检索结果-内蒙古大学图书馆

A novel uncertainty-aware deep learning technique with an application on skin cancer diagnosis

NEURAL COMPUTING & APPLICATIONS 2023年第30期35卷 22179-22188页

作者： Shamsi, Afshar Asgharnezhad, Hamzeh Bouchani, Ziba Jahanian, Khadijeh Saberi, Morteza Wang, Xianzhi Razzak, Imran Alizadehsani, Roohallah Mohammadi, Arash Alinejad-Rokny, Hamid UNSW Sydney Grad Sch Biomed Engn Biomed Machine Learning Lab BML Sydney Australia Univ Tehran Dept Elect & Comp Engn Tehran Iran Univ Technol Sydney Fac Engn & IT Sydney Australia Univ New South Wales UNSW Sch Comp Sci & Engn Sydney Australia Deakin Univ Intelligent Syst Res & Innovat IISRI Waurn Ponds Vic 3216 Australia Concordia Univ Inst Informat Syst Engn Montreal PQ Canada Univ New South Wales UNSW SYDNEY UNSW Data Sci Hub Sydney Australia Macquarie Univ Ctr Appl Artificial Intelligence Hlth Data Analyt Program Sydney Australia

Skin cancer, primarily resulting from the abnormal growth of skin cells, is among the most common cancer types. In recent decades, the incidence of skin cancer cases worldwide has risen significantly (one in every three newly diagnosed cancer cases is a skin cancer). Such an increase can be attributed to changes in our social and lifestyle habits coupled with devastating man-made alterations to the global ecosystem. Despite such a notable increase, diagnosis of skin cancer is still challenging, which becomes critical as its early detection is crucial for increasing the overall survival rate. This calls for advancements of innovative computer-aided systems to assist medical experts with their decision making. In this context, there has been a recent surge of interest in machine learning (ML), in particular, deep neural networks (DNNs), to provide complementary assistance to expert physicians. While DNNs have a high processing capacity far beyond that of human experts, their outputs are deterministic, i.e., providing estimates without prediction confidence. Therefore, it is of paramount importance to develop DNNs with uncertainty-awareness to provide confidence in their predictions. Monte Carlo dropout (MCD) is vastly used for uncertainty quantification;however, MCD suffers from overconfidence and being miss calibrated. In this paper, we use MCD algorithm to develop an uncertainty-aware DNN that assigns high predictive entropy to erroneous predictions and enable the model to optimize the hyper-parameters during training, which leads to more accurate uncertainty quantification. We use two synthetic (two moons and blobs) and a real dataset (skin cancer) to validate our algorithm. Our experiments on these datasets prove effectiveness of our approach in quantifying reliable uncertainty. Our method achieved 85.65 & PLUSMN;0.18 prediction accuracy, 83.03 & PLUSMN;0.25 uncertainty accuracy, and 1.93 & PLUSMN;0.3 expected calibration error outperforming vanilla MCD and MCD wi

关键词： Deep learning machine learning Classification Uncertainty quantification Skin cancer

来源：评论

学校读者我要写书评

暂无评论

A deep survival interpretable radiomics model of hepatocellular carcinoma patients

引用

PHYSICA MEDICA-EUROPEAN JOURNAL OF MEDICAL PHYSICS 2021年 82卷 295-305页

作者： Wei, Lise Owen, Dawn Rosen, Benjamin Guo, Xinzhou Cuneo, Kyle Lawrence, Theodore S. Ten Haken, Randall El Naqa, Issam Univ Michigan Dept Radiat Oncol Ann Arbor MI 48109 USA Mayo Clin Dept Radiat Oncol Rochester MN USA Harvard Med Sch Harvard Program Therapeut Sci Boston MA 02115 USA H Lee Moffitt Canc Ctr & Res Inst Dept Machine Learning Tampa FL USA

This work aims to identify a new radiomics signature using imaging phenotypes and clinical variables for risk prediction of overall survival (OS) in hepatocellular carcinoma (HCC) patients treated with stereotactic body radiation therapy (SBRT). 167 patients were retrospectively analyzed with repeated nested cross-validation to mitigate overfitting issues. 56 radiomic features were extracted from pre-treatment contrast-enhanced (CE) CT images. 37 clinical factors were obtained from patients? electronic records. Variational autoencoders (VAE) based survival models were designed for radiomics and clinical features and a convolutional neural network (CNN) survival model was used for the CECT. Finally, radiomics, clinical and raw image deep learning network (DNN) models were combined to predict the risk probability for OS. The final models yielded c-indices of 0.579 (95%CI: 0.544?0.621), 0.629 (95%CI: 0.601?0.643), 0.581 (95%CI: 0.553?0.613) and 0.650 (95%CI: 0.635?0.683) for radiomics, clinical, image input and combined models on nested cross validation scheme, respectively. Integrated gradients method was used to interpret the trained models. Our interpretability analysis of the DNN showed that the top ranked features were clinical liver function and liver exclusive of tumor radiomics features, which suggests a prominent role of side effects and toxicities in liver outside the tumor region in determining the survival rate of these patients. In summary, novel deep radiomic analysis provides improved performance for risk assessment of HCC prognosis compared with Cox survival models and may facilitate stratification of HCC patients and personalization of their treatment strategies. Liver function was found to contribute most to the OS for these HCC patients and radiomics can aid in their management.

关键词： Hepatocellular Carcinoma (HCC) Overall survival Radiomics Deep learning Variational autoencoder (VAE) Convolutional neural network (CNN) Computed tomography (CT)

来源：评论

学校读者我要写书评

暂无评论

Finite-time thermodynamics: Multistage separation processes consuming mechanical energy

引用

CHEMICAL ENGINEERING SCIENCE 2022年第PartB期248卷 117250-117250页

作者： Sukin, Ivan Tsirlin, Anatoly Andresen, Bjarne Financial Univ Govt Russian Federat Dept Data Anal & Machine Learning Moscow Russia Russian Acad Sci Complex Syst Dept Ailamazyan Program Syst Inst Yaroslavl Oblast Russia Univ Copenhagen Niels Bohr Inst Copenhagen Denmark

We analyze limiting capabilities of mechanical separation processes using finite-time thermodynamics and obtain estimates for the lower bound of the energy consumption for systems of a given productivity. We show that this consumption does not tend to zero when the molar fraction of one of the components tends to unity. The estimates obtained are used to analyze multistage separation processes, containing recycles, especially, isotope separation systems. For such systems we obtain relations between fluxes, mass transfer surfaces, and stage number. These relations are deduced from the condition of minimum dissipation, assuming that the enrichment factor is constant. We also obtain the optimality condition for the sequence of separations needed for a multicomponent mixture in a mechanical separation systems. (c) 2021 Elsevier Ltd. All rights reserved.

关键词： Irreversible separation processes Mechanical systems Reachable set Multistage systems Isotope separation Surface distribution Entropy generation Separation sequence

来源：评论

学校读者我要写书评

暂无评论

Radiomics for Predicting Oxygen Necessity in COVID-19 Patients Using Longitudinal Lung Computed Tomography

Radiomics for Predicting Oxygen Necessity in COVID-19 Patien...

引用

IEEE Symposium on Bioinformatics and Bioengineering (BIBE)

作者： Rafaela Ravazio Christian Mattjie Luis Vinicius De Moura Lucas Kupssinskü Rodrigo C. Barros Denise Cantarelli Machado Ana Maria Marques Da Silva Machine Learning Theory and Applications Lab (MALTA) Pontifical Catholic University of Rio Grande do Sul Porto Alegre Brazil Graduate Program in Biomedical Gerontology Pontifical Catholic University of Rio Grande do Sul Porto Alegre Brazil Graduate Program in Radiology University of São Paulo São Paulo Brazil

The COVID-19 pandemic has overwhelmed healthcare systems worldwide. Computed tomography imaging has emerged as an essential tool in diagnosing and monitoring cases, allowing for the detection of pulmonary changes even in the early stages of the disease. This study introduces a radiomics-based predictive model aimed at predicting the need for oxygen support in COVID-19 patients. Utilizing a private dataset collected from a local hospital with 81 patients that underwent two longitudinal chest CT scans, we employed two machine learning algorithms, Random Forest and XGBoost, to analyze radiomic features extracted from lung segmentation. We also explore incorporating clinical features in addition to the radiomic ones and using feature selection techniques to handle the high-dimensionality of the dataset. Our best model achieves an AUC of 0.81 in the test set. Our results indicate that only using radiomic features from the last time point reach a higher performance than with additional data. Therefore, it should be practical to implement a framework to predict the need of oxygen support in a clinical setting, as all the information required by the model comes from a single CT scan.

关键词：

来源：评论

学校读者我要写书评

暂无评论

A probabilistic knowledge graph for target identification

引用

PLOS COMPUTATIONAL BIOLOGY 2024年第4期20卷 e1011945页

作者： Liu, Chang Xiao, Kaimin Yu, Cuinan Lei, Yipin Lyu, Kangbo Tian, Tingzhong Zhao, Dan Zhou, Fengfeng Tang, Haidong Zeng, Jianyang Tsinghua Univ Inst Interdisciplinary Informat Sci Beijing Peoples R China Tsinghua Univ Sch Pharmaceut Sci Beijing Peoples R China Tsinghua Univ Sch Life Sci Joint Grad Program Peking Tsinghua NIBS Beijing Peoples R China Silexon AI Technol Co Ltd Machine Learning Dept Nanjing Jiangsu Peoples R China Jilin Univ Coll Comp Sci & Technol Key Lab Symbol Computat & Knowledge Engn Minist Educ Changchun Jilin Peoples R China Westlake Univ Sch Engn Hangzhou Peoples R China Westlake Lab Life Sci & Biomed Hangzhou Peoples R China Westlake Univ Res Ctr Ind Future Hangzhou Zhejiang Peoples R China Westlake Univ Sch Engn Hangzhou Zhejiang Peoples R China

Early identification of safe and efficacious disease targets is crucial to alleviating the tremendous cost of drug discovery projects. However, existing experimental methods for identifying new targets are generally labor-intensive and failure-prone. On the other hand, computational approaches, especially the machine learning-based frameworks, have shown remarkable application potential in drug discovery. In this work, we propose Progeni, a novel machine learning-based framework for target identification. In addition to fully exploiting the known heterogeneous biological networks from various sources, Progeni integrates literature evidence about the relations between biological entities to construct a probabilistic knowledge graph. Graph neural networks are then employed in Progeni to learn the feature embeddings of biological entities to facilitate the identification of biologically relevant target candidates. A comprehensive evaluation of Progeni demonstrated its superior predictive power over the baseline methods on the target identification task. In addition, our extensive tests showed that Progeni exhibited high robustness to the negative effect of exposure bias, a common phenomenon in recommendation systems, and effectively identified new targets that can be strongly supported by the literature. Moreover, our wet lab experiments successfully validated the biological significance of the top target candidates predicted by Progeni for melanoma and colorectal cancer. All these results suggested that Progeni can identify biologically effective targets and thus provide a powerful and useful tool for advancing the drug discovery process. The selection of a disease target, a biological entity with which potential drugs interact to treat diseases, is often the first step of a drug discovery project. Known to be extremely costly, drug discovery projects spend tremendous resources on failed drug instances, most of which result from inadequate choices of targets. To help

关键词： Colorectal cancer Melanoma Protein interaction networks Drug discovery machine learning Cancer treatment Drug interactions Neural networks

来源：评论

学校读者我要写书评

暂无评论

Churn Prediction in Financial Risk Management using Deep learning Techniques 2

Churn Prediction in Financial Risk Management using Deep Lea...

引用

2nd IEEE International Conference on Advanced Technologies in Intelligent Control, Environment, Computing and Communication Engineering, ICATIECE 2022

作者： Kumar, V Sunil Sowmyashree, S. Pareek, Piyush Kumar Khanna, Ashish Costa De Albuquerque, Victor Hugo Gupta, Deepak Nitte Meenakshi Institute of Technology Department of Artificial Intelligence & Machine Learning Bengaluru India Computer Science & Engineering East West Institute of Technology Bengaluru India Graduate Program in Telecommunication Engineering Federal Institute of Education Science and Technology of Cearά Fortaleza Brazil Maharaja Agrasen Institute of Technology GGSIPU CSE Dept India Federal University of Cearά Department of Teleinformatics Engineering Brazil

ISBN: (纸本)9781665493963

It is essential to develop an efficient Client Relationship Model in order to make use of the estimations about customer turnover (CRM). Finding and breaking down the facts surrounding the business information understanding is necessary for client produce prediction, which should be made possible effectively by altering the strategies used for business insight. The tools that are provided by company insight allow for the historical, current, and prospective perspectives of business activity to be anticipated and dissected. It is possible to deal with it in an efficient way by providing the information mining approaches that choose the data from the created informative index that is the most helpful. Fundamental leadership should be possible to achieve in a fruitful way if one has this information. The actions that consumers have taken in the recent past are going to be analyzed, and an accurate forecast is going to be formed about those customers who are likely going to get unhappy in the near future based on the results of that prediction. The problem is that deciding on a certain period to investigate the client's activities and also establishing on a specific day and age for each of the clients won't be appropriate6. This is due to the fact that the problem is caused by the fact that deciding on a certain period to investigate the client's activities will take too long. The challenge lies just in this aspect. How to decide on a certain amount of time to spend researching the activities of the customer. For instance, when planning a model, 100 customers are used, 70 of those customers are considered 'active,' which indicates that they are still conducting business with the organisation, and the remaining 30 customers are considered 'agitate,' which indicates that they have severed their ties with the organisation. The primary purpose of the work that has been suggested is to carry out exploratory research and management as a multicriteria basic leadership issue, a

关键词： Forecasting

来源：评论

学校读者我要写书评

暂无评论

Making transport more robust and interpretable by moving data through a small number of anchor points 38

Making transport more robust and interpretable by moving dat...

引用

International Conference on machine learning (ICML)

作者： Lin, Chi-Heng Azabou, Mehdi Dyer, Eva L. Georgia Tech Dept Elect & Comp Engn Atlanta GA 30332 USA Georgia Tech Machine Learning Program Atlanta GA 30332 USA Georgia Tech Coulter Dept Biomed Engn Atlanta GA 30332 USA Emory Univ Atlanta GA 30322 USA

ISBN: (纸本)9781713845065

Optimal transport (OT) is a widely used technique for distribution alignment, with applications throughout the machine learning, graphics, and vision communities. Without any additional structural assumptions on transport, however, OT can be fragile to outliers or noise, especially in high dimensions. Here, we introduce Latent Optimal Transport (LOT), a new approach for OT that simultaneously learns low-dimensional structure in data while leveraging this structure to solve the alignment task. The idea behind our approach is to learn two sets of "anchors" that constrain the flow of transport between a source and target distribution. In both theoretical and empirical studies, we show that LOT regularizes the rank of transport and makes it more robust to outliers and the sampling density. We show that by allowing the source and target to have different anchors, and using LOT to align the latent spaces between anchors, the resulting transport plan has better structural interpretability and highlights connections between both the individual data points and the local geometry of the datasets.

关键词： Anchors

来源：评论

学校读者我要写书评

暂无评论

Coarse-graining conformational dynamics with multi-dimensional generalized Langevin equation: how, when, and why

arXiv

引用

arXiv 2024年

作者： Xie, Pinchen Qiu, Yunrui Weinan, E. Program in Applied and Computational Mathematics Princeton University PrincetonNJ08544 United States Department of Chemistry University of Wisconsin-Madison MadisonWI53706 United States AI for Science Institute Beijing100080 China Center for Machine Learning Research School of Mathematical Sciences Peking University Beijing100084 China

A data-driven ab initio generalized Langevin equation (AIGLE) approach is developed to learn and simulate high-dimensional, heterogeneous, coarse-grained conformational dynamics. Constrained by the fluctuation-dissipation theorem, the approach can build coarse-grained models in dynamical consistency with all-atom molecular dynamics. We also propose practical criteria for AIGLE to enforce long-term dynamical consistency. Case studies of a toy polymer, with 20 coarse-grained sites, and the alanine dipeptide, with two dihedral angles, elucidate why one should adopt AIGLE or its Markovian limit for modeling coarse-grained conformational dynamics in practice. © 2024, CC BY.

关键词： Coarse-grained modeling

来源：评论

学校读者我要写书评

暂无评论

Predicting High School Students' Academic Performance: A Comparative Study of Supervised machine learning Techniques 1

Predicting High School Students' Academic Performance: A Com...

引用

machine learning-Driven Digital Technologies for Educational Innovation Workshop

作者： Sanchez-Pozo, Nadia N. Mejia-Ordonez, Juan S. Chamorro, Diana C. Mayorca-Torres, Dagoberto Peluffo-Ordonez, Diego H. SDAS Res Grp Machine Learning Res Program Ben Guerir Morocco Univ Tecn Machala Fac Ciencias Quim & Salud Carrera Ingn Quim Machala Ecuador Univ Mariana Programa Mecatron Fac Ingn Pasto Colombia Mohamed VI Polytech Univ Modeling Simulat & Data Anal MSDA Res Program Ben Guerir Morocco

ISBN: (纸本)9781665427630

The proliferation of mobile devices and the rapid development of information and communication technologies have revolutionized education. Educational data has evolved to be voluminously massive, broadly various, and produced at high velocity. Therefore, computerized techniques for integrating, processing, and transforming data into valuable knowledge have become necessary to improve internal academic processes. Specifically, educational data mining is an emerging discipline concerned with analyzing the massive amounts of academic data generated and stored by educational institutions. In this sense, machine learning algorithms aid decision-makers who are establishing strategies to improve students' learning experience and institutional effectiveness by revealing hidden patterns in academic performance. Thus, this paper describes our comparative study of machine learning techniques to predict academic performance. We selected the features that best fit the discovery of patterns in the academic performance of high school students, resulting in a balance between accuracy and interpretability. We implemented six supervised learning algorithms for pattern recognition: Light Gradient Boosting machine, Gradient Boosting, AdaBoost, Logistic Regression, Random Forest, and K-nearest Neighbors. The experimental results showed that the Gradient Boosting (Gbc) algorithm achieved the highest accuracy (96.77%), superior to other classification techniques considered.

关键词： Educational data mining Academic performance Classification High school education Supervised learning Educational innovation

来源：评论

学校读者我要写书评

暂无评论

Multimodal Sentiment Analysis using Kernel Based Support Vector machine 2

Multimodal Sentiment Analysis using Kernel Based Support Vec...

引用

2nd IEEE International Conference on Advanced Technologies in Intelligent Control, Environment, Computing and Communication Engineering, ICATIECE 2022

作者： Kumar, V Sunil Pareek, Piyush Kumar De Albuquerque, Victor Hugo Costa Khanna, Ashish Gupta, Deepak Madhumala, R.B. Nitte Meenakshi Institute of Technology Department Artificial Intelligence and Machine Learning Bengaluru India Graduate Program in Telecommunication Engineering Federal Institute of Education Science and Technology of Ceará Fortaleza Brazil Federal University of Ceará Department of Teleinformatics Engineering Brazil Maharaja Agrasen Institute of Technology GGSIPU CSE Dept India Dayananda Sagar Academy of Technology and Management Department of ISE Bangalore India

ISBN: (纸本)9781665493963

There has been a steady expansion in the number of social websites, electronic media outlets, and blogging services, all of which contribute a large number of user messages such as comments, reviews, and opinions from customers. The quantity of these signals, which are used to evaluate the quality of a wide variety of goods and services, is rather substantial. As a result, it is challenging to monitor and identify the feelings of customers. As a result, a method known as Sentiment Analysis (SA), also known as Opinion Mining (OM), was developed. This method gathers data from the internet by using Natural Language Processing (NLP) strategies and then analyses people's thoughts. It does this by extracting subjective information from the words, assessing opinions, and quantifying positive and negative evaluations. The end result is a conclusion about the specific entity, which may be an organisation, the public, or a product. The NLP first gathers the views, and then those opinions are sorted into positive, neutral, and negative categories. Both the marketing promotion and the customer's ability to choose the most appropriate product for their needs may be improved by doing an analysis. The purpose of this study is to evaluate the use of different machine learning and swarm intelligence optimization methods to perform sentiment analysis on data derived from movie reviews. This study helps many businesses increase their profits while simultaneously lowering the number of problems caused by a certain product. Within the scope of this investigation, three distinct models of sentiment analysis were investigated. An effective method for doing sentiment analysis known as a Hybrid Kernel based Support Vector machine (HKSVM) has been presented. The preprocessing that is described here is meant to cut down on the data inconsistencies and get rid of the data that isn't needed. After that, the Expectation Maximization (EM) technique is used in order to identify an outlier. After t

关键词： Sentiment analysis

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：