Щоб переглянути інші типи публікацій з цієї теми, перейдіть за посиланням: Synergistic regularization.
Статті в журналах з теми "Synergistic regularization"
Оформте джерело за APA, MLA, Chicago, Harvard та іншими стилями
Ознайомтеся з топ-20 статей у журналах для дослідження на тему "Synergistic regularization".
Біля кожної праці в переліку літератури доступна кнопка «Додати до бібліографії». Скористайтеся нею – і ми автоматично оформимо бібліографічне посилання на обрану працю в потрібному вам стилі цитування: APA, MLA, «Гарвард», «Чикаго», «Ванкувер» тощо.
Також ви можете завантажити повний текст наукової публікації у форматі «.pdf» та прочитати онлайн анотацію до роботи, якщо відповідні параметри наявні в метаданих.
Переглядайте статті в журналах для різних дисциплін та оформлюйте правильно вашу бібліографію.
1
Cueva, Evelyn, Alexander Meaney, Samuli Siltanen, and Matthias J. Ehrhardt. "Synergistic multi-spectral CT reconstruction with directional total variation." Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences 379, no. 2204 (July 5, 2021): 20200198. http://dx.doi.org/10.1098/rsta.2020.0198.
Повний текст джерелаАнотація:
This work considers synergistic multi-spectral CT reconstruction where information from all available energy channels is combined to improve the reconstruction of each individual channel. We propose to fuse these available data (represented by a single sinogram) to obtain a polyenergetic image which keeps structural information shared by the energy channels with increased signal-to-noise ratio. This new image is used as prior information during a channel-by-channel minimization process through the directional total variation. We analyse the use of directional total variation within variational regularization and iterative regularization. Our numerical results on simulated and experimental data show improvements in terms of image quality and in computational speed. This article is part of the theme issue ‘Synergistic tomographic image reconstruction: part 2’.
2
Mehranian, Abolfazl, Martin A. Belzunce, Claudia Prieto, Alexander Hammers, and Andrew J. Reader. "Synergistic PET and SENSE MR Image Reconstruction Using Joint Sparsity Regularization." IEEE Transactions on Medical Imaging 37, no. 1 (January 2018): 20–34. http://dx.doi.org/10.1109/tmi.2017.2691044.
Повний текст джерела
3
Perelli, Alessandro, and Martin S. Andersen. "Regularization by denoising sub-sampled Newton method for spectral CT multi-material decomposition." Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences 379, no. 2200 (May 10, 2021): 20200191. http://dx.doi.org/10.1098/rsta.2020.0191.
Повний текст джерелаАнотація:
Spectral Computed Tomography (CT) is an emerging technology that enables us to estimate the concentration of basis materials within a scanned object by exploiting different photon energy spectra. In this work, we aim at efficiently solving a model-based maximum-a-posterior problem to reconstruct multi-materials images with application to spectral CT. In particular, we propose to solve a regularized optimization problem based on a plug-in image-denoising function using a randomized second order method. By approximating the Newton step using a sketching of the Hessian of the likelihood function, it is possible to reduce the complexity while retaining the complex prior structure given by the data-driven regularizer. We exploit a non-uniform block sub-sampling of the Hessian with inexact but efficient conjugate gradient updates that require only Jacobian-vector products for denoising term. Finally, we show numerical and experimental results for spectral CT materials decomposition. This article is part of the theme issue ‘Synergistic tomographic image reconstruction: part 1’.
4
Jørgensen, J. S., E. Ametova, G. Burca, G. Fardell, E. Papoutsellis, E. Pasca, K. Thielemans, et al. "Core Imaging Library - Part I: a versatile Python framework for tomographic imaging." Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences 379, no. 2204 (July 5, 2021): 20200192. http://dx.doi.org/10.1098/rsta.2020.0192.
Повний текст джерелаАнотація:
We present the Core Imaging Library (CIL), an open-source Python framework for tomographic imaging with particular emphasis on reconstruction of challenging datasets. Conventional filtered back-projection reconstruction tends to be insufficient for highly noisy, incomplete, non-standard or multi-channel data arising for example in dynamic, spectral and in situ tomography. CIL provides an extensive modular optimization framework for prototyping reconstruction methods including sparsity and total variation regularization, as well as tools for loading, preprocessing and visualizing tomographic data. The capabilities of CIL are demonstrated on a synchrotron example dataset and three challenging cases spanning golden-ratio neutron tomography, cone-beam X-ray laminography and positron emission tomography. This article is part of the theme issue ‘Synergistic tomographic image reconstruction: part 2’.
5
Bahadur, Rabya, Saeed ur Rehman, Ghulam Rasool, and Muhammad AU Khan. "Synergy Estimation Method for Simultaneous Activation of Multiple DOFs Using Surface EMG Signals." NUST Journal of Engineering Sciences 14, no. 2 (January 31, 2022): 66–73. http://dx.doi.org/10.24949/njes.v14i2.661.
Повний текст джерелаАнотація:
Surface electromyography signals are routinely used for designing prosthetic control systems. The concept of synergy estimation for muscle control interpretation is being explored extensively. Synergies estimated for a single active degree of freedom (DoF) are found to be uncorrelated and provide better results when used for single movement classification; however, an increase of simultaneously active DoFs leads to complex limb movements and multiple DoF detection becomes a challenge. Synergy estimation is a non-convex optimization technique, to provide better estimation this paper proposes the use of regularized non-negative matrix factorization for the evaluation of synergistic weights in complex movements. The use of regularization constraint makes the overall problem bounded and provide smoothness. The proposed technique showed better accuracy when tested for activation of multiple DoF simultaneously at a significantly lower computational time, i.e., by 34%.
6
Zhong, Lihua, Tong Ye, Yuyao Yang, Feng Pan, Lei Feng, Shuzhe Qi, and Yuping Huang. "Deep Reinforcement Learning-Based Joint Low-Carbon Optimization for User-Side Shared Energy Storage–Distribution Networks." Processes 12, no. 9 (August 23, 2024): 1791. http://dx.doi.org/10.3390/pr12091791.
Повний текст джерелаАнотація:
As global energy demand rises and climate change poses an increasing threat, the development of sustainable, low-carbon energy solutions has become imperative. This study focuses on optimizing shared energy storage (SES) and distribution networks (DNs) using deep reinforcement learning (DRL) techniques to enhance operation and decision-making capability. An innovative dynamic carbon intensity calculation method is proposed, which more accurately calculates indirect carbon emissions of the power system through network topology in both spatial and temporal dimensions, thereby refining carbon responsibility allocation on the user side. Additionally, we integrate user-side SES and ladder-type carbon emission pricing into DN to create a low-carbon economic dispatch model. By framing the problem as a Markov decision process (MDP), we employ the DRL, specifically the deep deterministic policy gradient (DDPG) algorithm, enhanced with prioritized experience replay (PER) and orthogonal regularization (OR), to achieve both economic efficiency and environmental sustainability. The simulation results indicate that this method significantly reduces the operating costs and carbon emissions of DN. This study offers an innovative perspective on the synergistic optimization of SES with DN and provides a practical methodology for low-carbon economic dispatch in power systems.
7
Du, Lehui, Baolin Qu, Fang Liu, Na Ma, Shouping Xu, Wei Yu, Xiangkun Dai, and Xiang Huang. "Precise prediction of the radiation pneumonitis with RPI: An explorative preliminary mathematical model using genotype information." Journal of Clinical Oncology 37, no. 15_suppl (May 20, 2019): e14569-e14569. http://dx.doi.org/10.1200/jco.2019.37.15_suppl.e14569.
Повний текст джерелаАнотація:
e14569 Background: Radiation pneumonitis (RP) is the most significant dose-limiting toxicity and is one major obstacle for the radiotherapy of lung cancer. Reliable predictive factors or methods are strongly demanded by radiation oncologists. The purpose of this study is by determining the effectiveness of both genetic and non-genetic factors on their impact on the development of RP, to develop a clinically practicable approach for the risk assessment of RP. Methods: One hundred eighteen lung cancer patients who received radiotherapy were enrolled. RP events were prospectively scored using the National Cancer Institute Common Terminology Criteria for Adverse Events version 4.0 (CTCAE4.0). Seven hundred thousand single-nucleotide polymorphism (SNP) sites of each patient were tested via Generalized Linear Models via Lasso and Elastic-Net Regularization (GLMNET) to determine their synergistic effects on the risk of grade≥2 RP prediction. Non-genetic factors including patient characteristics and dosimetric parameters were separately assessed by statistic test for their association with the risk of grade ≥2 RP. Based on the results of the aforementioned analysis, a multiple linear regression model named Radiation Pneumonitis Index (RPI) was built, for the assessment of grade ≥2 RP risk. Results: No statistically significant association were found between the RP risk (grade ≥2) and any of the non-genetic factors. Twenty five effective SNPs for predicting the grade≥2 RP risk were discovered and their coefficients of the synergistic effect were determined. An RPI score defined only by the information about these 25 SNPs can successfully distinguish the grade ≥2 RP population with 100% specificity and 97.8% accuracy. Conclusions: Non-genetic factors including important dosimetric parameters may not play dominant roles in the development of RP. Genotype information alone can effectively predict the risk of grade ≥2 RP. The combination of genetics and mathematical algorithms can be a new direction for radiotherapy in the field of precision medicine.
8
Di Sciacca, G., L. Di Sieno, A. Farina, P. Lanka, E. Venturini, P. Panizza, A. Dalla Mora, A. Pifferi, P. Taroni, and S. R. Arridge. "Enhanced diffuse optical tomographic reconstruction using concurrent ultrasound information." Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences 379, no. 2204 (July 5, 2021): 20200195. http://dx.doi.org/10.1098/rsta.2020.0195.
Повний текст джерелаАнотація:
Multimodal imaging is an active branch of research as it has the potential to improve common medical imaging techniques. Diffuse optical tomography (DOT) is an example of a low resolution, functional imaging modality that typically has very low resolution due to the ill-posedness of its underlying inverse problem. Combining the functional information of DOT with a high resolution structural imaging modality has been studied widely. In particular, the combination of DOT with ultrasound (US) could serve as a useful tool for clinicians for the formulation of accurate diagnosis of breast lesions. In this paper, we propose a novel method for US-guided DOT reconstruction using a portable time-domain measurement system. B-mode US imaging is used to retrieve morphological information on the probed tissues by means of a semi-automatical segmentation procedure based on active contour fitting. A two-dimensional to three-dimensional extrapolation procedure, based on the concept of distance transform, is then applied to generate a three-dimensional edge-weighting prior for the regularization of DOT. The reconstruction procedure has been tested on experimental data obtained on specifically designed dual-modality silicon phantoms. Results show a substantial quantification improvement upon the application of the implemented technique. This article is part of the theme issue ‘Synergistic tomographic image reconstruction: part 2’.
9
Lu, Yifan, Ziqi Zhang, Chunfeng Yuan, Peng Li, Yan Wang, Bing Li, and Weiming Hu. "Set Prediction Guided by Semantic Concepts for Diverse Video Captioning." Proceedings of the AAAI Conference on Artificial Intelligence 38, no. 4 (March 24, 2024): 3909–17. http://dx.doi.org/10.1609/aaai.v38i4.28183.
Повний текст джерелаАнотація:
Diverse video captioning aims to generate a set of sentences to describe the given video in various aspects. Mainstream methods are trained with independent pairs of a video and a caption from its ground-truth set without exploiting the intra-set relationship, resulting in low diversity of generated captions. Different from them, we formulate diverse captioning into a semantic-concept-guided set prediction (SCG-SP) problem by fitting the predicted caption set to the ground-truth set, where the set-level relationship is fully captured. Specifically, our set prediction consists of two synergistic tasks, i.e., caption generation and an auxiliary task of concept combination prediction providing extra semantic supervision. Each caption in the set is attached to a concept combination indicating the primary semantic content of the caption and facilitating element alignment in set prediction. Furthermore, we apply a diversity regularization term on concepts to encourage the model to generate semantically diverse captions with various concept combinations. These two tasks share multiple semantics-specific encodings as input, which are obtained by iterative interaction between visual features and conceptual queries. The correspondence between the generated captions and specific concept combinations further guarantees the interpretability of our model. Extensive experiments on benchmark datasets show that the proposed SCG-SP achieves state-of-the-art (SOTA) performance under both relevance and diversity metrics.
10
Anacleto, Adilson, Karina Beatriz dos Santos Ferreira da Rocha, Raíssa Leal Calliari, Maike dos Santos, and Sandro Deretti. "Production Arrangement of Cachaça: Comparative Study Between Morretes in the Paraná Coast and Luiz Alves in Itajaí Valley - Santa Catarina." Revista de Gestão Social e Ambiental 18, no. 2 (June 26, 2024): e07510. http://dx.doi.org/10.24857/rgsa.v18n2-158.
Повний текст джерелаАнотація:
Objective: The research sought to promote a characterization of the cachaça LPA in Morretes, Paraná in comparison to an LPA in another Brazilian region, whose production is similar and is classified as one of the most developed in Brazil. Theoretical Framework: Studies about Local Productive Arrangements have been relevant to proposals and regional development, so the basis of the research was inserted in the introductory phase. Method: Between May 2022 and February 2023, an exploratory descriptive study was carried out with leaders of two cachaça production arrangements located in the south of Brazil, one consolidated in Santa Catarina and the other one under development in Paraná. Results and Discussion: The results obtained revealed that despite the similar number of producers, the Morretes LPA has a production capacity equivalent to 35% of what is produced in the Luiz Alves LPA. The difference in productivity is not linked to factors such as proximity to the consumer market, distribution logistics, labor or even geoclimatic characteristics. The main factors limiting the easier development of LPAs were the degree of difficulty in regularizing production, the lack of technical assistance and the almost non-existence of public policies, with the Morretes LPA on the coast of Paraná being most impacted. It is concluded that the insertion of tourism associated with beverage consumption, investment in marketing actions and public policies that facilitate the regularization of informal producers can strengthen the Morretes LPA and promote its consolidation as an economic activity that generates regional development. Research Implications: The Morretes LPA has all the favorable and synergistic conditions for the development of the cachaça LPA, and in this context the offer of new services and products arising from production and trade activities can generate a set of entertainment offered as a new experience for tourists, in addition from visits to stills and the creation of parties and fairs with the drink theme. Originality/Value: Comparative studies between local productive arrangements are still little explored in scientific literature and may guide regional development in a more organized way.
11
Chakravarti, Sayak, Suman Mazumder, Harish Kumar, Neeraj Sharma, Ujjal Kumar Mukherjee, Shaji Kumar, Linda B. Baughn, Brian G. Van Ness, and Amit Kumar Mitra. "Establishing a Novel Pipeline That Combines in-Silico Prediction with in-Vitro and Ex-Vivo Validation to Discover Secondary Drug Combinations Against Relapsed and/or Refractory Multiple Myeloma." Blood 138, Supplement 1 (November 5, 2021): 1615. http://dx.doi.org/10.1182/blood-2021-154521.
Повний текст джерелаАнотація:
Abstract Multiple myeloma (MM) is the second-most common hematological malignancy in the US. MM is an incurable, age-dependent plasma cell neoplasm with a 5-year survival rate of less than 50%. Extensive inter-individual variation in response to standard-of-care drugs like proteasome inhibitors (PIs) and immunomodulatory drugs (IMiDs), drug resistance, and dose-limiting toxicities are critical problems for the treatment of MM. Clinical success in anti-myeloma treatment, therefore, warrants continuous development of novel combination therapy strategies with the explicit goal to improve the therapeutic efficacy by concomitantly targeting multiple signaling pathways. Previously, we have reported the development of an in-house computational pipeline called secDrug that applies greedy algorithm-based set-covering computational optimization method followed by a regularization technique to predict secondary drugs that can be repurposed as novel synergistic partners of standard-of-care drugs for the management of refractory/ resistant MM. Top among these secondary drugs (secDrugs) were the HSP90 inhibitor 17-AAG. In this study, we used 17-AAG as a proof of principle to establish a pipeline that integrates our in silico predictions with in vitro and ex vivo validation as well as multi-omics technologies to identify, validate, and characterize therapeutic agents that could be used either alone or in combination with standard-of-care drugs for the treatment of R/R MM patients (Figure 1). To screen and validate our in silico prediction results, we performed in vitro cytotoxicity assays using 17-AAG on a panel of human myeloma cell lines (HMCLs; in vitro model systems) that captures a wide range of biological and genetic heterogeneity representing the complexities encountered in clinical settings. These cell lines include HMCLs representing innate sensitive/resistance, >10 pairs of parental and clonally-derived PI- and IMiD-resistant pairs (P vs VR or LenR; representing acquired/emerging resistance/relapse), NRAS mutants which leads to the constitutive activation of oncogenic Ras signaling, and CRISPR-edited HSP90 knockdown cell line. Our results showed that 17-AAG has high synergistic activity in combination with PI in inducing apoptosis even in innate and acquired PI-resistant HMCLs and significantly reduces the effective dose of PI required to achieve IC 50 (Chou-Talalay's Dose Reduction Index or DRI 7±1.4). Moreover, 17-AAG+IMiD showed synergistic cell killing activity in clonally-derived IMiD resistant HMCL. Further, 17-AAG induced cell death was comparable with Hsp90 knockdown as evident from the cytotoxicity assay using PI and 17-AAG in combination in RPMI8226-wild type and RPMI-HSP90AA1 knocked down cell line. Notably, 17-AAG was strikingly effective against the NRAS-mutant cell line indicating an additional niche (NRas mutant myeloma) where 17-AAG could be most effective. Next, we performed RNA sequencing to elucidate the molecular mechanisms behind 17-AAG drug action, drug synergy, 17-AAG-induced cell death. Our gene expression profiling (GEP) followed by Ingenuity Pathway Analysis (IPA) analysis revealed protein ubiquitination, aryl hydrocarbon receptor signalling pathway as the top canonical pathways. 17-AAG induced apoptosis via mitochondrial mediated pathway in myeloma. 17-AAG exerts its cytotoxic effect by activating intrinsic pathway of apoptosis which we further confirmed through the increase in reactive oxygen species generation and decrease in mitochondrial membrane potential. 17-AAG was also effective in reducing the expression of hallmarks of MM such as p65/NF-kB, IRF4, c-Myc. Finally, we performed mass cytometry (CyTOF; Cytometry by time of flight) on primary bone-marrow cells (PMCs) from myeloma patients for further validation of proteomic signatures at the single-cell level. CyTOF analysis confirmed 17-AAG-induced cell death and key changes in MM-specific proteomic markers. 17-AAG treated PMCs showed elevated cleaved caspase levels and down-regulation of IRF4 and phospho-STAT3. GEP and CyTOF results were confirmed using immunoblotting assays. Together, our study demonstrates a unique pipeline for drug repositioning that has the potential to revolutionize clinical decision-making by minimizing the number of drugs required for discovering successful combination chemotherapy regimens against drug-resistant myeloma. Figure 1 Figure 1. Disclosures Kumar: BMS: Consultancy, Research Funding; Abbvie: Consultancy, Membership on an entity's Board of Directors or advisory committees, Research Funding; Amgen: Consultancy, Research Funding; Takeda: Consultancy, Membership on an entity's Board of Directors or advisory committees, Research Funding; Tenebio: Research Funding; Beigene: Consultancy; Oncopeptides: Consultancy; Antengene: Consultancy, Honoraria; Carsgen: Research Funding; Janssen: Consultancy, Membership on an entity's Board of Directors or advisory committees, Research Funding; KITE: Consultancy, Membership on an entity's Board of Directors or advisory committees, Research Funding; Merck: Research Funding; Astra-Zeneca: Consultancy, Membership on an entity's Board of Directors or advisory committees, Research Funding; Novartis: Research Funding; Roche-Genentech: Consultancy, Research Funding; Bluebird Bio: Consultancy; Celgene: Membership on an entity's Board of Directors or advisory committees, Research Funding; Adaptive: Membership on an entity's Board of Directors or advisory committees, Research Funding; Sanofi: Research Funding.
12
Chakravarti, Sayak, Ujjal Kumar Mukherjee, Suman Mazumder, Timothy Moore, and Amit Kumar Mitra. "In silico Prediction Followed By I n Vitro validation Identifies a Survivin Inhibitor and an MCL-1 Inhibitor As a Potent Secondary Drug Against Refractory or Relapsed Mantle Cell Lymphoma." Blood 138, Supplement 1 (November 5, 2021): 1191. http://dx.doi.org/10.1182/blood-2021-154479.
Повний текст джерелаАнотація:
Abstract Mantle cell lymphoma (MCL) is an aggressive lymphoid neoplasm that develops from malignant B-lymphocytes in the outer edge or mantle zone of a lymph node. This is a sub-type of B-cell non-Hodgkin lymphoma characterized by rapid clinical progression and poor response rate to conventional chemotherapeutic drugs with recurrent relapse resulting in a short estimated 5-year overall survival (OS) of 2-5 years depending on the clinical risk. Combination therapies such as R-CHOP, R-DHAP, Hyper-CVAD, VcR-CAP constitute the front-line chemotherapeutic treatment landscape for MCL. Despite good initial response to the combination regimens, all patients develop resistance over time. The Bruton's tyrosine kinase inhibitor (BTKi) Ibrutinib and the proteasome inhibitor (PI) Bortezomib are FDA-approved therapies for refractory or relapsed (R/R) MCL with demonstrated high initial response rate in clinical trials. However, highly variable treatment response along with dose-limiting toxicities has limited the efficacy in real-world settings with the median progression-free survival (PFS) of <15 months and Over-al of 1-2 years. Thus, the identification of novel drugs that function either alone or as combination to curb the oncogenic progression as well as to reduce drug-associated toxicities is of high clinical significance. We have designed a novel optimization-regularization-based computational prediction algorithm called "secDrug" that uses large-scale pharmacogenomics databases like the GDSC1000 to identify novel secondary drugs for the management of treatment-resistant B-cell malignancies. We hypothesize that combination of our predicted secDrugs with BTKi/ PI will be useful in curbing oncogenic progressions of R/R MCL and abrogate drug resistance through simultaneous inhibition of multiple oncogenic factors/pathways. When applied to BTKi/PI-resistant R/R MCL, the top predicted secondary drugs (secDrugs) were YM155 (Survivin inhibitor) and S63845 (selective MCL-1 inhibitor). Interestingly, both Survivin and MCL-1 are reported to be over-expressed in MCL, and their expression is strongly correlated with the oncogenic progression and survivability of the patients. To validate our in-silico predictions, we performed in vitro cytotoxicity assays with the top predicted secDrugs (YM155 and S63845) as single agents (IC50 for YM155 4.87±0.66 nM, for S63845 0.9±1.1 uM) as well as in combination with BTKi/PI against a panel of MCL cell lines representing PI/BTKi sensitive, innate resistant (representing refractory MCL) and clonally-derived acquired resistant (representing relapsed MCL). Our results showed that the YM155 and S63845 exhibited significant synergistic cell killing activities (Combination index/ CI value of 0.31±0.49 as calculated using Chou-Talalay's CI theorem, C.I>1 depicts synergism) alone and in combination with Bortezomib (PI) and Ibrutinib (BTKi), especially in R/R MCL cell lines. Further, our results also showed that both YM155 and S63845 in combination with BTKi/PI were able to significantly lower the effective dose of both BTKi/PI required to achieve desired therapeutic response by >12 times (Dose Reduction Index or DRI for YM155 in combination is 15.87±4.93; DRI for S63845 in combination is 12.34±2.67), thereby making the cell lines relatively more BTKi/PI sensitive. Next, we performed next-generation RNA sequencing analysis to identify mechanisms of secDrug action and synergy. Our Gene expression profiling and Ingenuity pathway analysis of the RNAseq data among YM155-treated MCL cell lines revealed eIF4-p70S6K signaling and mTOR signaling as the top canonical pathways. Our study thus identified YM155 and S63845 as potential novel candidates for repurposing as secondary drugs in combination with BTKi/PI for the treatment of R/R MCL. Currently, we are exploring the probable subclonal molecular mechanisms governing the synergistic drug action by using single-cell transcriptomics analysis. As both YM155 and S63845 have reported activity against cancer stem-ness, we will further investigate the effect of our novel drugs on the cancer stem-like cells in MCL, which have a potential role in treatment resistance. The secDrug algorithm promises to serve as a universal prototype for the discovery of novel drug combination regimens for treatment outcomes in any cancer type by enhancing sensitivity or overcoming resistance to standard of care drugs. Figure 1 Figure 1. Disclosures No relevant conflicts of interest to declare.
13
Tsai, Yen-Hsi Richard, and Stanley Osher. "Total variation and level set methods in image science." Acta Numerica 14 (April 19, 2005): 509–73. http://dx.doi.org/10.1017/s0962492904000273.
Повний текст джерелаАнотація:
We review level set methods and the related techniques that are common in many PDE-based image models. Many of these techniques involve minimizing the total variation of the solution and admit regularizations on the curvature of its level sets. We examine the scope of these techniques in image science, in particular in image segmentation, interpolation, and decomposition, and introduce some relevant level set techniques that are useful for this class of applications. Many of the standard problems are formulated as variational models. We observe increasing synergistic progression of new tools and ideas between the inverse problem community and the ‘imagers’. We show that image science demands multi-disciplinary knowledge and flexible, but still robust methods. That is why the level set method and total variation methods have become thriving techniques in this field.Our goal is to survey recently developed techniques in various fields of research that are relevant to diverse objectives in image science. We begin by reviewing some typical PDE-based applications in image processing. In typical PDE methods, images are assumed to be continuous functions sampled on a grid. We will show that these methods all share a common feature, which is the emphasis on processing the level lines of the underlying image. The importance of level lines has been known for some time. See, e.g., Alvarez, Guichard, Morel and Lions (1993). This feature places our slightly general definition of the level set method for image science in context. In Section 2 we describe the building blocks of a typical level set method in the continuum setting. Each important task that we need to do is formulated as the solution to certain PDEs. Then, in Section 3, we briefly describe the finite difference methods developed to construct approximate solutions to these PDEs. Some approaches to interpolation into small subdomains of an image are reviewed in Section 4. In Section 5 we describe the Chan–Vese segmentation algorithm and two new fast implementation methods. Finally, in Section 6, we describe some new techniques developed in the level set community.
14
Sun Yan-Xu, Huang Juan, Gao Wei, Chang Jia-Feng, Zhang Wei, Shi Chang, and Li Yun-He. "Tomography of fast ion distribution function under neutral beam injection and ion cyclotron resonance heating on EAST." Acta Physica Sinica, 2023, 0. http://dx.doi.org/10.7498/aps.72.20230846.
Повний текст джерелаАнотація:
In magnetic confinement fusion devices, velocity-space tomography of fast-ion velocity distribution functions is crucial for investigating fast-ion distribution and transport. In the neutral beam injection (NBI) and ion cyclotron resonance heating (ICRF) synergistic heating experiments in Experimental Advanced Superconducting Tokamak (EAST), high-energy particles with energy exceeding NBI was observed. Simulations of synergistic effect on fast-ion velocity distribution functions given by TRANSP also showed the existence of particle with energy higher than NBI. To investigate the behavior of fast ion distribution and calculate the velocity distribution function under different heating conditions, the first-order Tikhonov regularization tomographic inversion method with higher inversion accuracy by comparing various regularization techniques was introduced. The limitations of the dual-view fast-ion D-alpha (FIDA) diagnostic measurements in velocity space was addressed by incorporating prior information such as null measurement and the known peaks and effectively mitigated the occurrence of artifacts. This method is firstly employed in the case of NBI heating. The NBI peaks was successfully reconstructed at the expected location in velocity space, which shows significant improvement in the inversion results. In order to further validate the synergistic effect of NBI-ICRF heating and study the mechanism of fast ion distribution under synergistic heating, the combination of FIDA and neutron emission spectrometers (NES) was employed to the first-order Tikhonov regularization tomographic inversion method for enhancing the coverage of velocity space, through which the issue of artifacts in the inversion results is significantly improved, and thus the precision of the obtained fast-ion velocity distribution functions is enhanced. Based on the benefit described above, the method of combining NES and FIDA diagnostics was used to calculate fast-ion velocity distribution functions in the NBI and ICRF synergistic heating discharge. The synergistic heating effect is manifested in the fast-ion velocity distribution. The availability of this inversion method in reconstructing fast-ion velocity distribution functions during high-performance operations of NBI-ICRF synergistic heating in the EAST experiment was confirmed. In the nextstep EAST research, high performance discharge will demand more efficiency NBI and ICRF synergistic heating, the work in this paper set the stage for investigating the underlying mechanisms of synergistic heating and the intricate behaviors associated with fast ion distribution and transport.
15
Lv, Ji, Guixia Liu, Yuan Ju, Ying Sun, and Weiying Guo. "Prediction of Synergistic Antibiotic Combinations by Graph Learning." Frontiers in Pharmacology 13 (March 8, 2022). http://dx.doi.org/10.3389/fphar.2022.849006.
Повний текст джерелаАнотація:
Antibiotic resistance is a major public health concern. Antibiotic combinations, offering better efficacy at lower doses, are a useful way to handle this problem. However, it is difficult for us to find effective antibiotic combinations in the vast chemical space. Herein, we propose a graph learning framework to predict synergistic antibiotic combinations. In this model, a network proximity method combined with network propagation was used to quantify the relationships of drug pairs, and we found that synergistic antibiotic combinations tend to have smaller network proximity. Therefore, network proximity can be used for building an affinity matrix. Subsequently, the affinity matrix was fed into a graph regularization model to predict potential synergistic antibiotic combinations. Compared with existing methods, our model shows a better performance in the prediction of synergistic antibiotic combinations and interpretability.
16
Liu, Huan, Xinglin Zhang, Congyu Liao, Haobin Dong, Zheng Liu, and Xiangyun Hu. "Synergistic Hankel structured low-rank approximation with total variation regularization for complex magnetic anomaly detection." IEEE Transactions on Instrumentation and Measurement, 2023, 1. http://dx.doi.org/10.1109/tim.2023.3264036.
Повний текст джерела
17
Qin, Limu, Gang Yang, and Wen He. "Generalized Shannon entropy sparse wavelet packet transform for fault detection of traction motor bearings in high-speed trains." Structural Health Monitoring, May 9, 2024. http://dx.doi.org/10.1177/14759217241245320.
Повний текст джерелаАнотація:
An effective structural health monitoring method of traction motor bearings is a powerful guarantee for the safety operation of high-speed trains. However, it is exceptionally difficult to detect bearing fault characteristics from the vibration signals of traction motor bearings operating at high rotational speeds. In this scenario, a generalized Shannon entropy sparse wavelet packet transform (GSWPT) for fault detection of motor bearings is proposed in this paper. Firstly, a generalized Shannon entropy sparse regularization method is proposed to obtain sparse wavelet reconstruction coefficients by extending the definition of the Shannon information entropy, and the non-convex sparse regularization function is minimized by synergistic swarm optimization algorithm. Then, the wavelet node coefficients are weighted according to the second-order cyclostationarity index of the wavelet packet node to further enhance the sparsity of the reconstructed signal. Moreover, the optimal decomposition level of GSWPT is adaptively selected by the maximum sparsity and cyclostationarity criterion. Particularly, in order to verify the bearing fault detection performance of GSWPT in practical engineering, a bearing fault dynamic model of traction motor in high-speed train was established based on Hertz contact theory and the fourth-order Runge-Kutta method to obtain simulated data under strong Gaussian white noise, and a corresponding test platform was constructed to collect experimental data under different operating conditions. Finally, the applications on the simulated and experimental signals of traction motor bearings in high-speed trains demonstrate that GSWPT significantly outperforms the conventional wavelet packet transform, dual-tree complex wavelet packet transform, blind deconvolution, modal decomposition, and Infogram methods to some extent for fault detection.
18
Yang, Huili, KyungPyo Hong, Justin J. Baraboo, Lexiaozi Fan, Andrine Larsen, Michael Markl, Joshua D. Robinson, Cynthia K. Rigsby, and Daniel Kim. "GRASP reconstruction amplified with view‐sharing and KWIC filtering reduces underestimation of peak velocity in highly‐accelerated real‐time phase‐contrast MRI: A preliminary evaluation in pediatric patients with congenital heart disease." Magnetic Resonance in Medicine, December 12, 2023. http://dx.doi.org/10.1002/mrm.29974.
Повний текст джерелаАнотація:
AbstractPurposeTo develop a highly‐accelerated, real‐time phase contrast (rtPC) MRI pulse sequence with 40 fps frame rate (25 ms effective temporal resolution).MethodsHighly‐accelerated golden‐angle radial sparse parallel (GRASP) with over regularization may result in temporal blurring, which in turn causes underestimation of peak velocity. Thus, we amplified GRASP performance by synergistically combining view‐sharing (VS) and k‐space weighted image contrast (KWIC) filtering. In 17 pediatric patients with congenital heart disease (CHD), the conventional GRASP and the proposed GRASP amplified by VS and KWIC (or GRASP + VS + KWIC) reconstruction for rtPC MRI were compared with respect to clinical standard PC MRI in measuring hemodynamic parameters (peak velocity, forward volume, backward volume, regurgitant fraction) at four locations (aortic valve, pulmonary valve, left and right pulmonary arteries).ResultsThe proposed reconstruction method (GRASP + VS + KWIC) achieved better effective spatial resolution (i.e., image sharpness) compared with conventional GRASP, ultimately reducing the underestimation of peak velocity from 17.4% to 6.4%. The hemodynamic metrics (peak velocity, volumes) were not significantly (p > 0.99) different between GRASP + VS + KWIC and clinical PC, whereas peak velocity was significantly (p < 0.007) lower for conventional GRASP. RtPC with GRASP + VS + KWIC also showed the ability to assess beat‐to‐beat variation and detect the highest peak among peaks.ConclusionThe synergistic combination of GRASP, VS, and KWIC achieves 25 ms effective temporal resolution (40 fps frame rate), while minimizing the underestimation of peak velocity compared with conventional GRASP.
19
Sivakumar, Nikita, Cameron Mura, and Shayn M. Peirce. "Innovations in integrating machine learning and agent-based modeling of biomedical systems." Frontiers in Systems Biology 2 (November 10, 2022). http://dx.doi.org/10.3389/fsysb.2022.959665.
Повний текст джерелаАнотація:
Agent-based modeling (ABM) is a well-established computational paradigm for simulating complex systems in terms of the interactions between individual entities that comprise the system’s population. Machine learning (ML) refers to computational approaches whereby algorithms use statistical methods to “learn” from data on their own, i.e., without imposing any a priori model/theory onto a system or its behavior. Biological systems—ranging from molecules, to cells, to entire organisms, to whole populations and even ecosystems—consist of vast numbers of discrete entities, governed by complex webs of interactions that span various spatiotemporal scales and exhibit nonlinearity, stochasticity, and variable degrees of coupling between entities. For these reasons, the macroscopic properties and collective dynamics of biological systems are generally difficult to accurately model or predict via continuum modeling techniques and mean-field formalisms. ABM takes a “bottom-up” approach that obviates common difficulties of other modeling approaches by enabling one to relatively easily create (or at least propose, for testing) a set of well-defined “rules” to be applied to the individual entities (agents) in a system. Quantitatively evaluating a system and propagating its state over a series of discrete time-steps effectively simulates the system, allowing various observables to be computed and the system’s properties to be analyzed. Because the rules that govern an ABM can be difficult to abstract and formulate from experimental data, at least in an unbiased way, there is a uniquely synergistic opportunity to employ ML to help infer optimal, system-specific ABM rules. Once such rule-sets are devised, running ABM calculations can generate a wealth of data, and ML can be applied in that context too—for example, to generate statistical measures that accurately and meaningfully describe the stochastic outputs of a system and its properties. As an example of synergy in the other direction (from ABM to ML), ABM simulations can generate plausible (realistic) datasets for training ML algorithms (e.g., for regularization, to mitigate overfitting). In these ways, one can envision a variety of synergistic ABM⇄ML loops. After introducing some basic ideas about ABMs and ML, and their limitations, this Review describes examples of how ABM and ML have been integrated in diverse contexts, spanning spatial scales that include multicellular and tissue-scale biology to human population-level epidemiology. In so doing, we have used published studies as a guide to identify ML approaches that are well-suited to particular types of ABM applications, based on the scale of the biological system and the properties of the available data.
20
Patra, Aswini Kumar, and Lingaraj Sahoo. "Explainable light-weight deep learning pipeline for improved drought stress identification." Frontiers in Plant Science 15 (November 28, 2024). http://dx.doi.org/10.3389/fpls.2024.1476130.
Повний текст джерелаАнотація:
IntroductionEarly identification of drought stress in crops is vital for implementing effective mitigation measures and reducing yield loss. Non-invasive imaging techniques hold immense potential by capturing subtle physiological changes in plants under water deficit. Sensor-based imaging data serves as a rich source of information for machine learning and deep learning algorithms, facilitating further analysis that aims to identify drought stress. While these approaches yield favorable results, real-time field applications require algorithms specifically designed for the complexities of natural agricultural conditions.MethodsOur work proposes a novel deep learning framework for classifying drought stress in potato crops captured by unmanned aerial vehicles (UAV) in natural settings. The novelty lies in the synergistic combination of a pre-trained network with carefully designed custom layers. This architecture leverages the pre-trained network’s feature extraction capabilities while the custom layers enable targeted dimensionality reduction and enhanced regularization, ultimately leading to improved performance. A key innovation of our work is the integration of gradient-based visualization inspired by Gradient-Class Activation Mapping (Grad-CAM), an explainability technique. This visualization approach sheds light on the internal workings of the deep learning model, often regarded as a ”black box”. By revealing the model’s focus areas within the images, it enhances interpretability and fosters trust in the model’s decision-making process.Results and discussionOur proposed framework achieves superior performance, particularly with the DenseNet121 pre-trained network, reaching a precision of 97% to identify the stressed class with an overall accuracy of 91%. Comparative analysis of existing state-of-the-art object detection algorithms reveals the superiority of our approach in achieving higher precision and accuracy. Thus, our explainable deep learning framework offers a powerful approach to drought stress identification with high accuracy and actionable insights.