检索结果-内蒙古大学图书馆

Enhancing Automated Microorganism Classification: A Study on the Influence of Compression, Non-Uniform Scaling, and Data Augmentation Using Deep Learning

SSRN

引用

SSRN 2024年

作者： Boukouvalas, Dimitria Theophanis Bissaco, Márcia A.S. Dellê, Humberto Deana, Alessandro M. Belan, Peterson A. de Araújo, Sidnei Alves Universidade Nove de Julho Informatics and Knowledge Management Graduate Program Rua Vergueiro 235/249 SP São Paulo01504-001 Brazil Universidade de Mogi das Cruzes Biomedical Engineering Graduate Program Av. Dr. Cândido Xavier de Almeida e Souza 200 Mogi das Cruzes SP São Paulo08780-911 Brazil Universidade Nove de Julho Medicine Graduate Program Rua Vergueiro 235/249 SP São Paulo01504-001 Brazil

With the advent of "Laboratory 4.0", a concept that represents the recent technological breakthroughs in laboratory automation, there is an increasing demand for solutions that can automate, standardize, and enhance the accuracy and speed of microorganism classification. One promising approach is the use of convolutional neural networks (CNNs), a powerful deep learning technique that has shown significant results in biomedical image analysis. However, its application in microorganism classification is still nascent mainly due to the unavailability of extensive datasets, which is a basic requirement to ensure generalization ability in learning CNNs. This study aims to fill a significant gap in the literature by exploring the influence of factors such as image compression, non-uniform scaling, and data augmentation on the performance of CNNs, specifically when using small image datasets of microorganisms. We focus on two CNN architectures, AlexNet and DenseNet-121, which have shown promising results in previous microorganism classification studies. Our experiments reveal that certain factors like the type of data compression and non-uniform scaling applied to the images can negatively impact the performance of CNNs. On the other hand, data augmentation techniques such as mirroring, rotation, and noise addition can enhance their performance. By utilizing training datasets and optimized parameters based on our findings, we were able to achieve classification accuracies of 98.61% and 99.82% with AlexNet and DenseNet-121, respectively, surpassing the results reported in current state-of-the-art studies. This research has significant implications for Laboratory 4.0, potentially reducing dataset preparation and CNN training time, lowering computational costs, and improving the consistency and accuracy of microorganism classification. It paves the way for more efficient and accurate microorganism classification methods in automated laboratories. © 2024, The Authors. All righ

关键词： Classification (of information)

来源：评论

学校读者我要写书评

暂无评论

The effect of stereochemical constraints on the radius of gyration of folded proteins

arXiv

引用

arXiv 2025年

作者： Logan, Jack A. Sumner, Jacob Grigas, Alex T. Shattuck, Mark D. O'Hern, Corey S. Department of Mechanical Engineering Yale University New HavenCT06520 United States Graduate Program in Computational Biology & Biomedical Informatics Yale University New HavenCT06520 United States Integrated Graduate Program in Physical and Engineering Biology Yale University New HavenCT06520 United States Benjamin Levich Institute and Physics Department The City College of New York New YorkNY10031 United States Department of Applied Physics Yale University New HavenCT06520 United States Department of Physics Yale University New HavenCT06520 United States

Proteins are composed of linear chains of amino acids that fold into complex three-dimensional structures. A general feature of folded proteins is that they are compact with a radius of gyration Rg(N) ∼ Nν that obeys power-law scaling with the number of amino acids N in each protein and ν ∼ 1/3. In this study, we investigate the internal scaling of the radius of gyration Rg(n) versus the chemical separation n between amino acids for all subchains of length n. We show that for globular proteins Rg(n) ∼ nν1,2 with a larger exponent ν1 > 1/3 for small n and a smaller exponent ν2 g(N) ∝ Nν. To describe this scaling behavior for Rg(n), we carry out folding simulations for a series of coarse-grained models for proteins beginning with the freely-jointed and freely-rotating chain models composed of spherical monomers and varying degrees of stereochemical constraints. We show that a minimal model, which coarse-grains amino acids into a single spherical backbone atom and one variable-sized spherical side-chain atom, and enforces bend- and dihedral-angle constraints, can recapitulate Rg(n) for x-ray crystal structures of globular proteins. In addition, this model predicts the correct average packing fraction and size of the hydrophobic core, which are two key physical features that can be used to distinguish between computational 'decoys' and correctly folded proteins in protein design applications. © 2025, CC BY.

关键词： Spheres

来源：评论

学校读者我要写书评

暂无评论

A ghost mechanism: An analytical model of abrupt learning

arXiv

引用

arXiv 2025年

作者： Dinc, Fatih Cirakman, Ege Jiang, Yiqi Yuksekgonul, Mert Schnitzer, Mark Tanaka, Hidenori CNC Program Stanford University StanfordCA94305 United States Physics & Informatics Laboratories NTT Research Inc. SunnyvaleCA94085 United States Stanford University StanfordCA94305 United States James H. Clark Center for Biomedical Engineering & Sciences Stanford University Stanford United States Howard Hughes Medical Institute Stanford University StanfordCA94305 United States CBS-NTT Program in Physics of Intelligence Harvard University CambridgeMA94305 United States

Abrupt learning is commonly observed in neural networks, where long plateaus in network performance are followed by rapid convergence to a desirable solution. Yet, despite its common occurrence, the complex interplay of task, network architecture, and learning rule has made it difficult to understand the underlying mechanisms. Here, we introduce a minimal dynamical system trained on a delayed-activation task and demonstrate analytically how even a one-dimensional system can exhibit abrupt learning through ghost points rather than bifurcations. Through our toy model, we show that the emergence of a ghost point destabilizes learning dynamics. We identify a critical learning rate that prevents learning through two distinct loss landscape features: a no-learning zone and an oscillatory minimum. Testing these predictions in recurrent neural networks (RNNs), we confirm that ghost points precede abrupt learning and accompany the destabilization of learning. We demonstrate two complementary remedies: lowering the model output confidence prevents the network from getting stuck in no-learning zones, while increasing trainable ranks beyond task requirements (i.e., adding sloppy parameters) provides more stable learning trajectories. Our model reveals a bifurcation-free mechanism for abrupt learning and illustrates the importance of both deliberate uncertainty and redundancy in stabilizing learning dynamics. © 2025, CC BY.

关键词： Recurrent neural networks

来源：评论

学校读者我要写书评

暂无评论

Improving quantitation accuracy in isobaric-labeling mass spectrometry experiments with spectral library searching and feature-based peptide-spectrum match filter

引用

Scientific Reports 2023年第1期13卷 1-10页

作者： Kuo, Tzu-Yun Wang, Jen-Hung Huang, Yung-Wen Sung, Ting-Yi Chen, Ching-Tai Department of Biochemical Science and Technology National Taiwan University Taipei 10617 Taiwan Bioinformatics Program Taiwan International Graduate Program Institute of Statistical Science Academia Sinica Taipei 11529 Taiwan Institute of Information Science Academia Sinica Taipei 11529 Taiwan Institute of Biomedical Informatics National Yang Ming Chiao Tung University Taipei 11221 Taiwan Department of Computer Science and Information Engineering National Taiwan University Taipei 10617 Taiwan Department of Bioinformatics and Biomedical Engineering Asia University Taichung 41354 Taiwan Center for Precision Health Research Asia University Taichung 41354 Taiwan

Isobaric labeling relative quantitation is one of the dominating proteomic quantitation technologies. Traditional quantitation pipelines for isobaric-labeled mass spectrometry data are based on sequence database searching. In this study, we present a novel quantitation pipeline that integrates sequence database searching, spectral library searching, and a feature-based peptide-spectrum-match (PSM) filter using various spectral features for filtering. The combined database and spectral library searching results in larger quantitation coverage, and the filter removes PSMs with larger quantitation errors, retaining those with higher quantitation accuracy. Quantitation results show that the proposed pipeline can improve the overall quantitation accuracy at the PSM and protein levels. To our knowledge, this is the first study that utilizes spectral library searching to improve isobaric labeling-based quantitation. For users to conveniently perform the proposed pipeline, we have implemented the feature-based filter being executable on both Windows and Linux platforms;its executable files, user manual, and sample data sets are freely available at https://***/comics/Software_*** . Furthermore, with the developed filter, the proposed pipeline is fully compatible with the Trans-Proteomic Pipeline. © 2023, Springer Nature Limited.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Generalization is not a universal guarantee: Estimating similarity to training data with an ensemble out-of-distribution metric

arXiv

引用

arXiv 2025年

作者： Schreyer, W. Max Anderson, Christopher Thompson, Reid F. Computational Biology Program Oregon Health & Science University PortlandOR United States Department of Biomedical Engineering Oregon Health & Science University PortlandOR United States Portland VA Research Foundation PortlandOR United States Department of Radiation Medicine Oregon Health & Science University PortlandOR United States Department of Medical Informatics and Clinical Epidemiology Oregon Health & Science University PortlandOR United States Division of Hospital and Specialty Medicine VA Portland Healthcare System PortlandOR United States

Failure of machine learning models to generalize to new data is a core problem limiting the reliability of AI systems, partly due to the lack of simple and robust methods for comparing new data to the original training dataset. We propose a standardized approach for assessing data similarity in a model-agnostic manner by constructing a supervised autoencoder for generalizability estimation (SAGE). We compare points in a low-dimensional embedded latent space, defining empirical probability measures for k-Nearest Neighbors (kNN) distance, reconstruction of inputs and task-based performance. As proof of concept for classification tasks, we use MNIST and CIFAR-10 to demonstrate how an ensemble output probability score can separate deformed images from a mixture of typical test examples, and how this SAGE score is robust to transformations of increasing severity. As further proof of concept, we extend this approach to a regression task using non-imaging data (UCI Abalone). In all cases, we show that out-of-the-box model performance increases after SAGE score filtering, even when applied to data from the model’s own training and test datasets. Our out-of-distribution scoring method can be introduced during several steps of model construction and assessment, leading to future improvements in responsible deep learning implementation. Copyright © 2025, The Authors. All rights reserved.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Fair Patient Model: Mitigating Bias in the Patient Representation Learned from the Electronic Health Records

arXiv

引用

arXiv 2023年

作者： Sivarajkumar, Sonish Huang, Yufei Wang, Yanshan Intelligent Systems Program School of Computing and Information University of Pittsburgh PittsburghPA United States Department of Electrical and Computer Engineering University of Pittsburgh PittsburghPA United States Department of Health Information Management University of Pittsburgh PittsburghPA United States Department of Biomedical Informatics University of Pittsburgh PittsburghPA United States Clinical and Translational Science Institute University of Pittsburgh PittsburghPA United States

Objective: To pre-train fair and unbiased patient representations from Electronic Health Records (EHRs) using a novel weighted loss function that reduces bias and improves fairness in deep representation learning models. Methods: We defined a new loss function, called weighted loss function, in the deep representation learning model to balance the importance of different groups of patients and features. We applied the proposed model, called Fair Patient Model (FPM), to a sample of 34,739 patients from the MIMIC-III dataset and learned patient representations for four clinical outcome prediction tasks. Results: FPM outperformed the baseline models in terms of three fairness metrics: demographic parity, equality of opportunity difference, and equalized odds ratio. FPM also achieved comparable predictive performance with the baselines, with an average accuracy of 0.7912. Feature analysis revealed that FPM captured more information from clinical features than the baselines. Conclusion: FPM is a novel method to pre-train fair and unbiased patient representations from EHR data using a weighted loss function. The learned representations can be used for various downstream tasks in healthcare and can be extended to other domains where bias and fairness are important. Copyright © 2023, The Authors. All rights reserved.

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

A Clinical Decision Support System of Pressure Ulcers Tissue Classification 21

A Clinical Decision Support System of Pressure Ulcers Tissue...

引用

5th International Conference on Medical and Health informatics, ICMHI 2021

作者： Chang, Dun-Hao Chu, Po-Jui Li, Yi-Jhen Ning, Chun-Kai Chien, Ting-Ying Department of Information Management Yuan Ze University Taoyuan City Taiwan Department of Plastic and Aesthetic Surgery Far Eastern Memorial Hospital Taiwan Department of Computer Science and Engineering Yuan Ze University Taoyuan City Taiwan Program in Biomedical Informatics Yuan Ze University Taoyuan City Taiwan Innovation Center for Big Data and Digital Convergence Yuan Ze University Taoyuan City Taiwan

ISBN: (纸本)9781450389846

Pressure ulcers (PUs) are a common problem associated with great morbidity and cost. Because of the lack of professional personnel in the institutions or home care system, the diagnosis and treatment of PUs were sometimes delayed. We aim to develop an automatic tissue classification and severity evaluation system of PUs using AI technology. All PU images were collected from patients in the Far Eastern Memorial Hospital (FEMH) in recent 3 years and labeled according to different tissue types. The labeled images were used to train the tissue segmentation model with U-net convolutional neural network (CNN). The percentage of the different tissue can be calculated automatically and would be compared with the manually labeled one. The output of tissue classification was transformed to a clinical decision support system (CDSS) with a rule-base formula. Our study showed that proposed system can classify each tissue automatically and reach an accuracy about 93.6%. However, the CDSS's accuracy rate was only 64% for the referral suggestion and 62% for the care recommendation. Hence, the system should be modified with more data and information. © 2021 ACM.

关键词： Tissue

来源：评论

学校读者我要写书评

暂无评论

Multi-Task Learning U-Net for Functional Shoulder Sub-Task Segmentation

Multi-Task Learning U-Net for Functional Shoulder Sub-Task S...

引用

Annual International Conference of the IEEE engineering in Medicine and Biology Society (EMBC)

作者： En-Ping Chu Kai-Chun Liu Chia-Yeh Hsieh Chih-Ya Chang Yu Tsao Chia-Tai Chan Department of Biomedical Engineering National Yang Ming Chiao Tung University Taipei Taiwan Research Center for Information Technology Innovation (CITI) at Academia Sinica Taipei Taiwan Bachelor’s Program in Medical Informatics and Innovative Applications Fu Jen Catholic University New Taipei City Taiwan Department of Physical Medicine and Rehabilitation School of Medicine National Defense Medical Center Tri-Service General Hospital Taipei Taiwan

The assessment of a frozen shoulder (FS) is critical for evaluating outcomes and medical treatment. Analysis of functional shoulder sub-tasks provides more crucial information, but current manual labeling methods are time-consuming and prone to errors. To address this challenge, we propose a deep multi-task learning (MTL) U-Net to provide an automatic and reliable functional shoulder sub-task segmentation (STS) tool for clinical evaluation in FS. The proposed approach contains the main task of STS and the auxiliary task of transition point detection (TPD). For the main STS task, a U-Net architecture including an encoder-decoder with skip connection is presented to perform shoulder sub-task classification for each time point. The auxiliary TPD task uses lightweight convolutional neural networks architecture to detect the boundary between shoulder sub-tasks. A shared structure is implemented between two tasks and their objective functions of them are optimized jointly. The fine-grained transition-related information from the auxiliary TPD task is expected to help the main STS task better detect boundaries between functional shoulder sub-tasks. We conduct the experiments using wearable inertial measurement units to record 815 shoulder task sequences collected from 20 healthy subjects and 43 patients with FS. The experimental results present that the deep MTL U-Net can achieve superior performance compared to using single-task models. It shows the effectiveness of the proposed method for functional shoulder STS. The code has been made publicly available at https://***/RobinChu9890/*** Relevance— This work provides an automatic and reliable functional shoulder sub-task segmentation tool for clinical evaluation in frozen shoulder.

关键词：

来源：评论

学校读者我要写书评

暂无评论

No winners: Performance of lung cancer prediction models depends on screening-detected, incidental, and biopsied pulmonary nodule use cases.

arXiv

引用

arXiv 2024年

作者： Li, Thomas Z. Xu, Kaiwen Krishnan, Aravind Gao, Riqiang Kammer, Michael N. Antic, Sanja Xiao, David Knight, Michael Martinez, Yency Paez, Rafael Lentz, Robert J. Deppen, Stephen Grogan, Eric L. Lasko, Thomas A. Sandler, Kim L. Maldonado, Fabien Landman, Bennett A. Medical Scientist Training Program Vanderbilt University NashvilleTN37235 United States Biomedical Engineering Vanderbilt University NashvilleTN37235 United States Computer Science Vanderbilt University NashvilleTN37235 United States Electrical and Computer Engineering Vanderbilt University NashvilleTN37235 United States Digital Technology and Innovation Siemens Healthineers PrincetonNJ08540 United States Division of Allergy Pulmonary and Critical Care Medicine Department of Medicine Vanderbilt University Medical Center NashvilleTN37232 United States Department of Thoracic Surgery Vanderbilt University Medical Center NashvilleTN37232 United States Biomedical Informatics Vanderbilt University Medical Center NashvilleTN37232 United States Radiology and Radiological Sciences Vanderbilt University Medical Center NashvilleTN37232 United States

Purpose: Statistical models for predicting lung cancer have the potential to facilitate earlier diagnosis of malignancy and avoid invasive workup of benign disease. Many models have been published, but comparative studies of their utility in different clinical settings in which patients would arguably most benefit are scarce. Materials and Methods: This study retrospectively evaluated promising predictive models for lung cancer prediction in three clinical settings: lung cancer screening with low-dose computed tomography, incidentally detected pulmonary nodules, and nodules deemed suspicious enough to warrant a biopsy. We leveraged 9 cohorts (n=898, 896, 882, 219, 364, 117, 131, 115, 373) from multiple institutions to assess the area under the receiver operating characteristic curve (AUC) of validated models including logistic regressions on clinical variables and radiologist nodule characterizations, artificial intelligence on chest CTs, longitudinal imaging AI, and multi-modal approaches. We implemented each model from their published literature, re-training the models if necessary, and curated each cohort from primary data sources. Results: We observed that model performance varied greatly across clinical use cases. No single predictive model emerged as a clear winner across all cohorts, but certain models excelled in specific clinical contexts. Single timepoint chest CT AI performed well in lung screening, but struggled to generalize to other clinical settings. Longitudinal imaging and multimodal models demonstrated comparatively promising performance on incidentally-detected nodules. However, when applied to nodules that underwent biopsy, all models underperformed. Conclusion: These results underscore the strengths and limitations of 8 validated predictive models and highlight promising directions towards personalized, noninvasive lung cancer diagnosis. © 2024, CC BY-NC-SA.

关键词： Biopsy

来源：评论

学校读者我要写书评

暂无评论

Computational prediction and validation of specific EmbR binding site on PknH

引用

Synthetic and Systems Biotechnology 2021年第4期6卷 429-436页

作者： Insung Na Huanqin Dai Hantian Li Anvita Gupta David Kreda Powell Zhang Xiangyin Chen Lixin Zhang Gil Alterovitz Computational Health Informatics Program Boston Children’s Hospital/Harvard Medical SchoolBostonMA 02115USA Department of Medicine Brigham and Women’s Hospital/Harvard Medical SchoolBostonMA 02115USA State Key Laboratory of Mycology Institute of MicrobiologyBeijing100101 China Department of Computer Sciences Stanford UniversityStanford CA 94305USA Center for Biomedical Informatics Harvard Medical SchoolBostonMA 02115USA State Key Laboratory of Bioreactor Engineering East China University of Science and TechnologyShanghaiChina Department of Biology The Johns Hopkins UniversityBaltimoreMD 21218USA Lexington High School BostonMA 02115USA

Tuberculosis drug resistance continues to threaten global health but the underline molecular mechanisms are not ***(EMB),one of the well-known first-line drugs in tuberculosis treatment is,unfortunately,not free from drug resistance *** studies have shown that some genetic mutations in Mycobacterium tuberculosis(Mtb)EmbR,and EmbC/A/B genes cause EMB ***-PknH pair controls embC/A/B operon,which encodes EmbC/A/B genes,and EMB interacts with EmbA/B ***,the EmbR binding site on PknH was *** conducted molecular simulation on the EmbR-peptides binding structures and discovered phosphorylated PknH 273-280(N′-HEALS^(P)DPD-C′)makesβstrand with the EmbR FHA domain,asβ-MoRF(MoRF;molecular recognition feature)does at its binding *** bond number analysis also supported the peptides’β-MoRF forming activity at the EmbR FHA ***,we discovered that previously known phosphorylation residues might have their chronological order according to the phosphorylation *** discovery validated that Mtb PknH 273-280(N′-HEALSDPD-C′)has reliable EmbR binding *** approach is revolutionary in the computer-aided drug discovery field,because it is the first trial to discover the protein-protein interaction site,and find binding partner in nature from this site.

关键词： Disorder-to-order transition Protein intrinsic disorder Binding site prediction Drug resistance Molecular simulation

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：