A5100781205- Photoacoustic and Ultrasonic Imaging
- Thermography and Photoacoustic Techniques
- Additive Manufacturing Materials and Processes
- Bayesian Modeling and Causal Inference
- Additive Manufacturing and 3D Printing Technologies
- Infrared Thermography in Medicine
- Data Quality and Management
- Gear and Bearing Dynamics Analysis
- Wireless Signal Modulation Classification
- Library Science and Information Systems
- Fractal and DNA sequence analysis
- Radar Systems and Signal Processing
- High Entropy Alloys Studies
- Mechanical Failure Analysis and Simulation
- Computational and Text Analysis Methods
- Fire Detection and Safety Systems
- Advanced Battery Materials and Technologies
- Optical Imaging and Spectroscopy Techniques
- Industrial Technology and Control Systems
- Digital Imaging in Medicine
- Machine Fault Diagnosis Techniques
- Evacuation and Crowd Dynamics
- Natural Language Processing Techniques
- Retinal Imaging and Analysis
- Semantic Web and Ontologies
Anhui University of Science and Technology
2025
Beijing University of Chemical Technology
2025
Zhejiang University
2024
University of Pennsylvania
2024
Hebei University of Technology
2024
Union Hospital
2023
Huazhong University of Science and Technology
2023
Liaoning Technical University
2021-2023
Shandong University of Science and Technology
2023
California University of Pennsylvania
2023
A three-dimensional laser absorption model based on ray tracing was established to describe the coupled interaction of a beam with particles in powder layers pure tungsten (W) material processed by selective melting (SLM). The influence particle size powder-to-laser absorptivity and underlying behavior investigated. An intrinsic relationship between absorption, distribution absorbed irradiance within layers, surface morphology geometric characteristics (e.g., contact angle, width height...
Nowadays, breast cancer has increasingly threatened the health of human, especially females. However, is still hard to detect in early stage, and diagnostic procedure can be time-consuming with abundant expertise needed. In this paper, we explored deep learning algorithms emerging photoacoustic tomography for diagnostics. Specifically, used a pre-processing algorithm enhance quality uniformity input images transfer method achieve better classification performance. Besides, by comparing area...
To explore an intelligent detection technology based on deep learning algorithms to assist the clinical diagnosis of distal radius fractures (DRFs), and further compare it with human performance verify feasibility this method.A total 3,240 patients (fracture: n = 1,620, normal: 1,620) were included in study, a 3,276 wrist joint anteroposterior (AP) X-ray films (1,639 fractured, 1,637 normal) 3,260 lateral (1,623 normal). We divided into training set, validation set test ratio 7:1.5:1.5. The...
Modern machine learning algorithms have been adopted in a range of signal-processing applications spanning computer vision, natural language processing, and artificial intelligence. Many relevant problems involve subspace-structured features, orthogonality constrained or low-rank objective functions, subspace distances. These mathematical characteristics are expressed naturally using the Grassmann manifold. Unfortunately, this fact is not yet explored many traditional algorithms. In last few...
Coal gangue and flue gas desulfurization gypsum (DG) are industrial by-products whose utilization is critical for sustainable development. This study explores the synergistic activation of coal using DG lime to develop eco-friendly cementitious materials. Three systems-CF (coal gangue-DG), CL gangue-lime), CFL gangue-DG-lime)-were designed investigate effects on reactivity. Unconfined compressive strength tests identified optimal dosages lime, while XRD, SEM, TGA-DTG, NMR analyses were...
Hard carbon (HC) has emerged as a highly promising anode material for sodium-ion batteries (SIBs) attributed to its characteristic low-potential charge and discharge plateau. Recent studies have shown that the plateau capacity of HC mainly originates from filling nanoscale closed pores by sodium. However, precise design pore structure remains great challenge. Herein, C60 with diameter 0.7 nm is used promote formation in phenolic resin-based HC. The spherical facilitates oriented...
Photoacoustic tomography is a technique to reconstruct the image of light energy absorption distribution in tissues based on detected photoacoustic signals. In recent years, this research field has been greatly developed, and its application range wide, including anatomy, functionality, molecular imaging. However, conversion efficiency effect from sound quite low, which leads low signal-to-noise ratio signal poor quality reconstructed image. The traditional method improve signals data...
Zhaowei Wang, Quyet V. Do, Hongming Zhang, Jiayao Weiqi Tianqing Fang, Yangqiu Song, Ginny Wong, Simon See. Proceedings of the 61st Annual Meeting Association for Computational Linguistics (Volume 1: Long Papers). 2023.
Recent breakthroughs in machine learning especially artificial intelligence shift the paradigm of wireless communication towards radios. One their core operations is automatic modulation recognition (AMR). Existing research focuses on coherent schemes such as QAM, PSK and FSK. The AMR (non-coherent) space-time remains an uncharted area despite its wide deployment modern multiple-input-multiple-output (MIMO) systems. scheme using a so called Grassmann constellation enables rate-enhancement...
Photoacoustic (PA) imaging breaks the diffusion limit of conventional optical by listening to PA wave. As a new kind functional method, it has experienced tremendous growth in research community with wide range applications. However, is still an open and fundamental challenge that conversion efficiency from light sound based on effect extremely low. The consequence poor signal‐to‐noise ratio (SNR) signal especially scenarios low laser power deep penetration. Conventional approach enhance SNR...
Abstract Many recent studies have probed status bias in the peer-review process of academic journals and conferences. In this article, we investigated association between author metadata area chairs’ final decisions (Accept/Reject) using our compiled database 5,313 borderline submissions to International Conference on Learning Representations from 2017 2022 under a matched observational study framework. We carefully defined elements cause-and-effect analysis, including treatment its timing,...
This paper discusses the construction and implementation of forest early warning system from perspective large data storage, analysis structure.In this paper, architecture model combination video alert forecast are proposed, it analyzes it.Experimental results show that method has good expansibility superiority.
Recent breakthroughs in machine learning especially artificial intelligence shift the paradigm of wireless communication towards radios. One their core operations is automatic modulation recognition (AMR). Existing research focuses on coherent schemes such as QAM, PSK and FSK. The AMR (non- coherent) space-time remains an uncharted area despite its wide deployment modern multiple-input-multiple-output (MIMO) systems. scheme using a so called Grassmann constellation (comprising unitary...
1. Bagri S., Manwar A, Varghese A., Mujumdar Joshi S.S., 2021, Tool wear and remaining useful life prediction in micro-milling along complex tool paths using neural networks, Journal of Manufacturing Processes, 71, 679-698. CrossRef Google Scholar
Detecting commonsense causal relations (causation) between events has long been an essential yet challenging task. Given that are complicated, event may have different causes under various contexts. Thus, exploiting context plays role in detecting relations. Meanwhile, previous works about causation only consider two and ignore their context, simplifying the task formulation. This paper proposes a new to detect sequence (i.e., context), called contextualized reasoning. We also design...
We conducted an experiment during the review process of 2023 International Conference on Machine Learning (ICML) that requested authors with multiple submissions to rank their own papers based perceived quality. received 1,342 rankings, each from a distinct author, pertaining 2,592 submissions. In this paper, we present empirical analysis how author-provided rankings could be leveraged improve peer processes at machine learning conferences. focus Isotonic Mechanism, which calibrates raw...
With the flourishing of data-driven applications, data marketplaces, which can dramatically facilitate utilization, have emerged recently. However, determining appropriate price for datasets presents a significant challenge due to intangible nature data. DataPrice alleviates by providing an interactive system based on pricing model trained real collected from commercial marketplaces. The estimate according dataset description as reference both sellers and buyers. By leveraging Shapley values...
In the entire diabetic population, total number of patients with retinopathy is more than 50%, and longer diabetes, higher incidence rate blindness. Besides, blood vessel ocular fundus only that can be directly observed, which has excellent application value in medical diagnostics. Photoacoustic Tomography (PAT) an emerging technique obtain high-resolution 3D in-vivo images optical absorption by sensing laser-generated ultrasound. Therefore, this paper, we applied U-net neural network for...
We examine the role of textual data as study units when conducting causal inference by drawing parallels between human subjects and organized texts. %in population research. elaborate on key concepts principles, expose some ambiguity sometimes fallacies. To facilitate better framing a query, we discuss two strategies: (i) shifting from immutable traits to perceptions them, (ii) abstract concept/property its constituent parts, i.e., adopting constructivist perspective an concept. hope this...