Simone Scalise, Giorgio Gosti, Giancarlo Ruocco, Giovanna Peruzzi, Mattia Miotto
The ability of a cancer cell population to achieve heterogeneity in their phenotype distributions offers advantages in tumor invasiveness and drug resistance. Studying the mechanisms behind such observed heterogeneity in mammalian cells presents challenges due for instance to the prolonged proliferation times compared to widely studied unicellular organisms like bacteria and yeast. Here, we studied the response of leukemia cell populations to serum starvation via a protocol, we recently developed, that makes use of live cell fluorescence and flow cytometry in combination with a quantitative analytical model to follow the population proliferation while monitoring the dynamics of its phenotype distributions. We found that upon switching between a serum-rich to a serum-poor media, leukemia cells (i) maintain a memory of the previous environment up to one generation even in the presence of severe medium-depletion, before (ii) adapting their growth and division rates to the novel environment while preserving a sizer-like division strategy. Finally, looking at the mitochondria content of the proliferating vs non-proliferating cells, we found that the latter is characterized by a higher number of older mitochondria, suggesting a possible functional role of the observed asymmetric partitioning of (aged) mitochondria in leukemia cells.
{"title":"Probing leukemia cells behavior under starvation","authors":"Simone Scalise, Giorgio Gosti, Giancarlo Ruocco, Giovanna Peruzzi, Mattia Miotto","doi":"arxiv-2408.09219","DOIUrl":"https://doi.org/arxiv-2408.09219","url":null,"abstract":"The ability of a cancer cell population to achieve heterogeneity in their\u0000phenotype distributions offers advantages in tumor invasiveness and drug\u0000resistance. Studying the mechanisms behind such observed heterogeneity in\u0000mammalian cells presents challenges due for instance to the prolonged\u0000proliferation times compared to widely studied unicellular organisms like\u0000bacteria and yeast. Here, we studied the response of leukemia cell populations\u0000to serum starvation via a protocol, we recently developed, that makes use of\u0000live cell fluorescence and flow cytometry in combination with a quantitative\u0000analytical model to follow the population proliferation while monitoring the\u0000dynamics of its phenotype distributions. We found that upon switching between a\u0000serum-rich to a serum-poor media, leukemia cells (i) maintain a memory of the\u0000previous environment up to one generation even in the presence of severe\u0000medium-depletion, before (ii) adapting their growth and division rates to the\u0000novel environment while preserving a sizer-like division strategy. Finally,\u0000looking at the mitochondria content of the proliferating vs non-proliferating\u0000cells, we found that the latter is characterized by a higher number of older\u0000mitochondria, suggesting a possible functional role of the observed asymmetric\u0000partitioning of (aged) mitochondria in leukemia cells.","PeriodicalId":501266,"journal":{"name":"arXiv - QuanBio - Quantitative Methods","volume":"111 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-08-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142213359","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Honggen Zhang, Xiangrui Gao, June Zhang, Lipeng Lai
Messenger RNA (mRNA)-based vaccines are accelerating the discovery of new drugs and revolutionizing the pharmaceutical industry. However, selecting particular mRNA sequences for vaccines and therapeutics from extensive mRNA libraries is costly. Effective mRNA therapeutics require carefully designed sequences with optimized expression levels and stability. This paper proposes a novel contextual language model (LM)-based embedding method: mRNA2vec. In contrast to existing mRNA embedding approaches, our method is based on the self-supervised teacher-student learning framework of data2vec. We jointly use the 5' untranslated region (UTR) and coding sequence (CDS) region as the input sequences. We adapt our LM-based approach specifically to mRNA by 1) considering the importance of location on the mRNA sequence with probabilistic masking, 2) using Minimum Free Energy (MFE) prediction and Secondary Structure (SS) classification as additional pretext tasks. mRNA2vec demonstrates significant improvements in translation efficiency (TE) and expression level (EL) prediction tasks in UTR compared to SOTA methods such as UTR-LM. It also gives a competitive performance in mRNA stability and protein production level tasks in CDS such as CodonBERT.
{"title":"mRNA2vec: mRNA Embedding with Language Model in the 5'UTR-CDS for mRNA Design","authors":"Honggen Zhang, Xiangrui Gao, June Zhang, Lipeng Lai","doi":"arxiv-2408.09048","DOIUrl":"https://doi.org/arxiv-2408.09048","url":null,"abstract":"Messenger RNA (mRNA)-based vaccines are accelerating the discovery of new\u0000drugs and revolutionizing the pharmaceutical industry. However, selecting\u0000particular mRNA sequences for vaccines and therapeutics from extensive mRNA\u0000libraries is costly. Effective mRNA therapeutics require carefully designed\u0000sequences with optimized expression levels and stability. This paper proposes a\u0000novel contextual language model (LM)-based embedding method: mRNA2vec. In\u0000contrast to existing mRNA embedding approaches, our method is based on the\u0000self-supervised teacher-student learning framework of data2vec. We jointly use\u0000the 5' untranslated region (UTR) and coding sequence (CDS) region as the input\u0000sequences. We adapt our LM-based approach specifically to mRNA by 1)\u0000considering the importance of location on the mRNA sequence with probabilistic\u0000masking, 2) using Minimum Free Energy (MFE) prediction and Secondary Structure\u0000(SS) classification as additional pretext tasks. mRNA2vec demonstrates\u0000significant improvements in translation efficiency (TE) and expression level\u0000(EL) prediction tasks in UTR compared to SOTA methods such as UTR-LM. It also\u0000gives a competitive performance in mRNA stability and protein production level\u0000tasks in CDS such as CodonBERT.","PeriodicalId":501266,"journal":{"name":"arXiv - QuanBio - Quantitative Methods","volume":"1 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-08-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142213361","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Mate NagyMTA-ELTE Statistical and Biological Physics Research Group, Budapest, HungaryMTA-ELTE Lendulet Collective Behaviour Research Group, Budapest, HungaryDepartment of Biological Physics, Eotvos Lorand University, Budapest, HungaryDepartment of Mechanical and Aerospace Engineering, Princeton University, Princeton, NJ, USAPrinceton Neuroscience Institute, Princeton University, Princeton, NJ, USALewis-Sigler Institute for Integrative Genomics, Princeton University, Princeton, NJ, USA, Jacob D. DavidsonDepartment of Mechanical and Aerospace Engineering, Princeton University, Princeton, NJ, USAPrinceton Neuroscience Institute, Princeton University, Princeton, NJ, USALewis-Sigler Institute for Integrative Genomics, Princeton University, Princeton, NJ, USA, Gabor VasarhelyiMTA-ELTE Statistical and Biological Physics Research Group, Budapest, HungaryDepartment of Biological Physics, Eotvos Lorand University, Budapest, Hungary, Daniel AbelMTA-ELTE Statistical and Biological Physics Research Group, Budapest, Hungary, Eniko KubinyiDepartment of Ethology, Eotvos Lorand University, Budapest, HungaryMTA-ELTE Comparative Ethology Research Group, Budapest, HungaryResearch Centre for Natural Sciences, Budapest, Hungary, Ahmed El HadyDepartment of Mechanical and Aerospace Engineering, Princeton University, Princeton, NJ, USAPrinceton Neuroscience Institute, Princeton University, Princeton, NJ, USALewis-Sigler Institute for Integrative Genomics, Princeton University, Princeton, NJ, USA, Tamas VicsekMTA-ELTE Statistical and Biological Physics Research Group, Budapest, HungaryDepartment of Biological Physics, Eotvos Lorand University, Budapest, Hungary
Rodents serve as an important model for examining both individual and collective behavior. Dominance within rodent social structures can determine access to critical resources, such as food and mating opportunities. Yet, many aspects of the intricate interplay between individual behaviors and the resulting group social hierarchy, especially its evolution over time, remain unexplored. In this study, we utilized an automated tracking system that continuously monitored groups of male rats for over 250 days to enable an in-depth analysis of individual behavior and the overarching group dynamic. We describe the evolution of social structures within a group and additionally investigate how past behaviors influence the emergence of new social hierarchies when group composition and experimental area changes. Notably, we find that conventional individual and pairwise tests exhibit a weak correlation with group behavior, highlighting their limited accuracy in predicting behavioral outcomes in a collective context. These results emphasize the context-dependence of social behavior as an emergent property of interactions within a group and highlight the need to measure and quantify social behavior in more naturalistic environments.
{"title":"Long-term tracking of social structure in groups of rats","authors":"Mate NagyMTA-ELTE Statistical and Biological Physics Research Group, Budapest, HungaryMTA-ELTE Lendulet Collective Behaviour Research Group, Budapest, HungaryDepartment of Biological Physics, Eotvos Lorand University, Budapest, HungaryDepartment of Mechanical and Aerospace Engineering, Princeton University, Princeton, NJ, USAPrinceton Neuroscience Institute, Princeton University, Princeton, NJ, USALewis-Sigler Institute for Integrative Genomics, Princeton University, Princeton, NJ, USA, Jacob D. DavidsonDepartment of Mechanical and Aerospace Engineering, Princeton University, Princeton, NJ, USAPrinceton Neuroscience Institute, Princeton University, Princeton, NJ, USALewis-Sigler Institute for Integrative Genomics, Princeton University, Princeton, NJ, USA, Gabor VasarhelyiMTA-ELTE Statistical and Biological Physics Research Group, Budapest, HungaryDepartment of Biological Physics, Eotvos Lorand University, Budapest, Hungary, Daniel AbelMTA-ELTE Statistical and Biological Physics Research Group, Budapest, Hungary, Eniko KubinyiDepartment of Ethology, Eotvos Lorand University, Budapest, HungaryMTA-ELTE Comparative Ethology Research Group, Budapest, HungaryResearch Centre for Natural Sciences, Budapest, Hungary, Ahmed El HadyDepartment of Mechanical and Aerospace Engineering, Princeton University, Princeton, NJ, USAPrinceton Neuroscience Institute, Princeton University, Princeton, NJ, USALewis-Sigler Institute for Integrative Genomics, Princeton University, Princeton, NJ, USA, Tamas VicsekMTA-ELTE Statistical and Biological Physics Research Group, Budapest, HungaryDepartment of Biological Physics, Eotvos Lorand University, Budapest, Hungary","doi":"arxiv-2408.08945","DOIUrl":"https://doi.org/arxiv-2408.08945","url":null,"abstract":"Rodents serve as an important model for examining both individual and\u0000collective behavior. Dominance within rodent social structures can determine\u0000access to critical resources, such as food and mating opportunities. Yet, many\u0000aspects of the intricate interplay between individual behaviors and the\u0000resulting group social hierarchy, especially its evolution over time, remain\u0000unexplored. In this study, we utilized an automated tracking system that\u0000continuously monitored groups of male rats for over 250 days to enable an\u0000in-depth analysis of individual behavior and the overarching group dynamic. We\u0000describe the evolution of social structures within a group and additionally\u0000investigate how past behaviors influence the emergence of new social\u0000hierarchies when group composition and experimental area changes. Notably, we\u0000find that conventional individual and pairwise tests exhibit a weak correlation\u0000with group behavior, highlighting their limited accuracy in predicting\u0000behavioral outcomes in a collective context. These results emphasize the\u0000context-dependence of social behavior as an emergent property of interactions\u0000within a group and highlight the need to measure and quantify social behavior\u0000in more naturalistic environments.","PeriodicalId":501266,"journal":{"name":"arXiv - QuanBio - Quantitative Methods","volume":"43 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-08-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142213360","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Complex behavioral contagion in collective evasion of mobile animal groups can be predicted by reconstructing quantitative interaction networks. Based on the assumption of time-scale separation between a fast contagion process and a slower movement response, the underlying interaction networks have been previously assumed to be static, determined by the spatial structure at the onset of the collective escape response. This idealization does not account for the temporal evolution of the spatial network structure, which may have a major impact on the behavioral contagion dynamics. Here, we propose a spatially-explicit, agent-based model for the coupling between behavioral contagion and the network dynamics originating from the spreading movement response. We explore the impact of movement parameters (startle speed, initial directionality, and directional noise) on average cascade size. By conducting numerical simulations for different density levels, we show that increasing escape speed suppresses the cascade size in most cases, that the cascade size depends strongly on the movement direction of the initially startled individual, and that large variability in the direction of individual escape movements (rotational noise) will typically promote the spread of behavioral contagion through spatial groups. Our work highlights the importance of accounting for movement dynamics in behavioral contagion, and facilitates our understanding of rapid coordinated response and collective information processing in animal groups.
{"title":"Escape cascades as a behavioral contagion process with adaptive network dynamics","authors":"Wenhan Wu, Xiaoping Zheng, Pawel Romanczuk","doi":"arxiv-2408.05096","DOIUrl":"https://doi.org/arxiv-2408.05096","url":null,"abstract":"Complex behavioral contagion in collective evasion of mobile animal groups\u0000can be predicted by reconstructing quantitative interaction networks. Based on\u0000the assumption of time-scale separation between a fast contagion process and a\u0000slower movement response, the underlying interaction networks have been\u0000previously assumed to be static, determined by the spatial structure at the\u0000onset of the collective escape response. This idealization does not account for\u0000the temporal evolution of the spatial network structure, which may have a major\u0000impact on the behavioral contagion dynamics. Here, we propose a\u0000spatially-explicit, agent-based model for the coupling between behavioral\u0000contagion and the network dynamics originating from the spreading movement\u0000response. We explore the impact of movement parameters (startle speed, initial\u0000directionality, and directional noise) on average cascade size. By conducting\u0000numerical simulations for different density levels, we show that increasing\u0000escape speed suppresses the cascade size in most cases, that the cascade size\u0000depends strongly on the movement direction of the initially startled\u0000individual, and that large variability in the direction of individual escape\u0000movements (rotational noise) will typically promote the spread of behavioral\u0000contagion through spatial groups. Our work highlights the importance of\u0000accounting for movement dynamics in behavioral contagion, and facilitates our\u0000understanding of rapid coordinated response and collective information\u0000processing in animal groups.","PeriodicalId":501266,"journal":{"name":"arXiv - QuanBio - Quantitative Methods","volume":"42 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-08-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141935340","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Richard D. Paul, Johannes Seiffarth, Hanno Scharr, Katharina Nöh
Live-cell microscopy allows to go beyond measuring average features of cellular populations to observe, quantify and explain biological heterogeneity. Deep Learning-based instance segmentation and cell tracking form the gold standard analysis tools to process the microscopy data collected, but tracking in particular suffers severely from low temporal resolution. In this work, we show that approximating cell cycle time distributions in microbial colonies of C. glutamicum is possible without performing tracking, even at low temporal resolution. To this end, we infer the parameters of a stochastic multi-stage birth process model using the Bayesian Synthetic Likelihood method at varying temporal resolutions by subsampling microscopy sequences, for which ground truth tracking is available. Our results indicate, that the proposed approach yields high quality approximations even at very low temporal resolution, where tracking fails to yield reasonable results.
{"title":"Robust Approximate Characterization of Single-Cell Heterogeneity in Microbial Growth","authors":"Richard D. Paul, Johannes Seiffarth, Hanno Scharr, Katharina Nöh","doi":"arxiv-2408.04501","DOIUrl":"https://doi.org/arxiv-2408.04501","url":null,"abstract":"Live-cell microscopy allows to go beyond measuring average features of\u0000cellular populations to observe, quantify and explain biological heterogeneity.\u0000Deep Learning-based instance segmentation and cell tracking form the gold\u0000standard analysis tools to process the microscopy data collected, but tracking\u0000in particular suffers severely from low temporal resolution. In this work, we\u0000show that approximating cell cycle time distributions in microbial colonies of\u0000C. glutamicum is possible without performing tracking, even at low temporal\u0000resolution. To this end, we infer the parameters of a stochastic multi-stage\u0000birth process model using the Bayesian Synthetic Likelihood method at varying\u0000temporal resolutions by subsampling microscopy sequences, for which ground\u0000truth tracking is available. Our results indicate, that the proposed approach\u0000yields high quality approximations even at very low temporal resolution, where\u0000tracking fails to yield reasonable results.","PeriodicalId":501266,"journal":{"name":"arXiv - QuanBio - Quantitative Methods","volume":"7 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-08-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141935344","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Uncertainty quantification enables users to assess the reliability of responses generated by large language models (LLMs). We present a novel Question Rephrasing technique to evaluate the input uncertainty of LLMs, which refers to the uncertainty arising from equivalent variations of the inputs provided to LLMs. This technique is integrated with sampling methods that measure the output uncertainty of LLMs, thereby offering a more comprehensive uncertainty assessment. We validated our approach on property prediction and reaction prediction for molecular chemistry tasks.
{"title":"Question Rephrasing for Quantifying Uncertainty in Large Language Models: Applications in Molecular Chemistry Tasks","authors":"Zizhang Chen, Pengyu Hong, Sandeep Madireddy","doi":"arxiv-2408.03732","DOIUrl":"https://doi.org/arxiv-2408.03732","url":null,"abstract":"Uncertainty quantification enables users to assess the reliability of\u0000responses generated by large language models (LLMs). We present a novel\u0000Question Rephrasing technique to evaluate the input uncertainty of LLMs, which\u0000refers to the uncertainty arising from equivalent variations of the inputs\u0000provided to LLMs. This technique is integrated with sampling methods that\u0000measure the output uncertainty of LLMs, thereby offering a more comprehensive\u0000uncertainty assessment. We validated our approach on property prediction and\u0000reaction prediction for molecular chemistry tasks.","PeriodicalId":501266,"journal":{"name":"arXiv - QuanBio - Quantitative Methods","volume":"307 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-08-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141935345","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
João César Reis Alves, Gabriel Rodrigues Palma, Idemauro Antonio Rodrigues de Lara
Sensory analysis is an important area that the food industry can use to innovate and improve its products. This study involves a sample of individuals who can be trained or not to assess a product using a hedonic scale or notes, where the experimental design is a balanced incomplete block design. In this context, integrating sensory analysis with effective statistical methods, which consider the nature of the response variables, is essential to answer the aim of the experimental study. Some techniques are available to analyse sensory data, such as response surface models or categorical models. This article proposes using beta regression as an alternative to the proportional odds model, addressing some convergence problems, especially regarding the number of parameters. Moreover, the beta distribution is flexible for heteroscedasticity and asymmetry data. To this end, we conducted simulation studies that showed agreement rates in product selection using both models. Also, we presented a motivational study that was developed to select prebiotic drinks based on cashew nuts added to grape juice. In this application, the beta regression mixed model results corroborated with the selected formulations using the proportional mixed model.
{"title":"Beta regression mixed model applied to sensory analysis","authors":"João César Reis Alves, Gabriel Rodrigues Palma, Idemauro Antonio Rodrigues de Lara","doi":"arxiv-2408.03240","DOIUrl":"https://doi.org/arxiv-2408.03240","url":null,"abstract":"Sensory analysis is an important area that the food industry can use to\u0000innovate and improve its products. This study involves a sample of individuals\u0000who can be trained or not to assess a product using a hedonic scale or notes,\u0000where the experimental design is a balanced incomplete block design. In this\u0000context, integrating sensory analysis with effective statistical methods, which\u0000consider the nature of the response variables, is essential to answer the aim\u0000of the experimental study. Some techniques are available to analyse sensory\u0000data, such as response surface models or categorical models. This article\u0000proposes using beta regression as an alternative to the proportional odds\u0000model, addressing some convergence problems, especially regarding the number of\u0000parameters. Moreover, the beta distribution is flexible for heteroscedasticity\u0000and asymmetry data. To this end, we conducted simulation studies that showed\u0000agreement rates in product selection using both models. Also, we presented a\u0000motivational study that was developed to select prebiotic drinks based on\u0000cashew nuts added to grape juice. In this application, the beta regression\u0000mixed model results corroborated with the selected formulations using the\u0000proportional mixed model.","PeriodicalId":501266,"journal":{"name":"arXiv - QuanBio - Quantitative Methods","volume":"160 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-08-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141935346","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Amir Heydari, Abbas Ahmadi, Tae Hyung Kim, Berkin Bilgic
Quantification of tissue parameters using MRI is emerging as a powerful tool in clinical diagnosis and research studies. The need for multiple long scans with different acquisition parameters prohibits quantitative MRI from reaching widespread adoption in routine clinical and research exams. Accelerated parameter mapping techniques leverage parallel imaging, signal modelling and deep learning to offer more practical quantitative MRI acquisitions. However, the achievable acceleration and the quality of maps are often limited. Joint MAPLE is a recent state-of-the-art multi-parametric and scan-specific parameter mapping technique with promising performance at high acceleration rates. It synergistically combines parallel imaging, model-based and machine learning approaches for joint mapping of T1, T2*, proton density and the field inhomogeneity. However, Joint MAPLE suffers from prohibitively long reconstruction time to estimate the maps from a multi-echo, multi-flip angle (MEMFA) dataset at high resolution in a scan-specific manner. In this work, we propose a faster version of Joint MAPLE which retains the mapping performance of the original version. Coil compression, random slice selection, parameter-specific learning rates and transfer learning are synergistically combined in the proposed framework. It speeds-up the reconstruction time up to 700 times than the original version and processes a whole-brain MEMFA dataset in 21 minutes on average, which originally requires ~260 hours for Joint MAPLE. The mapping performance of the proposed framework is ~2-fold better than the standard and the state-of-the-art evaluated reconstruction techniques on average in terms of the root mean squared error.
利用核磁共振成像对组织参数进行定量分析正在成为临床诊断和研究的有力工具。由于需要使用不同的采集参数进行多次长时间扫描，定量 MRI 无法在常规临床和研究检查中得到广泛应用。加速参数映射技术利用并行成像、信号建模和深度学习来提供更实用的定量 MRI 采集。然而，可实现的加速度和制图质量往往受到限制。JointMAPLE是一种最新的多参数和特定扫描参数成像技术，在高加速度下具有良好的性能。它协同结合了并行成像、基于模型和机器学习的方法，用于联合绘制 T1、T2*、质子密度和同质性场图。然而，Joint MAPLE 在以特定扫描方式从高分辨率多回波、多翻转角度（MEMFA）数据集估算图谱时，存在重建时间过长的问题。在这项工作中，我们提出了一种更快的联合 MAPLE 版本，它保留了原始版本的映射性能。在提出的框架中，线圈压缩、随机切片选择、特定参数学习率和迁移学习被协同结合在一起。它将重建时间加快到原始版本的 700 倍，处理全脑 MEMFA 数据集的平均时间为 21 分钟，而联合 MAPLE 原本需要约 260 个小时。
{"title":"Fast Whole-Brain MR Multi-Parametric Mapping with Scan-Specific Self-Supervised Networks","authors":"Amir Heydari, Abbas Ahmadi, Tae Hyung Kim, Berkin Bilgic","doi":"arxiv-2408.02988","DOIUrl":"https://doi.org/arxiv-2408.02988","url":null,"abstract":"Quantification of tissue parameters using MRI is emerging as a powerful tool\u0000in clinical diagnosis and research studies. The need for multiple long scans\u0000with different acquisition parameters prohibits quantitative MRI from reaching\u0000widespread adoption in routine clinical and research exams. Accelerated\u0000parameter mapping techniques leverage parallel imaging, signal modelling and\u0000deep learning to offer more practical quantitative MRI acquisitions. However,\u0000the achievable acceleration and the quality of maps are often limited. Joint\u0000MAPLE is a recent state-of-the-art multi-parametric and scan-specific parameter\u0000mapping technique with promising performance at high acceleration rates. It\u0000synergistically combines parallel imaging, model-based and machine learning\u0000approaches for joint mapping of T1, T2*, proton density and the field\u0000inhomogeneity. However, Joint MAPLE suffers from prohibitively long\u0000reconstruction time to estimate the maps from a multi-echo, multi-flip angle\u0000(MEMFA) dataset at high resolution in a scan-specific manner. In this work, we\u0000propose a faster version of Joint MAPLE which retains the mapping performance\u0000of the original version. Coil compression, random slice selection,\u0000parameter-specific learning rates and transfer learning are synergistically\u0000combined in the proposed framework. It speeds-up the reconstruction time up to\u0000700 times than the original version and processes a whole-brain MEMFA dataset\u0000in 21 minutes on average, which originally requires ~260 hours for Joint MAPLE.\u0000The mapping performance of the proposed framework is ~2-fold better than the\u0000standard and the state-of-the-art evaluated reconstruction techniques on\u0000average in terms of the root mean squared error.","PeriodicalId":501266,"journal":{"name":"arXiv - QuanBio - Quantitative Methods","volume":"115 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-08-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141935361","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Antonio Martinez-SanchezUniversity of Murcia, Spain, Ulrike HombergThermo Fisher Scientific, José María AlmiraUniversity of Murcia, Spain, Harold PhelippeauThermo Fisher Scientific
Object detection is a main task in computer vision. Template matching is the reference method for detecting objects with arbitrary templates. However, template matching computational complexity depends on the rotation accuracy, being a limiting factor for large 3D images (tomograms). Here, we implement a new algorithm called tensorial template matching, based on a mathematical framework that represents all rotations of a template with a tensor field. Contrary to standard template matching, the computational complexity of the presented algorithm is independent of the rotation accuracy. Using both, synthetic and real data from tomography, we demonstrate that tensorial template matching is much faster than template matching and has the potential to improve its accuracy
{"title":"Tensorial template matching for fast cross-correlation with rotations and its application for tomography","authors":"Antonio Martinez-SanchezUniversity of Murcia, Spain, Ulrike HombergThermo Fisher Scientific, José María AlmiraUniversity of Murcia, Spain, Harold PhelippeauThermo Fisher Scientific","doi":"arxiv-2408.02398","DOIUrl":"https://doi.org/arxiv-2408.02398","url":null,"abstract":"Object detection is a main task in computer vision. Template matching is the\u0000reference method for detecting objects with arbitrary templates. However,\u0000template matching computational complexity depends on the rotation accuracy,\u0000being a limiting factor for large 3D images (tomograms). Here, we implement a\u0000new algorithm called tensorial template matching, based on a mathematical\u0000framework that represents all rotations of a template with a tensor field.\u0000Contrary to standard template matching, the computational complexity of the\u0000presented algorithm is independent of the rotation accuracy. Using both,\u0000synthetic and real data from tomography, we demonstrate that tensorial template\u0000matching is much faster than template matching and has the potential to improve\u0000its accuracy","PeriodicalId":501266,"journal":{"name":"arXiv - QuanBio - Quantitative Methods","volume":"27 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-08-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141935349","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
EM Wolkovich, T Jonathan Davies, William D Pearse, Michael Betancourt
Growing anthropogenic pressures have increased the need for robust predictive models. Meeting this demand requires approaches that can handle bigger data to yield forecasts that capture the variability and underlying uncertainty of ecological systems. Bayesian models are especially adept at this and are growing in use in ecology. Yet many ecologists today are not trained to take advantage of the bigger ecological data needed to generate more flexible robust models. Here we describe a broadly generalizable workflow for statistical analyses and show how it can enhance training in ecology. Building on the increasingly computational toolkit of many ecologists, this approach leverages simulation to integrate model building and testing for empirical data more fully with ecological theory. In turn this workflow can fit models that are more robust and well-suited to provide new ecological insights -- allowing us to refine where to put resources for better estimates, better models, and better forecasts.
{"title":"A four-step Bayesian workflow for improving ecological science","authors":"EM Wolkovich, T Jonathan Davies, William D Pearse, Michael Betancourt","doi":"arxiv-2408.02603","DOIUrl":"https://doi.org/arxiv-2408.02603","url":null,"abstract":"Growing anthropogenic pressures have increased the need for robust predictive\u0000models. Meeting this demand requires approaches that can handle bigger data to\u0000yield forecasts that capture the variability and underlying uncertainty of\u0000ecological systems. Bayesian models are especially adept at this and are\u0000growing in use in ecology. Yet many ecologists today are not trained to take\u0000advantage of the bigger ecological data needed to generate more flexible robust\u0000models. Here we describe a broadly generalizable workflow for statistical\u0000analyses and show how it can enhance training in ecology. Building on the\u0000increasingly computational toolkit of many ecologists, this approach leverages\u0000simulation to integrate model building and testing for empirical data more\u0000fully with ecological theory. In turn this workflow can fit models that are\u0000more robust and well-suited to provide new ecological insights -- allowing us\u0000to refine where to put resources for better estimates, better models, and\u0000better forecasts.","PeriodicalId":501266,"journal":{"name":"arXiv - QuanBio - Quantitative Methods","volume":"131 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-08-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141935347","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：