VNC-Dist: A machine learning-based tool for quantification of neuronal positioning in the ventral nerve cord of C. elegans DOI Open Access

Saber Saharkhiz,

Mearhyn Petite,

Tony Roenspies

et al.

bioRxiv (Cold Spring Harbor Laboratory), Journal Year: 2024, Volume and Issue: unknown

Published: Nov. 18, 2024

Abstract The ventral nerve cord (VNC) of newly hatched C. elegans contains 22 motoneurons organized into three distinct classes: DD, DA, and DB, that show stereotypical positioning arrangement along its length. VNC represents a genetically tractable model to investigate mechanisms involved in neuron sorting positioning. However, accurately efficiently mapping quantifying all motoneuron positions within large datasets is major challenge. Here, we introduce VNC-Dist, semi-automated software toolbox designed overcome the limitations subjective analysis microscopy. VNC-Dist uses an annotator for localization automated contour-based method measuring relative distances neurons based on deep learning numerical analysis. To demonstrate robustness versatility applied it multiple genetic mutants known disrupt VNC. This will enable acquisition neuronal positioning, thereby advancing investigations cellular molecular control

Language: Английский

Physics-constrained wind power forecasting aligned with probability distributions for noise-resilient deep learning DOI
Jiaxin Gao, Yong Cheng, Dongxiao Zhang

et al.

Applied Energy, Journal Year: 2025, Volume and Issue: 383, P. 125295 - 125295

Published: Jan. 15, 2025

Language: Английский

Citations

5
Enhancing wind power forecasting accuracy through LSTM with adaptive wind speed calibration (C-LSTM) DOI Creative Commons
Ding Wang, Min Xu, Guangming Zhu

et al.

Scientific Reports, Journal Year: 2025, Volume and Issue: 15(1)

Published: Feb. 13, 2025

Language: Английский

Citations

2
Deep probabilistic solar power forecasting with Transformer and Gaussian process approximation DOI
Binyu Xiong, Yuntian Chen, Dali Chen

et al.

Applied Energy, Journal Year: 2025, Volume and Issue: 382, P. 125294 - 125294

Published: Jan. 13, 2025

Language: Английский

Citations

1
Physics-constrained robust learning of open-form partial differential equations from limited and noisy data DOI Creative Commons
Mengge Du, Yuntian Chen, Longfeng Nie

et al.

Physics of Fluids, Journal Year: 2024, Volume and Issue: 36(5)

Published: May 1, 2024

Unveiling the underlying governing equations of nonlinear dynamic systems remains a significant challenge. Insufficient prior knowledge hinders determination an accurate candidate library, while noisy observations lead to imprecise evaluations, which in turn result redundant function terms or erroneous equations. This study proposes framework robustly uncover open-form partial differential (PDEs) from limited and data. The operates through two alternating update processes: discovering embedding. phase employs symbolic representation novel reinforcement learning (RL)-guided hybrid PDE generator efficiently produce diverse PDEs with tree structures. A neural network-based predictive model fits system response serves as reward evaluator for generated PDEs. higher rewards are utilized iteratively optimize via RL strategy best-performing is selected by parameter-free stability metric. embedding integrates initially identified process physical constraint into robust training. traversal trees automates construction computational graph without human intervention. Numerical experiments demonstrate our framework's capability highly data outperform other physics-informed discovery methods. work opens new potential exploring real-world understanding.

Language: Английский

Citations

6
Energy consumption dynamic prediction for HVAC systems based on feature clustering deconstruction and model training adaptation DOI
Huiheng Liu, Yanchen Liu, Huakun Huang

et al.

Building Simulation, Journal Year: 2024, Volume and Issue: 17(9), P. 1439 - 1460

Published: July 19, 2024

Language: Английский

Citations

6
Adaptive Control Method for Morphing Trailing-Edge Wing Based on Deep Supervision Network and Reinforcement Learning DOI
Jiahua Dai, Peiqing Liu,

Chuihuan Kong

et al.

Aerospace Science and Technology, Journal Year: 2024, Volume and Issue: 153, P. 109424 - 109424

Published: July 27, 2024

Language: Английский

Citations

4
Transfer Learning on Physics-Informed Neural Networks for Tracking the Hemodynamics in the Evolving False Lumen of Dissected Aorta DOI Creative Commons

Mitchell Daneker,

Shengze Cai, Ying Qian

et al.

Deleted Journal, Journal Year: 2024, Volume and Issue: 1(2), P. 100016 - 100016

Published: May 22, 2024

Aortic dissection is a life-threatening event that responsible for significant morbidity and mortality in individuals ranging age from children to older adults. A better understanding of the complex hemodynamic environment inside aorta enables clinicians assess patient-specific risk complications administer timely interventions. In this study, we propose develop validate new computational framework, warm-start physics-informed neural networks (WS-PINNs), address limitations current approaches analyzing hemodynamics false lumen (FL) type B aortic vessels reconstructed apolipoprotein null mice infused with AngII, thereby significantly reducing amount required measurement data eliminating dependency predictions on accuracy availability inflow/outflow boundary conditions. Specifically, demonstrate WS-PINN models allow us focus assessing 3D flow field FL without modeling true various branched vessels. Furthermore, investigate impact spatial temporal resolutions MRI prediction PINN model, which can guide acquisition reduce time financial costs. Finally, consider use transfer learning provide faster results when looking at similar but geometries. Our indicate proposed framework enhance capacity analysis dissections, promise eventually leading improved prognostic ability development aneurysms.

Language: Английский

Citations

3
Complex-valued physics-informed machine learning for efficient solving of quintic nonlinear Schrödinger equations DOI Creative Commons
Lei Zhang, Mengge Du, Xiaodong Bai

et al.

Physical Review Research, Journal Year: 2025, Volume and Issue: 7(1)

Published: Feb. 13, 2025

The Gross-Pitaevskii equation (GPE), a specialized form of the nonlinear Schrödinger (NLSE), plays pivotal role in quantum mechanics, optics, and condensed matter physics, modeling phenomena such as superfluidity, turbulence, solitons, while serving cornerstone for advancing study wave propagation its technological applications. In this paper, we propose solver, Complex-Valued Physics Informed Neural Network (CV-PINN) NLSEs using physics-informed learning machines with complex representation. This method integrates values algebraic properties directly into neural network, structure mirroring computation process numbers, thereby significantly enhancing ability to effectively solve problems. Additionally, introduce collocation-point sampling called Predictive Dynamic Monitoring Sampling (PDM sampling), which adaptively adjusts distribution collocation points during training based on model's historical performance. We conducted extensive empirical evaluations CV-PINN PDM series NLSE/GPEs (a total 16 solved examples). Compared traditional real-valued PINN, demonstrated higher accuracy, faster convergence, greater robustness, better stability these cases. Moreover, proved be effective preventing model degradation convergence speed predictive accuracy when compared methods. advancement provides an approach perspective solving partial differential equations, offering insights broader application PINNs across various fields. Published by American Physical Society 2025

Language: Английский

Citations

0
Mechanism-data dual-driven research framework for whole-process water conservation and carbon emission reduction within industrial parks DOI Creative Commons

Yuehong Zhao,

Hongbin Cao

Published: Feb. 1, 2025

Language: Английский

Citations

0
Computational Super-Resolution: An Odyssey in Harnessing Priors to Enhance Optical Microscopy Resolution DOI Creative Commons

Wei Tian,

Riwang Chen,

Liangyi Chen

et al.

Analytical Chemistry, Journal Year: 2025, Volume and Issue: unknown

Published: Feb. 27, 2025

Language: Английский

Citations

0