Article
Engineering, Biomedical
Lin Yao, Bingzhao Zhu, Mahsa Shoaran
Summary: This work introduces the use of Riemannian-space features and temporal dynamics of electrocorticography (ECoG) signal combined with modern machine learning tools to improve the decoding accuracy of individual finger movements. By selecting informative biomarkers and exploring the concatenation of features, the proposed method achieved significant improvements in both classification and regression tasks. The results show that the approach outperformed previous methods in detecting individual finger movements and continuous decoding of movement trajectory, with a low time complexity.
JOURNAL OF NEURAL ENGINEERING
(2022)
Article
Computer Science, Information Systems
Eric Rigall, Xianglong Wang, Shu Zhang, Junyu Dong
Summary: In this paper, a phase-based SAR tag positioning method is proposed for fast and accurate item localization in large-scale warehousing and inventorying applications. The method replaces traditional holograms with a one-step hologram mask to accelerate the tag positioning task. Hash tables are also implemented to reduce hologram construction time. Comparative experiments show that the proposed model achieves similar positioning accuracy to state-of-the-art methods while offering a much lower inference time.
COMPUTER COMMUNICATIONS
(2023)
Article
Optics
Siddharth Rawat, Anna Wang
Summary: This article introduces a method that utilizes deep learning techniques to overcome noise limitations in holographic microscopy imaging, accurately extracting cell features. By using conditional generative adversarial networks to extract continuous phase values from holograms, phase maps can be effectively generated, with successful demonstrations on test objects, expanding potential applications.
Article
Automation & Control Systems
Tuan-Tang Le, Trung-Son Le, Yu-Ru Chen, Joel Vidal, Chyi-Yeu Lin
Summary: This paper proposes a novel real-time 3D object recognition and grasping solution with the potential to handle multi-object class scenes. Experimental results show high accuracy and efficiency, with significant improvements in performance metrics compared to existing methods.
ROBOTICS AND AUTONOMOUS SYSTEMS
(2021)
Article
Computer Science, Artificial Intelligence
Sjoerd Boeschoten, Cagatay Catal, Bedir Tekinerdogan, Arjen Lommen, Marco Blokland
Summary: This article proposes a framework that combines feature engineering techniques to compare different prediction models and select the best one. The framework can automatically generate various classification models and has been demonstrated in practical and research settings.
EXPERT SYSTEMS WITH APPLICATIONS
(2023)
Article
Materials Science, Multidisciplinary
Jiawei Xi, Jian Shen, Man To Chow, Tan Li, Jack Ng, Jensen Li
Summary: Multiplexing holography combined with metasurfaces using different degrees of light freedom has provided new applications in display and information processing. Polarization-multiplexed holograms can store the maximum amount of information, but it requires using bianisotropic metasurfaces instead of conventional single-layer nanostructures, which complicates the design and generation of holograms. In this study, an integrated neural network approach is developed to directly obtain metasurface profiles from independent holograms, allowing for complex polarization holograms without detailed knowledge of physical constraints.
ADVANCED OPTICAL MATERIALS
(2023)
Article
Chemistry, Multidisciplinary
Junhyeok Kang, Kanghan Oh, Il-Seok Oh
Summary: Recent studies have shown that using perturbation methods in deep learning can force networks to extract richer representations of images, leading to performance improvements. This study introduces effective perturbation approaches for medical landmark localization, with experiments demonstrating superior performance compared to traditional methods.
APPLIED SCIENCES-BASEL
(2021)
Article
Biochemistry & Molecular Biology
Phasit Charoenkwan, Chanin Nantasenamat, Md Mehedi Hasan, Mohammad Ali Moni, Balachandran Manavalan, Watshara Shoombuatong
Summary: A novel machine-learning meta-predictor UMPred-FRL was developed for improved umami peptide identification, combining six machine learning algorithms and seven feature encodings to achieve more accurate performance compared to baseline models.
INTERNATIONAL JOURNAL OF MOLECULAR SCIENCES
(2021)
Article
Oncology
Rajasekhar Chaganti, Furqan Rustam, Isabel De la Torre Diez, Juan Luis Vidal Mazon, Carmen Lili Rodriguez, Imran Ashraf
Summary: The study presents an approach for thyroid disease prediction using random forest-based features and achieves high accuracy. The proposed approach investigates feature engineering and utilizes machine learning models, showing superior performance in thyroid disease detection.
Article
Chemistry, Multidisciplinary
Gabriele M. Coli, Marjolein Dijkstra
Summary: Colloidal suspensions of two species have the ability to form binary crystals, which exhibit synergetic and collective properties. Investigations on the nucleation process of the exotic AB(13) crystal show that it follows classical nucleation theory, and neural network can accurately identify the AB(13) phase from competing crystals.
Article
Genetics & Heredity
Ramon Vinas, Tiago Azevedo, Eric R. Gamazon, Pietro Lio
Summary: Novel deep learning methods, PMI and GAIN-GTEx, were proposed for gene expression imputation, showing advantages in predictive performance and runtime over standard methods. PMI performed best in inductive imputation for protein-coding genes, while GAIN-GTEx outperformed in in-place imputation, indicating robust generalization on RNA-Seq data across cancer types.
FRONTIERS IN GENETICS
(2021)
Article
Multidisciplinary Sciences
Hadrien M. L. Robert, Kristyna Holanova, Lukasz Bujak, Milan Vala, Verena Henrichs, Zdenek Lansky, Marek Piliarik
Summary: The study introduces a high-speed photothermal spatial light modulator capable of generating a step-like wavefront change without diffraction artifacts, used for quantitative phase imaging to capture nanometer-scale 3D displacements.
NATURE COMMUNICATIONS
(2021)
Article
Chemistry, Analytical
Yulin He, Wei Chen, Chen Li, Xin Luo, Libo Huang
Summary: The study proposes an efficient method for extracting lane features in lane detection, involving two phases: local feature extraction and global feature aggregation. Additionally, the feature compression module based on decoupling representation learning effectively reduces redundancy and retains more critical information.
Article
Engineering, Civil
Chengju Zhou, Meiqing Wu, Siew-Kei Lam
Summary: In this paper, a unified multi-task learning architecture for fast and accurate pedestrian detection is proposed. By integrating a lightweight semantic segmentation branch and optimized modules, the architecture effectively combines pedestrian detection and semantic segmentation tasks, achieving improved detection performance without increasing computational overhead and achieving high detection performance with low resolution input images.
IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS
(2022)
Article
Biochemical Research Methods
Saleh Musleh, Mohammad Tariqul Islam, Rizwan Qureshi, Nihad Alajez, Tanvir Alam
Summary: In this article, a machine learning-based method called MSLP is proposed for predicting the subcellular localization of mRNA. By combining four types of features, including k-mer, PseKNC, physicochemical properties of nucleotides, and 3D representation of sequences based on Z-curve transformation, our method achieved state-of-the-art results in mRNA subcellular localization prediction tasks for multiple benchmark datasets.
BMC BIOINFORMATICS
(2023)
Article
Chemistry, Medicinal
Annemarie Winters, Fook Chiong Cheong, Mary Ann Odete, Juliana Lumer, David B. Ruffner, Kimberly Mishra, David G. Grier, Laura A. Philips
JOURNAL OF PHARMACEUTICAL SCIENCES
(2020)
Article
Biochemical Research Methods
Lauren E. Altman, David G. Grier
BIOMEDICAL OPTICS EXPRESS
(2020)
Article
Multidisciplinary Sciences
Guolong Zhu, Mark Hannel, Ruojie Sha, Feng Zhou, Matan Yah Ben Zion, Yin Zhang, Kyle Bishop, David Grier, Nadrian Seeman, Paul Chaikin
Summary: This study extends the programmability of DNA oligonucleotides to the micrometer-colloidal scale, utilizing optical microscopy and holographic optical tweezers for real-time observations. The research team has designed a chemomechanical device based on DNA origami structures, demonstrating high energy storage/retrieval capability and remote activation. This work paves the way for easily designed micromechanical devices bridging the molecular and colloidal/cellular scales.
PROCEEDINGS OF THE NATIONAL ACADEMY OF SCIENCES OF THE UNITED STATES OF AMERICA
(2021)
Article
Chemistry, Medicinal
Kurt D. Benkstein, Gurusamy Balakrishnan, Ashwinkumar Bhirde, Pascal Chalus, Tapan K. Das, Ngoc Do, David L. Duewer, Nazar Filonov, Fook Chiong Cheong, Patrick Garidel, Nicole S. Gill, Adam D. Grabarek, David G. Grier, Judith Hadley, Andrew D. Hollingsworth, Wesley W. Howard, Maciej Jarzebski, Wim Jiskoot, Sambit R. Kar, Vikram Kestens, Harshit Khasa, Yoen Joo Kim, Atanas Koulov, Anja Matter, Laura A. Philips, Christine Probst, Yannic Ramaye, Theodore W. Randolph, Dean C. Ripple, Stefan Romeijn, Miguel Saggu, Franziska Schleinzer, Jared R. Snell, Jan Kuba Tatarkiewicz, Heather Anne Wright, Dennis T. Yang
Summary: This paper presents a study on the measurement variability of sub-micrometer polydisperse protein aggregates and particles in biotherapeutics. It includes an interlaboratory comparison involving 20 laboratories and different particle-counting instruments. The results show high variability between datasets from different laboratories and provide guidance for instrument performance assessment and the development of polydisperse reference materials.
JOURNAL OF PHARMACEUTICAL SCIENCES
(2022)
Article
Astronomy & Astrophysics
Michael J. O'Brien, Blakesley Burkhart, Michael J. Shelley
Summary: This study demonstrates the application of the bispectrum, the Fourier three-point correlation function, for studying driving scales of magnetohydrodynamic (MHD) turbulence in the interstellar medium. The bispectrum is calculated using a parallelized Monte Carlo direct measurement method. The study finds that the bicoherence, a related statistic, can identify turbulence-driving scales using density and column density fields.
ASTROPHYSICAL JOURNAL
(2022)
Article
Optics
Jatin Abacousnac, David G. Grier
Summary: This paper introduces a dark trap created by superimposing Gaussian modes with different waist diameters, which is used to manipulate light-seeking and dark-seeking particles in the same system. The interference-based difference-of-Gaussians (DoG) trap can rigidly capture dark-seeking particles in three dimensions.
Article
Optics
Rafe Abdulali, Lauren E. Altman, David G. Grier
Summary: Holographic particle characterization is a method that uses quantitative analysis of holographic microscopy data to measure the diameter and refractive index of particles precisely and rapidly. It can measure the properties of an effective sphere enclosing each particle when applied to inhomogeneous or aspherical particles. This technique can also measure the fractal dimensions of individual fractal clusters by probing the changes in their effective diameter and refractive index during rotational diffusion.
Article
Chemistry, Physical
Lauren E. Altman, David G. Grier
Summary: Holographic particle characterization uses video microscopy to track and characterize individual colloidal particles. It has a wide range of applications from fundamental research to product development. Conventional optimization algorithms yield high precision for position, size, and index of refraction, while machine learning automates the process. This study presents an updated neural network solution called CATCH, which is fast and accurate for high-throughput applications.
Article
Physics, Fluids & Plasmas
Lauren E. Altman, David G. Grier
Summary: This study investigates the diffusion and sedimentation of colloidal particles between parallel horizontal walls using holographic particle tracking. The results provide insights into the particles' mobility perpendicular to the walls and their composition.
Review
Multidisciplinary Sciences
Caroline Martin, Lauren E. Altman, Siddharth Rawat, Anna Wang, David G. Grier, Vinothan N. Manoharan
Summary: Holographic microscopy is an optical microscope that collects holograms and analyzes them to obtain information about specimen properties. It offers high-speed acquisition of precise results. This Primer introduces in-line holographic microscopy and three analysis methods and discusses applications, reproducibility, and limitations. It also provides an outlook on future development and the integration between experiment and computational analysis.
NATURE REVIEWS METHODS PRIMERS
(2022)
Article
Physics, Multidisciplinary
Mohammed A. Abdelaziz, Jairo A. A. Diaz, Jean-Luc Aider, David J. Pine, David G. Grier, Mauricio Hoyos
Summary: The study found that emulsion droplets trapped in an ultrasonic levitator organize themselves into single-file chains, unlike solid spheres which tend to form planar structures. The behavior of these chains suggests that the droplets are rapidly spinning around a common axis, distinguishing them from solid spheres. This acoustically induced spinning mechanism guides the self-organization of acoustically levitated matter, as demonstrated in the model system presented in the study.
PHYSICAL REVIEW RESEARCH
(2021)
Article
Chemistry, Physical
Lauren E. Altman, Rushna Quddus, Fook Chiong Cheong, David G. Grier
Summary: The Lorenz-Mie theory can be used to measure the three-dimensional position of colloidal spheres with nanometer-scale precision and the diameter and refractive index with part-per-thousand precision. Effective-sphere interpretation has been successfully applied to various types of particles, including porous, dimpled, coated spheres, and fractal clusters of nanoparticles. Experimental studies show that the effective-sphere estimation accurately distinguishes colloidal clusters and can be used to measure the position and orientation of colloidal dimers.
Article
Physics, Multidisciplinary
Mohammed A. Abdelaziz, David G. Grier
Summary: A focused acoustic standing wave can stably levitate a small sphere and exposing it to a transverse traveling sound wave can drive it away from mechanical equilibrium. The interference between the trapping wave and the driving wave shapes the driving force, leading to oscillations at different frequencies if the waves are detuned. The wave-driven harmonic oscillator exhibits a rich variety of dynamical behaviors, including oscillations at harmonics and subharmonics, period-doubling routes to chaos, and Fibonacci cascades, showcasing opportunities for dynamic acoustic manipulation based on spectral control of the sound field.
PHYSICAL REVIEW RESEARCH
(2021)
Article
Chemistry, Physical
Kaitlynn Snyder, Rushna Quddus, Andrew D. Hollingsworth, Kent Kirshenbaum, David G. Grier
Article
Physics, Multidisciplinary
Mohammed A. Abdelaziz, David G. Grier
PHYSICAL REVIEW RESEARCH
(2020)