检索结果-内蒙古大学图书馆

27th International Conference on Pattern Recognition, ICPR 2024

作者： Paolo, Domenico Russo, Ciro Russo, Giulio Greco, Carlo Cortellini, Alessio Russano, Marco Ramella, Sara Soda, Paolo Marrocco, Claudio Bria, Alessandro Sicilia, Rosa Unit of Computer Systems and Bioinformatics Department of Engineering Universita‘ Campus Bio-Medico di Roma Selcetta Italy Department of Electrical and Information Engineering University of Cassino and Lazio Meridionale FR Cassino03043 Italy Research Unit of Radiation Oncology Department of Medicine and Surgery University Campus Bio-Medico di Roma Selcetta Italy Operative Research Unit of Medical Oncology Fondazione Policlinico Universitario Campus Bio-Medico Portillo Italy Department of Diagnostics and Intervention Umeå University Umeå Sweden

ISBN: (纸本)9783031876592

Predicting pathologic complete response in non-small cell lung cancer is crucial for tailoring effective treatment strategies and to improve patient outcomes. With the increasing application of artificial intelligence in cancer research, machine learning is poised to play a significant role in prognostication and decision-making. This paper presents a novel approach that utilizes named entity recognition and attention mechanisms applied to electronic health records to predict the pathologic complete response. We first employ named entity recognition to extract relevant biomedical entities from unstructured clinical notes within reports. These entities, combined with structured data, are then processed using a hierarchical attention mechanism to generate comprehensive patient representations. This approach captures complex relationships and contextual information within electronic health records compared to traditional methods. The results highlight the potential of advanced natural language processing techniques to enhance clinical decision-making and support personalized treatment planning in oncology. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： Electronic health record

来源：评论

学校读者我要写书评

暂无评论

Not Another Imputation Method: A Transformer-based Model for Missing Values in Tabular Datasets

arXiv

引用

arXiv 2024年

作者： Caruso, Camillo Maria Soda, Paolo Guarrasi, Valerio The Research Unit of Computer Systems and Bioinformatics Department of Engineering Università Campus Bio-Medico di Roma Roma Italy The Research Unit of Computer Systems and Bioinformatics Department of Engineering Università Campus Bio-Medico di Roma Roma Italy The Department of Diagnostics and Intervention Radiation Physics Biomedical Engineering Umeå University Umeå Sweden

Handling missing values in tabular datasets presents a significant challenge in training and testing artificial intelligence models, an issue usually addressed using imputation techniques. Here we introduce "Not Another Imputation Method" (NAIM), a novel transformer-based model specifically designed to address this issue without the need for traditional imputation techniques. NAIM employs feature-specific embeddings and a masked self-attention mechanism that effectively learns from available data, thus avoiding the necessity to impute missing values. Additionally, a novel regularization technique is introduced to enhance the model's generalization capability from incomplete data. We extensively evaluated NAIM on 5 publicly available tabular datasets, demonstrating its superior performance over 6 state-of-the-art machine learning models and 4 deep learning models, each paired with 3 different imputation techniques when necessary. The results highlight the efficacy of NAIM in improving predictive performance and resilience in the presence of missing data. To facilitate further research and practical application in handling missing data without traditional imputation methods, we made the code for NAIM available at https://***/cosbidev/NAIM. © 2024, CC BY-NC-ND.

关键词： Embeddings

来源：评论

学校读者我要写书评

暂无评论

Building an AI-Enabled Metaverse for Intelligent Healthcare: Opportunities and Challenges

Building an AI-Enabled Metaverse for Intelligent Healthcare:...

引用

2023 Italia Intelligenza Artificiale - Thematic Workshops, Ital-IA 2023

作者： Guarrasi, Valerio Tronchin, Lorenzo Caruso, Camillo Maria Rofena, Aurora Manni, Guido Aksu, Fatih Paolo, Domenico Iannello, Giulio Sicilia, Rosa Cordelli, Ermanno Soda, Paolo Unit of Computer Systems and Bioinformatics Department of Engineering University Campus Bio-Medico of Rome Italy Department of Biomedical Sciences Humanitas University Milan Italy Department of Radiation Sciences Radiation Physics Biomedical Engineering Umeå University Sweden

This abstract discusses the development of a metaverse for intelligent healthcare, which involves creating a virtual environment where healthcare professionals, patients, and researchers can interact and collaborate using digital technologies. The metaverse can improve the efficiency and effectiveness of healthcare services and provide new opportunities for research and innovation. AI models are necessary for analyzing patient data and providing personalized healthcare recommendations, but the data in a metaverse setting is inherently multimodal, unstructured, noisy, incomplete, limited, or partially inconsistent, which poses a challenge for AI models. However, it becomes necessary the integration of AI models for the development of virtual scanners to simulate image modalities, and robotics to simulate surgical procedures within a virtual environment. The ultimate goal is to leverage the power of AI to enhance the quality of healthcare in a metaverse for intelligent healthcare, which has the potential to transform the way healthcare services are delivered and improve health outcomes for patients worldwide. © 2022 Copyright for this paper by its authors. Use permitted under Creative Commons License Attribution 4.0 International (CC BY 4.0).

关键词： Virtual reality

来源：评论

学校读者我要写书评

暂无评论

Identification of the Optimal Meal Detection Strategy for Adults, Adolescents, and Children with Type 1 Diabetes: an in Silico Validation

Identification of the Optimal Meal Detection Strategy for Ad...

引用

IEEE Annual International computer Software and Applications Conference (COMPSAC)

作者： Federico D’Antoni Martina Bertazzoni Luca Vollero Mario Merone Department of Engineering Research Unit of Computer Systems and Bioinformatics Università Campus Bio-Medico di Roma Rome Italy

Current management of Type 1 Diabetes mellitus (T1D) resorts to manual meal announcements from the patient to manage postprandial glycemia; nevertheless, suboptimal glycemic control is observed in real data, with the presence of many hypoglycemic and hyperglycemic events. The utilization of Continuous Glucose Monitoring (CGM) sensors and Artificial Intelligence (AI) is paving the way for improved and automated glycemic control. A step in this direction is represented by the automation of meal detection, which would not require patients to perform tasks such as carbohydrate estimation and meal announcement that are error-prone, especially for children and elderly *** this work, we investigate several AI models for meal detection from in silico data of 10 adults, 10 adolescents, and 10 children with T1D using only CGM data, and compare them to the standard detection method based on the glycemic threshold. We generate 30 days of data per patient that include 5 meals per day and introduce human error on carbohydrate estimation to make data more similar to the real ones. The AI models can detect more than 81% of meals from any cohort of patients while producing a relatively small amount of false positives. The feedforward neural network, the support vector machine, and the threshold method are the most promising meal detection strategies for adult, adolescent, and child populations, respectively, and may improve patients’ health and disease management.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Boosting white shark optimizer for global optimization and cloud scheduling problem

引用

Neural Computing and Applications 2024年第18期36卷 10853-10879页

作者： Mostafa, Reham R. Chhabra, Amit Khedr, Ahmed M. Hashim, Fatma A. University of Sharjah Sharjah27272 United Arab Emirates Information Systems Department Faculty of Computers and Information Sciences Mansoura University Mansoura35516 Egypt Department of Computer Engineering and Technology Guru Nanak Dev University Amritsar143005 India Computer Science Department University of Sharjah Sharjah27272 United Arab Emirates Faculty of Engineering Helwan University Helwan Egypt MEU Research Unit Middle East University Amman11831 Jordan

With the growing adoption of cloud computing in both public and private sector enterprises, the industry has experienced rapid expansion. To fully unlock the potential of cloud computing, efficient task scheduling becomes crucial. In cloud computing, task scheduling involves optimizing the allocation of tasks to a diverse range of resources, such as virtual machines, with the goals of reducing makespan, maximizing resource utilization, and minimizing response times. This challenge becomes even more pronounced for large-scale tasks due to the NP-hard nature of the problem. Consequently, the integration of metaheuristic algorithms into task scheduling has emerged as a solution to equitably distribute complex and diverse tasks across limited resources within acceptable timeframes. To enhance the quality of cloud computing services, this research introduces the modified white shark optimizer (mWSO) as an alternative task scheduling technique. The improved variant mWSO boosts the performance of the original WSO by introducing the following three enhancement steps: (1) introduce memory-based WSO to boost the exploitation phase, (2) propose an exploration-exploitation balance phase to enhance the exploration phase, and (3) introduce a control randomization parameter to balance exploration and exploitation properly. The mWSO is subjected to testing on both the global optimization problems from CEC2020 and cloud task scheduling problems. The experimental results of mWSO demonstrate high performance for CEC2020 competition benchmarks compared to other state-of-the-art and recent metaheuristic algorithms. In the case of the task scheduling problem, the mWSO achieved − 0.01 to 13.53% and 0.62–10.42% makespan and energy consumption reduction, respectively, for CEA-Curie workloads. For HPC2N workloads, mWSO achieved 7.27–29.53% makespan reduction and 3.52–26.08% energy savings over the compared metaheuristics. The statistical validity of the performance is also verified using Wil

关键词： Global optimization

来源：评论

学校读者我要写书评

暂无评论

Identification of Optimal Training for Prediction of Glucose Levels in Type-1-Diabetes Using Edge Computing

Identification of Optimal Training for Prediction of Glucose...

引用

2022 International Conference on Electrical, computer, Communications and Mechatronics engineering, ICECCME 2022

作者： D'Antoni, Federico Petrosino, Lorenzo Velieri, Andrea Sasso, Daniele D'Angelis, Onorato Boscarino, Tamara Vollero, Luca Merone, Mario Piemonte, Vincenzo Università Campus Bio-Medico di Roma Research Unit of Computer Systems and Bioinformatics Department of Engineering Rome Italy Università Campus Bio-Medico di Roma Chemical-physics Fundamentals in Chemical Engineering Department of Engineering Rome Italy

ISBN: (数字)9781665470957

ISBN: (纸本)9781665470957

Type 1 Diabetes Mellitus (T1DM) is a chronic autoimmune disease involving high blood glucose levels. It is caused by the destruction of pancreatic β-cells, which are responsible for insulin production. Survival of subjects with T1DM is dependent on exogenous insulin administrations [1]. In order to calculate and then inject an appropriate insulin dose, subjects must frequently monitor their glucose values, taking into account not only the exact amount of carbohydrates ingested, but also physical activity [2], [3], disease states [4], and stress episodes [5]. The main goal of glycemic control is to avoid or at least mitigate events such as hypoglycemia and hyperglycemia [6]. Hypoglycemia, defined as a blood glucose level lower than 70 mg/dL, can result in serious immediate complications such as seizure or coma [7], whereas hyperglycemia, defined as a blood glucose level greater than 180 mg/dL, leads to several microvascular and macrovascular complications in the long term [7]. The advent of Continuous Glucose Monitoring (CGM) devices, i.e., sensors capable of measuring interstitial glucose concentration at 1-to 5-minute intervals, has improved the quality of life of individuals with T1DM, allowing a decrease in the occurrence of hypo/hyperglycemic events [8]. Despite the widespread use of these devices, adverse glycemic events still occur. Providing patients with a prediction of blood glucose levels can further improve their quality of life enabling proactive interventions and reducing their daily cognitive load [9]. Indeed, research is focusing on the development of data-driven models that can predict future glycemic levels as accurately as possible on both short-term [10] and long-term predictions [11], leveraging past CGM values alone [12] or in combination with other features such as insulin injections or carbohydrate intake [13]. © 2022 IEEE.

关键词： Insulin

来源：评论

学校读者我要写书评

暂无评论

Automated analysis of fundus images for the diagnosis of retinal diseases: a review

引用

research on Biomedical engineering 2024年第1期40卷 225-251页

作者： Berto, Alexa Scarpa, Fabio Tsiknakis, Nikos Manikis, Georgios Fotiadis, Dimitrios I. Marias, Kostas Scarpa, Alberto D-Eye s.r.l. Padua Italy Department of Information Engineering University of Padova Padua35131 Italy Heraklion70013 Greece Department of Biomedical Research Institute of Molecular Biology and Biotechnology FORTH Ioannina45115 Greece Department of Materials Science and Engineering Unit of Medical Technology and Intelligent Information Systems University of Ioannina Ioannina45110 Greece Department of Electrical and Computer Engineering Hellenic Mediterranean University Heraklion71004 Greece

Purpose: Colour fundus images are widely used in diagnosis treatment decision of several retinal diseases such as diabetic retinopathy (DR), glaucoma and age-related macular degeneration (AMD). These very common conditions must be detected early and monitored to prevent progression and avoid permanent damage. Fundus images revealed to be useful also for the diagnosis of cataracts. In recent years, various automatic diagnostic support methods have been proposed in the literature, with the aim of facilitating widespread screening and obtaining quantitative, objective, and reproducible information. Methods: In this review paper, an overview of traditional, machine learning and modern deep learning techniques for ophthalmic disease diagnosis (i.e. glaucoma, DR, AMD, and cataract) using retinal fundus images is presented. In addition, various publicly available image datasets used for such purposes are described. Results: The current main challenges and findings are identified, as well as common aspects and discrepancies between the various methods developed for the various diseases. Conclusion: The overview of what has been done about all pathologies, rather than only on a specific one, could favour a migration of the best solutions and, hopefully, the development of a more precise and clinically useful automatic analysis of all pathologies. Important critical insights and research trends are also discussed. Graphical abstract: (Figure presented.) © The Author(s), under exclusive licence to The Brazilian Society of Biomedical engineering 2023.

关键词： Pathology

来源：评论

学校读者我要写书评

暂无评论

Benchmarking GAN-Based vs Classical Data Augmentation on Biomedical Images 27th

Benchmarking GAN-Based vs Classical Data Augmentation on Bi...

引用

27th International Conference on Pattern Recognition, ICPR 2024

作者： Mantegna, Massimiliano Tronchin, Lorenzo Tortora, Matteo Soda, Paolo Unit of Computer Systems and Bioinformatics Department of Engineering University Campus Bio-Medico of Rome Rome Italy Department of Naval Electrical Electronics and Telecommunications Engineering University of Genova Via all’Opera Pia 11a Genoa16145 Italy Department of Diagnostics and Intervention Radiation Physics Biomedical Engineering Umeå University Umeå Sweden Department of Computing Science Umeå University Umeå Sweden Messina Italy

ISBN: (纸本)9783031876592

The medical field faces significant data shortages due to the high image acquisition and maintenance costs. Data Augmentation aims to mitigate this by increasing data availability and enhancing image generalization. However, traditional DA methods often produce data with limited quality and diversity. Generative Adversarial Networks present a promising alternative, offering potential solutions to data scarcity issues. This paper evaluates the impact of GAN-based data augmentation in medical imaging and provides a benchmark for the efficacy of synthetic data in downstream classification tasks. To this aim, we performed a wide set of tests using three different GAN architectures on six 2D datasets from the standardized MedMNIST biomedical image collection, conducting a total of 696 experiments. Our results reveal that while GAN-based DA methods show promise with low-dimensional datasets, traditional DA methods still outperform them. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： Mammography

来源：评论

学校读者我要写书评

暂无评论

Named Entity Recognition in Italian Lung Cancer Clinical Reports using Transformers

Named Entity Recognition in Italian Lung Cancer Clinical Rep...

引用

2023 IEEE International Conference on bioinformatics and Biomedicine, BIBM 2023

作者： Paolo, Domenico Bria, Alessandro Greco, Carlo Russano, Marco Ramella, Sara Soda, Paolo Sicilia, Rosa Università Campus Bio-Medico di Roma Unit of Computer Systems & Bioinformatics Department of Engineering Italy University of Cassino and Southern Latium Department of Electrical and Information Engineering Cassino Italy Università Campus Bio-Medico di Roma Research Unit of Radiation Oncology Department of Medicine and Surgery Italy Fondazione Policlinico Universitario Campus Bio-Medico Operative Research Unit of Radiation Oncology Italy Fondazione Policlinico Universitario Campus Bio-Medico Operative Research Unit of Medical Oncology Italy Umeå University Department of Radiation Sciences Radiation Physics Biomedical Engineering Sweden

ISBN: (纸本)9798350337488

The widespread adoption of electronic health records (EHRs) offers a valuable opportunity to support clinical research by containing crucial patient information, including diagnoses, symptoms, medications, lab tests, and more. Despite the success of deep learning for biomedical Named Entity Recognition (NER), the literature in this field still presents a gap regarding applications focused on lung cancer for the Italian language. Hence, this paper presents a transformer-based approach to extract named entities from Italian clinical notes related to Non-Small Cell Lung Cancer (NSCLC). We introduce a novel set of 25 clinical entities related to NSCLC building a corpus annotated for NER. We apply a state-of the-art model pre-trained on Italian biomedical texts to the manually annotated clinical reports of a cohort of 257 patients suffering from NSCLC, successfully dealing with class-imbalance problems and obtaining promising performance (average F1-score of 84.3%). We also compared our method with two other pre-trained state-of-the-art models showing that the domain specific knowledge offered by the proposed approach is necessary to achieve higher performance. These findings also showcase the feasibility of using transformers to extract biomedical information in the Italian language. © 2023 IEEE.

关键词： deep learning EHRs NER NSCLC trasformer

来源：评论

学校读者我要写书评

暂无评论

Not Another Imputation Method: A Transformer-Based Model for Missing Values in Tabular Datasets

SSRN

引用

SSRN 2025年

作者： Caruso, Camillo Maria Soda, Paolo Guarrasi, Valerio Research Unit of Computer Systems and Bioinformatics Department of Engineering Università Campus Bio-Medico di Roma Roma Italy Department of Diagnostics and Intervention Radiation Physics Biomedical Engineering Umeå University Umeå Sweden

Handling missing values in tabular datasets presents a significant challenge in training and testing artificial intelligence models, an issue usually addressed using imputation techniques. Here we introduce "Not Another Imputation Method" (NAIM), a novel transformer-based model specifically designed to address this issue without the need for traditional imputation techniques. NAIM employs feature-specific embeddings and a masked self-attention mechanism that effectively learns from available data, thus avoiding the necessity to impute missing values. Additionally, a novel regularization technique is introduced to enhance the model’s generalization capability from incomplete data. We extensively evaluated NAIM on 5 publicly available tabular datasets, demonstrating its superior performance over 6 state-of-the-art machine learning models and 4 deep learning models, each paired with 3 different imputation techniques when necessary. The results highlight the efficacy of NAIM in improving predictive performance and resilience in the presence of missing data. To facilitate further research and practical application in handling missing data without traditional imputation methods, we made the code for NAIM available at https://***/cosbidev/NAIM. © 2025, The Authors. All rights reserved.

关键词： Adversarial 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

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

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：