Yong, Jia-Xin Chen, Yan-Xing Qian, Li-Mei Zhang, Jin-Ling Zeng, Zhao-Lei Xu, Rui-Hua Wang, Feng Zhao, Qi Sun Yat Sen Univ Guangdong Prov Clin Res Ctr Canc State Key Lab Oncol South ChinaCanc Ctr Guangdong Key Lab Nasopharyngeal Carcinoma Diag & Guangzhou 510060 Peoples R China Chinese Acad Med Sci Res Unit Precis Diag & Treatment Gastrointestinal Guangzhou 510060 Peoples R China

The clinical implications of extrachromosomal DNA (ecDNA) in cancer therapy remain largely elusive. Here, we present a comprehensive analysis of ecDNA amplification spectra and their associationwith clinical andmolecular features in multiple cohorts comprising over 13,000 pan-cancer patients. Using our developed computational framework, GCAP, and validating it with multi-faceted approaches, we reveal a consistent pan-cancer pattern of mutual exclusivity between ecDNAamplification andmicrosatellite instability (MSI). In addition, we establish the role of ecDNA amplification as a risk factor and refine genomic subtypes in a cohort from 1015 colorectal cancer patients. Importantly, our investigation incorporates data from four clinical trials focused on anti-PD-1 immunotherapy, demonstrating the pivotal role of ecDNA amplification as a biomarker for guiding checkpoint blockade immunotherapy in gastrointestinal cancer. This finding represents clinical evidence linking ecDNA amplification to the effectiveness of immunotherapeutic interventions. Overall, our study provides a proof-of-concept of identifying ecDNA amplification from cancer whole-exome sequencing (WES) data, highlighting the potential of ecDNA amplification as a valuable biomarker for facilitating personalized cancer treatment.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Comparative Study on the Methods of Atmospheric Early Warning Based on machine learning 1

引用

7th EAI international conference on machine learning and Intelligent Communication, MLICOM 2022

作者： Yu, Guanghua Ou, Jianghong Fan, Dahua Heihe University Heilongjiang Heihe China School of Computer and Information Engineering College HeiHe University Heihe China Starway Communication No. 31 Kefeng Road Guangzhou Guangzhou Science City510663 China

ISBN: (数字)9783031302374

ISBN: (纸本)9783031302367

This paper takes Beijing Meteorological environment data as the analysis object, and uses six eigenvalues to analyze PM2.5. Normal equation, gradient descent, ridge regression and xgboost are used to predict and analyze and compare the results. The mean square error of xgboost algorithm is 0.322. It is significantly lower than the other three algorithms, and its effect is far better than the other three algorithms, which is suitable for the prediction of this data. © 2023, ICST Institute for Computer sciences, Social Informatics and Telecommunications Engineering.

关键词： machine learning

来源：评论

学校读者我要写书评

暂无评论

Improving Robustness to Multiple Spurious Correlations by Multi-Objective Optimization 41

Improving Robustness to Multiple Spurious Correlations by Mu...

引用

41st international conference on machine learning, ICML 2024

作者： Kim, Nayeong Kang, Juwon Ahn, Sungsoo Ok, Jungseul Kwak, Suha Department of Computer Science and Engineering POSTECH Pohang Korea Republic of Graduate School of Artificial Intelligence POSTECH Pohang Korea Republic of

We study the problem of training an unbiased and accurate model given a dataset with multiple biases. This problem is challenging since the multiple biases cause multiple undesirable shortcuts during training, and even worse, mitigating one may exacerbate the other. We propose a novel training method to tackle this challenge. Our method first groups training data so that different groups induce different shortcuts, and then optimizes a linear combination of group-wise losses while adjusting their weights dynamically to alleviate conflicts between the groups in performance;this approach, rooted in the multi-objective optimization theory, encourages to achieve the minimax Pareto solution. We also present a new benchmark with multiple biases, dubbed MultiCelebA, for evaluating debiased training methods under realistic and challenging scenarios. Our method achieved the best on three datasets with multiple biases, and also showed superior performance on conventional single-bias datasets. Copyright 2024 by the author(s)

关键词： Multiobjective optimization

来源：评论

学校读者我要写书评

暂无评论

Deep learning for Assistive Decision-Making in Robot-Aided Rehabilitation Therapy 10th

Deep Learning for Assistive Decision-Making in Robot-Aided R...

引用

10th international Work-conference on the Interplay Between Natural and Artificial Computation (IWINAC)

作者： Martinez-Pascual, David Catalan, Jose M. Liedo, Luis D. Blanco-Ivorra, Andrea Garcia-Aracil, Nicolas Miguel Hernandez Univ Bioengn Inst Robot & Artificial Intelligence Grp Ave Univ S-N Alicante 03202 Spain

ISBN: (纸本)9783031611360;9783031611377

The use of robotic rehabilitation devices has emerged as a promising approach to enhance motor recovery during rehabilitation. One of the significant challenges while using these devices is the ability to decide when to provide assistance to the patient. In this regard, we propose a Deep learning-based solution that can learn from a therapist's criteria when a patient requires assistance during robot-aided rehabilitation therapy. We have trained and evaluated the proposed model with data from different patients during a point-to-point game modality. To make the model more universal, we have applied a series of transformations to the trajectory data before using them as inputs. The proposed model has been evaluated using different metrics and has shown an accuracy of 93.21% and an F1-Score of 85.05% with the validation dataset. Furthermore, the model has achieved an accuracy of 69.32% and an F1-Score of 63.31% with users who were not involved in the model learning process.

关键词： Rehabilitation Robotics Neurorehabilitation Deep learning Exergames

来源：评论

学校读者我要写书评

暂无评论

RETRIEVAL-BASED DISENTANGLED REPRESENTATION learning WITH NATURAL LANGUAGE SUPERVISION 12

RETRIEVAL-BASED DISENTANGLED REPRESENTATION LEARNING WITH NA...

引用

12th international conference on learning Representations, ICLR 2024

作者： Zhou, Jiawei Li, Xiaoguang Shang, Lifeng Jiang, Xin Liu, Qun Chen, Lei The Hong Kong University of Science and Technology Hong Kong Huawei Noah's Ark Lab Canada

Disentangled representation learning remains challenging as the underlying factors of variation in the data do not naturally exist. The inherent complexity of real-world data makes it unfeasible to exhaustively enumerate and encapsulate all its variations within a finite set of factors. In light of this, we present Vocabulary Disentangled Retrieval (VDR), a retrieval-based framework that harnesses natural language as proxies of the underlying data variation to drive disentangled representation learning. Our approach employs a bi-encoder model to represent both data and natural language in a vocabulary space, enabling the model to distinguish dimensions that capture intrinsic characteristics within data through its natural language counterpart, thus facilitating disentanglement. We extensively assess the performance of VDR across 15 retrieval benchmark datasets, covering text-to-text and cross-modal retrieval scenarios, as well as human evaluation. Our experimental results compellingly demonstrate the superiority of VDR over previous bi-encoder retrievers with comparable model size and training costs, achieving an impressive 8.7% improvement in NDCG@10 on the BEIR benchmark, a 5.3% increase on MS COCO, and a 6.0% increase on Flickr30k in terms of mean recall in the zero-shot setting. Moreover, the results from human evaluation indicate that interpretability of our method is on par with SOTA captioning models. © 2024 12th international conference on learning Representations, ICLR 2024. All rights reserved.

关键词： Digital storage

来源：评论

学校读者我要写书评

暂无评论

Financial Evaluation of Urban Cooperative Banks - A machine learning Approach 2

Financial Evaluation of Urban Cooperative Banks - A Machine ...

引用

2nd international conference on machine learning and data Engineering, ICMLDE 2023

作者： Gautam, Triambica Srivastava, Amit Jain, Shruti Department Fo HSS Jaypee University of Information Technology Solan India Department of ECE Jaypee University of Information Technology Solan India

The banking industry is an essential component of the world's economy. In a globalized world, the strength of a banking system is determined by each of its components, whatever its size. This calls into question the financial viability of smaller institutions like cooperative banks. Despite their tiny size, these banks play a crucial role in extending and bolstering the reach of traditional banking channels. Scheduled Urban Cooperative Banks have been examined in this research to better understand the elements that affect these organizations' success or failure. A neural network model with five input and four output variables was developed over four years, from 2019 to 2022. A neural network model was developed to aid in assessing the various performance metrics and is used to examine a prediction model to help with output value prediction. R-values were monitored to achieve the lowest possible mean square error for training, testing, and validation. 1000 epochs of experimentation were completed. The mu value for the six validation tests was found to be 1e10, and the gradient value was seen to be 1e-5. The output or input for future years, if any are available, can be predicted by authors using a neural network prediction model. © 2024 Elsevier B.V.. All rights reserved.

关键词： Mean square error

来源：评论

学校读者我要写书评

暂无评论

Epicentral Region Estimation Using Convolutional Neural Networks 7th

Epicentral Region Estimation Using Convolutional Neural Netw...

引用

7th international conference on machine learning, Optimization, and data science (LOD) / 1st Symposium on Artificial Intelligence and Neuroscience (ACAIN)

作者： Cruz, Leonel Tous, Ruben Otero, Beatriz Alvarado, Leonardo Mus, Sergi Rojas, Otilio Univ Politecn Cataluna ES-08034 Barcelona Spain Univ Cent Venezuela Caracas 1070 Venezuela Fdn Venezolana Invest Sismol Caracas 1070 Venezuela Barcelona Supercomp Ctr Barcelona Spain

ISBN: (纸本)9783030954673;9783030954666

Recent works have assessed the capability of deep neural networks of estimating the epicentral source region of a seismic event from a single-station three-channel signal. In all the cases, the geographical partitioning is performed by automatic tessellation algorithms such as the Voronoi decomposition. This paper evaluates the hypothesis that the source region estimation accuracy is significantly increased if the geographical partitioning is performed considering the regional geological characteristics such as the tectonic plate boundaries. Also, it raises the transformation of the training data to increase the accuracy of the predictive model based on a Projected Coordinate Reference (PCR) System. A deep convolutional neural network (CNN) is applied over the data recorded by the broadband stations of the Venezuelan Foundation of Seismological Research (FUNVISIS) in the region of 9.5 to 11.5N and 67.0 to 69.0W between April 2018 and April 2019. In order to estimate the epicentral source region of a detected event, several geographical tessellations provided by seismologists from the area are employed. These tessellations, with different number of partitions, consider the fault systems of the study region (San Sebastian, La Victoria and Moron fault systems). The results are compared to the ones obtained with automatic partitioning performed by the k-means algorithm.

关键词： Earthquake location estimation Supervised learning Convolutional neural networks Deep learning

来源：评论

学校读者我要写书评

暂无评论

Multi-lingual Scene Text Detection Containing the Arabic Scripts Using an Optimal then Enhanced YOLO Model

Multi-lingual Scene Text Detection Containing the Arabic Scr...

引用

Workshop held at the 12th international conference on Model and data Engineering, MEDI 2023

作者： Turki, Houssem Elleuch, Mohamed Kherallah, Monji University of Sfax Sfax Tunisia University of Kairoun Kairoun Tunisia Faculty of Sciences University of Sfax Sfax Tunisia University of Sfax Sfax Tunisia

ISBN: (纸本)9783031557286

In the past few years, notable progress has been achieved in the field of deep learning, particularly in the realm of identifying text within images of natural scenes, owing to the advancements in machine learning and artificial intelligence. The effectiveness of deep learning and text detection in the wild, especially when dealing with Arabic language, is frequently hindered by the scarcity of diverse datasets encompassing multiple languages and scripts, which poses an additional challenge. Despite significant advancements, this shortage continues to be a limiting factor. The YOLO (You Only Look Once) deep learning neural network has gained widespread popularity for its adaptability in tackling various machine learning tasks, notably in the field of computer vision. The YOLO algorithm has garnered growing recognition for its remarkable capability to address intricate issues when dealing with images taken in natural environments, managing noisy data, and surmounting the diverse challenges encountered in the wild. Our experiments provide a concise evaluation of text detection algorithms centered around convolutional neural networks (CNNs). Specifically, we concentrate on different versions of the YOLO models, applying identical data augmentation methods to both the SYPHAX dataset and the ICDAR MLT-2019 dataset, both of which encompass Arabic scripts within images of natural scenes. The objective of this article is to pinpoint the most efficient YOLO algorithm for recognizing Arabic script in the wild, and subsequently, to improve upon the best-performing model. Additionally, we aim to investigate potential research directions that can further enhance the capabilities of the most robust architecture in this domain. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2024.

关键词： Arabic script Computer Vision Deep learning enhanced architecture Multi-lingual script Natural Scene images Optimal model Text detection in the wild YOLO

来源：评论

学校读者我要写书评

暂无评论

LIDL: Local Intrinsic Dimension Estimation Using Approximate Likelihood 39

LIDL: Local Intrinsic Dimension Estimation Using Approximate...

引用

39th international conference on machine learning (ICML)

作者： Tempczyk, Piotr Michaluk, Rafal Garncarek, Lukasz Spurek, Przemyslaw Tabor, Jacek Golinski, Adam Univ Warsaw Inst Informat Warsaw Poland Polish Natl Inst Machine Learning Warsaw Poland Deeptale AI Wroclaw Poland Applica Warsaw Poland Jagiellonian Univ GMUM Warsaw Poland Univ Oxford Oxford England

Most of the existing methods for estimating the local intrinsic dimension of a data distribution do not scale well to high dimensional data. Many of them rely on a non-parametric nearest neighbours approach which suffers from the curse of dimensionality. We attempt to address that challenge by proposing a novel approach to the problem: Local Intrinsic Dimension estimation using approximate Likelihood (LIDL). Our method relies on an arbitrary density estimation method as its subroutine, and hence tries to sidestep the dimensionality challenge by making use of the recent progress in parametric neural methods for likelihood estimation. We carefully investigate the empirical properties of the proposed method, compare them with our theoretical predictions, show that LIDL yields competitive results on the standard benchmarks for this problem, and that it scales to thousands of dimensions. What is more, we anticipate this approach to improve further with the continuing advances in the density estimation literature.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Diabetes Prediction Using machine learning Classifiers 5th

Diabetes Prediction Using Machine Learning Classifiers

引用

5th international conference on machine learning, Image Processing, Network Security, and data sciences, MIND 2023

作者： Choudhary, Dharna Gupta, Pradeep Gupta, Sonam Ajay Kumar Garg Engineering College Ghaziabad Ghaziabad India

ISBN: (纸本)9783031622168

Diabetes Mellitus is an endocrine disorder that arises when your blood glucose or blood sugar is too high. Your body either does not create enough insulin or does not use the insulin that it does produces adequately. Insulin is a hormonal substance that regulates glucose levels in blood. Hyperglycaemia refers to high blood sugar levels, commonly known as blood glucose. It can lead to serious health consequences in diabetics over time. This condition can lead to substantial and progressive harm across various bodily systems, with a notable impact on neurons and blood vessels. Diabetes affects 537 million people worldwide, making it the deadliest and most frequent non-communicable disease. Diabetes takes the lives of approximately 2–5 million people each year. In the year 2014, diabetes affected 8.5% of individuals aged 18 and older. In 2019, diabetes directly led to 1.5 million deaths with almost half of these occurring before the age of 65. Additionally, diabetes led to 560 000 deaths linked to chronic kidney disease and elevated blood glucose was a factor in around 20% of cardiovascular-related deaths. Diabetes increased age-specific death rates by 5% between 1999 and 2020. Diabetes related fatalities rose by 14% in low-and middle-income nations. The death rate due to diabetes is said to be increased to 629 million by 2045. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2024.

关键词： machine learning

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：