Article
Computer Science, Interdisciplinary Applications
Ruud J. G. van Sloun, Oren Solomon, Matthew Bruce, Zin Z. Khaing, Hessel Wijkstra, Yonina C. Eldar, Massimo Mischi
Summary: The study introduces a new method Deep-ULM, which utilizes deep learning and convolutional neural networks for super-resolution vascular imaging in dense contrast-enhanced scenarios, achieving high accuracy and fast imaging. The research demonstrates through experiments that deep learning performs well under high-density contrast agent conditions.
IEEE TRANSACTIONS ON MEDICAL IMAGING
(2021)
Review
Biochemistry & Molecular Biology
Yoonsuk Hyun, Doory Kim
Summary: With the development of super-resolution imaging techniques, understanding protein structure at the nanoscale in terms of clustering and organization in a cell is crucial. This review categorizes computational cluster analysis methods for single-molecule localization microscopy (SMLM) images into classical and machine learning-based methods, and discusses possible future directions for machine learning-based cluster analysis methods for SMLM data.
COMPUTATIONAL AND STRUCTURAL BIOTECHNOLOGY JOURNAL
(2023)
Article
Telecommunications
Sahar Imtiaz, Sebastian Schiessl, Georgios P. Koudouridis, James Gross
Summary: In this study, a coordinates-based resource allocation scheme using supervised machine learning techniques is proposed, which performs efficiently compared to the traditional approach under various propagation conditions. The performance of the coordinates-based scheme is close to the CSI-based scheme even when user coordinates are erroneous.
IEEE TRANSACTIONS ON COGNITIVE COMMUNICATIONS AND NETWORKING
(2021)
Article
Chemistry, Physical
Guannan Qian, Jin Zhang, Sheng-Qi Chu, Jizhou Li, Kai Zhang, Qingxi Yuan, Zi-Feng Ma, Piero Pianetta, Linsen Li, Keeyoung Jung, Yijin Liu
Summary: This study investigates the interplay among electronic structure, lattice configuration, and micro-morphology of nickel-rich cathode materials for lithium-ion batteries during prolonged electrochemical cycling. The research reveals mesoscale reaction heterogeneity in the battery cathode and suggests particle structure engineering as a viable approach to mitigate the chemomechanical degradation of cathode materials.
ACS ENERGY LETTERS
(2021)
Article
Telecommunications
Jun Yan, Chuanhui Ma, Bin Kang, Xiaohuan Wu, Huaping Liu
Summary: This letter presents a technique for indoor localization using channel state information and received signal strength indicators, tackling the random weight issue of ELM through AdaBoost optimization, to enhance localization performance and reduce data storage and computational needs.
IEEE COMMUNICATIONS LETTERS
(2021)
Article
Automation & Control Systems
Xueyan Oh, Ryan Lim, Shaohui Foong, U-Xuan Tan
Summary: This article proposes an infrastructure-free localization system that uses a single active pan-tilt-zoom camera to detect, track, and localize a circular LED marker, addressing the challenges of localization in GPS-denied environments without the need for on-site calibration. The approach utilizes a convolutional neural network trained on synthetic images to detect the LED marker as an ellipse, demonstrating improved robustness compared to traditional ellipse detection methods. The predicted elliptical angle is also utilized to improve marker range estimation and 3D localization. The system's performance is evaluated through real-world flight experiments and demonstrated in both indoor and outdoor environments.
IEEE-ASME TRANSACTIONS ON MECHATRONICS
(2023)
Article
Automation & Control Systems
Zhefu Wu, Xingda Pan, Kunpeng Fan, Kai Liu, Yun Xiang
Summary: This research aims to build a nonintrusive orientation detection system using WiFi infrastructure and channel state information (CSI), with a focus on exploring subcarrier correlations through a visibility graph (VG) network analysis method. The system demonstrates high accuracy and effectiveness in various applications, outperforming baseline approaches with at least 26% improvement in detection accuracy.
IEEE TRANSACTIONS ON SYSTEMS MAN CYBERNETICS-SYSTEMS
(2021)
Article
Computer Science, Information Systems
Maria Frasca, Genoveffa Tortora
Summary: The paper discusses the integration of computer science and medical research, focusing on analyzing PPMI data for Parkinson's disease. Information Retrieval and Machine Learning techniques are used to correlate data and classify patients, with a visualization system aiding in early diagnosis by supporting doctors in analyzing chronological data.
MULTIMEDIA TOOLS AND APPLICATIONS
(2022)
Article
Chemistry, Analytical
Zaynab El Mawas, Cindy Cappelle, Mohamad Daher, Maan El Badaoui El Najjar, Ikhlas Abdel-Qader
Summary: This paper proposes a fault-tolerant cooperative localization solution by coupling model-based and data-driven techniques, which effectively identifies and flags erroneous readings through the guidance of prior knowledge. Additionally, a comparative study on centralized and federated learning techniques is conducted, evaluating their performance in sensor fault detection within a multi-vehicle system.
Article
Nanoscience & Nanotechnology
Abhishek Talapatra, Udaykumar Gajera, P. Syam Prasad, Jeyaramane Arout Chelvane, Jyoti Ranjan Mohanty
Summary: In this study, advanced neural networks and image regression algorithms were used to analyze magnetic properties based on magnetic microstructure. The results showed that this method was able to accurately predict magnetic properties, which is important for understanding the changes in magnetic properties under external perturbations.
ACS APPLIED MATERIALS & INTERFACES
(2022)
Article
Computer Science, Interdisciplinary Applications
Leo Milecki, Jonathan Poree, Hatim Belgharbi, Chloe Bourquin, Rafat Damseh, Patrick Delafontaine-Martel, Frederic Lesage, Maxime Gasse, Jean Provost
Summary: A new approach using deep learning to recover dense vascular networks from high microbubble concentrations in ultrasound acquisitions was proposed. The study trained a 3D convolutional neural network on a realistic mouse brain microvascular network, and validated the approach both in silico and in vivo, showing improved precision in resolving micro vessels both in controlled and real-life environments.
IEEE TRANSACTIONS ON MEDICAL IMAGING
(2021)
Review
Biochemical Research Methods
Jun Wang, Marc Horlacher, Lixin Cheng, Ole Winther
Summary: RNA localization is important for spatial translation regulation, and this review discusses its molecular mechanisms, experimental techniques, and machine learning-based prediction tools. The three main molecular mechanisms controlling RNA localization to distinct cellular compartments, including directed transport, mRNA degradation protection, and diffusion/local entrapment, are reviewed. Advances in experimental methods provide ample data resources for the design of powerful machine learning models in RNA localization prediction. The review also covers publicly available predictive tools, serving as a guide for users and encouraging the development of more effective prediction models. Lastly, an overview of multimodal learning is presented as a potential new avenue for RNA localization prediction.
BRIEFINGS IN BIOINFORMATICS
(2023)
Article
Engineering, Industrial
Chengcheng Wang, Shubham Chandra, Sheng Huang, Shu Beng Tor, Xipeng Tan
Summary: A machine learning-based framework is developed to optimize process parameters and unravel the relationships between process, microstructure, and property (PMP). Using electron beam melting (EBM)-processed Ti-6Al-4V alloy as an example, process maps are constructed to discover the optimal process window. By using the machine learning-optimized process parameters, Ti-6Al-4V parts with higher strength and faster fabrication are achieved.
JOURNAL OF MATERIALS PROCESSING TECHNOLOGY
(2023)
Article
Chemistry, Physical
Rui Li, Aleksei Makogon, Tatiana Galochkina, Jean-Francois Lemineur, Frederic Kanoufi, Viacheslav Shkirskiy
Summary: Unsupervised machine learning algorithms are used to analyze optical microscopy images obtained dynamically and the results show distinct clusters of chemical reactivity in Al alloy particles. The analysis confirms the chemical communication of OH- fluxes within particles, and statistically significant patterns of reactivity under dynamic conditions are revealed. The results align well with numerical models of chemical communication, demonstrating the synergy between data-driven machine learning and physics-driven finite element modeling approaches.
Article
Management
Jente Van Belle, Tias Guns, Wouter Verbeke
Summary: This paper reviews empirical studies on incorporating downstream information in demand forecasting, investigating different modeling approaches and forecasting methods. The results show that integrating sell-through data into the forecasting process can effectively improve a manufacturer's short-term forecast accuracy.
EUROPEAN JOURNAL OF OPERATIONAL RESEARCH
(2021)
Article
Chemistry, Multidisciplinary
Bowen Wang, Michael Xiong, Josephine Susanto, Xue Li, Wai-Yee Leung, Ke Xu
Summary: We propose a strategy to optimize the photoswitching behavior of rhodamines for (d)STORM super-resolution microscopy. By replacing the benzene ring with a permanently charged 1,3-disubstituted imidazolium, the resulting dyes show enhanced photoswitching capability and exhibit excellent performance in (d)STORM with fast on-off switching, long-lasting blinking, and bright single-molecule emission. Our method achieves high-quality (d)STORM images under green excitation that are comparable to those obtained using ideal red-excited dyes, even for challenging structures like the mammalian actin cytoskeleton. We also demonstrate high-quality two-color three-dimensional (d)-STORM.
ANGEWANDTE CHEMIE-INTERNATIONAL EDITION
(2022)
Article
Cell Biology
Rui Yan, Kun Chen, Bowen Wang, Ke Xu
Summary: A highly elongated tubular ER-Golgi intermediate compartment (t-ERGIC) that selectively expedites the ER-to-Golgi transport for soluble cargoes of the receptor SURF4 has been discovered. The t-ERGIC has a distinctive elongated and thinned shape, and is marked by the small GTPases Rabl A/B instead of the canonical ERGIC marker ERGIC-53. It accelerates the trafficking of SURF4-bound cargoes with its large surface-to-volume ratio, high intracellular traveling speeds, and ER-Golgi recycling capabilities.
DEVELOPMENTAL CELL
(2022)
Article
Chemistry, Multidisciplinary
Yunqi Li, Bowen Wang, Wan Li, Ke Xu
Summary: This study reports that acidic solutions can induce the fast generation of solution-enclosing blisters on monolayer and few-layer graphene on silicon and glass substrates. The graphene blisters are nanoscale to micrometer-sized and are universally generated on hydrophilic surfaces. The blisters are highly dynamic and can be modulated by solution osmotic pressure.
Article
Engineering, Electrical & Electronic
Saeid Daneshgar, Hao Li, Taehwan Kim, Ganesh Balamurugan
Summary: This paper reviews the design trade-offs in CMOS inverter-based SF-TIA and presents a design example of a high-speed and energy efficient TIA using FinFET CMOS technology. The results demonstrate the potential for highly integrated, low-cost, and high-sensitivity CMOS optical receivers.
IEEE JOURNAL OF SOLID-STATE CIRCUITS
(2022)
Article
Chemistry, Multidisciplinary
Alexander A. Choi, Ha H. Park, Kun Chen, Rui Yan, Wan Li, Ke Xu
Summary: Recent studies have sparked debate over whether catalytic reactions enhance the diffusion coefficients of enzymes. Through high statistical analysis, it has been found that catalytic reactions do not affect the diffusion properties of hydrogenase, urease, aldolase, and alkaline phosphatase.
JOURNAL OF THE AMERICAN CHEMICAL SOCIETY
(2022)
Article
Multidisciplinary Sciences
Bowen Wang, Zhiheng Zhao, Michael Xiong, Rui Yan, Ke Xu
Summary: The endoplasmic reticulum (ER) is a versatile organelle with distinct forms of tubules, regulated by Rtn4 and Climp63, that together form a ribbon-like structure of well-defined width.
PROCEEDINGS OF THE NATIONAL ACADEMY OF SCIENCES OF THE UNITED STATES OF AMERICA
(2022)
Article
Multidisciplinary Sciences
Meiyan Jin, Cyna Shirazinejad, Bowen Wang, Amy Yan, Johannes Schoneberg, Srigokul Upadhyayula, Ke Xu, David G. Drubin
Summary: This study investigates the organization of actin networks at clathrin-mediated endocytosis (CME) sites and how actin assembly is regulated. The findings show that actin assembles preferentially at stalled CME sites, where it exerts an asymmetric force to pull vesicles into the cell. N-WASP and the Arp2/3 complex are recruited to one side of CME sites for stimulating asymmetric actin assembly.
NATURE COMMUNICATIONS
(2022)
Article
Multidisciplinary Sciences
Wojciech M. Smigiel, Luca Mantovanelli, Dmitrii S. Linnik, Michiel Punter, Jakob Silberberg, Limin Xiang, Ke Xu, Bert Poolman
Summary: This study analyzed the structure of the cytoplasm in Escherichia coli by performing single-molecule displacement mapping on native cytoplasmic proteins. It found that protein diffusion at the poles is consistently slower than in the center of the cell and that the diffusion coefficient scales with the mass of the used probes, while other parameters do not correlate with protein mobility.
Article
Chemistry, Multidisciplinary
Ilya O. Aparin, Rui Yan, Remi Pelletier, Alexander A. Choi, Dmytro I. Danylchuk, Ke Xu, Andrey S. Klymchenko
Summary: This study proposes a concept of ON/OFF switchable probes for super-resolution fluorescence microscopy based on bright cyanine dyes. The new probes provide higher brightness and slower diffusion at the cell surface, improving the localization precision. The study validated the concept and demonstrated the observation of nanoscale cellular structures with enhanced dynamics.
JOURNAL OF THE AMERICAN CHEMICAL SOCIETY
(2022)
Article
Biochemical Research Methods
Kun Chen, Wan Li, Ke Xu
Summary: Fluorescence microscopy is a valuable tool for studying complex biological systems, but the broad spectra of molecular fluorescence hinder the ability to distinguish numerous subcellular structures in a single sample. In this study, the authors propose a method based on multiple fluorescence bands to simultaneously image up to ten targets, with the potential for further expansion. By covering the excitation/emission bandwidth across the full visible range, they establish an ultra-broad excitation scheme, enabling superior spectral information and multiplexing capability.
BIOMEDICAL OPTICS EXPRESS
(2022)
Article
Chemistry, Physical
Ha H. Park, Alexander A. Choi, Ke Xu
Summary: By using expansion microscopy, we investigated the size-dependent suppression of molecular diffusivity in hydrogel nanomatrices with controlled meshwork size. Our research demonstrated that larger molecules experience more impeded diffusion within a fixed meshwork size, and diffusion is progressively suppressed as the meshwork size decreases, particularly for larger molecules. We also found that the obstruction of diffusion caused by the meshwork is independent of the suppression of diffusion due to increased solution viscosities. These mechanisms can scale down molecular diffusivity to produce diffusion slowdown in complex systems like cells.
JOURNAL OF PHYSICAL CHEMISTRY B
(2023)
Article
Chemistry, Multidisciplinary
Limin Xiang, Rui Yan, Kun Chen, Wan Li, Ke Xu
Summary: Using single-molecule displacement/diffusivity mapping (SMdM), we analyzed the diffusion of a typical fluorescent protein (FP) in living mammalian cells at nanoscale resolution. We found that the diffusion coefficients in the endoplasmic reticulum (ER) and mitochondrion are about 40% of that in the cytoplasm, with higher spatial inhomogeneities in the cytoplasm. Furthermore, we discovered that positive net charges impede diffusions in the ER lumen and mitochondrial matrix, while negative net charges do not.
Article
Biology
Ha H. Park, Bowen Wang, Suhong Moon, Tyler Jepson, Ke Xu
Summary: This study presents a machine-learning-enabled approach called Pix2D, which extracts the diffusion coefficient D directly from single molecule images and enables high-resolution spatial mapping of D.
COMMUNICATIONS BIOLOGY
(2023)
Article
Chemistry, Multidisciplinary
Changdong He, Chun Ying Wu, Wan Li, Ke Xu
Summary: The study reveals nanoscale heterogeneities in FUS condensates and their evolution over aging using multidimensional super-resolution microscopy tools. Hydrophobic nanodomains and amyloid aggregates are identified at the condensate surface, which significantly impede local diffusion. These nanoaggregates progressively expand on the condensate surface upon aging or mechanical shears, leading to a growing low-diffusivity shell while maintaining diffusion in the condensate interior.
JOURNAL OF THE AMERICAN CHEMICAL SOCIETY
(2023)
Article
Optics
Vu Nguyen, Wan Li, Joel Ager, Ke Xu, Hayden Taylor
Summary: Optical reflectance imaging is a simple and fast technique for characterizing 2D materials. It can reveal the interlayer coupling in stacked MoS2 and WS(2) bilayers through reflectance contrast. The determination of interlayer coupling requires optimization of illumination wavelength and SiO2 film thickness. This approach enables quick characterization of constructed 2D material systems.