Genomic prediction with NetGP based on gene network and multi‐omics data in plants DOI Creative Commons

Longyang Zhao,

Ping Tang, Jinjing Luo

et al.

Plant Biotechnology Journal, Journal Year: 2025, Volume and Issue: unknown

Published: Feb. 14, 2025

Summary Genomic selection (GS) is a new breeding strategy. Generally, traditional methods are used for predicting traits based on the whole genome. However, prediction accuracy of these models remains limited because they cannot fully reflect intricate nonlinear interactions between genotypes and traits. Here, novel single nucleotide polymorphism (SNP) feature extraction technique Pearson‐Collinearity Selection (PCS) firstly presented improves across several known models. Furthermore, gene network model (NetGP) deep learning approach designed phenotypic prediction. It utilizes transcriptomic dataset (Trans), genomic multi‐omics (Trans + SNP). The NetGP demonstrated better performance compared to other in predictions, predictions predictions. performed than independent or Prediction evaluations using plants' data showed good generalizability NetGP. Taken together, our study not only offers effective tool plant but also points avenues future research.

Language: Английский

Defense against adversarial attacks: robust and efficient compressed optimized neural networks DOI Creative Commons
Insaf Kraidia, Afifa Ghenai, Samir Brahim Belhaouari

et al.

Scientific Reports, Journal Year: 2024, Volume and Issue: 14(1)

Published: March 17, 2024

Abstract In the ongoing battle against adversarial attacks, adopting a suitable strategy to enhance model efficiency, bolster resistance threats, and ensure practical deployment is crucial. To achieve this goal, novel four-component methodology introduced. First, introducing pioneering batch-cumulative approach, exponential particle swarm optimization (ExPSO) algorithm was developed for meticulous parameter fine-tuning within each batch. A cumulative updating loss function employed overall optimization, demonstrating remarkable superiority over traditional techniques. Second, weight compression applied streamline deep neural network (DNN) parameters, boosting storage efficiency accelerating inference. It also introduces complexity deter potential attackers, enhancing accuracy in settings. This study compresses generative pre-trained transformer (GPT) by 65%, saving time memory without causing performance loss. Compared state-of-the-art methods, proposed method achieves lowest perplexity (14.28), highest (93.72%), an 8 × speedup central processing unit. The integration of preceding two components involves simultaneous training multiple versions compressed GPT. occurs across various rates different segments dataset ultimately associated with multi-expert architecture. enhancement significantly fortifies model's attacks into attackers' attempts anticipate prediction process. Consequently, leads average improvement 25% 14 attack scenarios datasets, surpassing capabilities current methods.

Language: Английский

Citations

5
Kidney Tumor Classification on CT images using Self-supervised Learning DOI
Erdal Özbay, Feyza Altunbey Özbay, Farhad Soleimanian Gharehchopogh

et al.

Computers in Biology and Medicine, Journal Year: 2024, Volume and Issue: 176, P. 108554 - 108554

Published: May 3, 2024

Language: Английский

Citations

5
Data reduction in big data: a survey of methods, challenges and future directions DOI

Tala Talaei Khoei,

Aditi Singh

International Journal of Data Science and Analytics, Journal Year: 2024, Volume and Issue: unknown

Published: July 10, 2024

Language: Английский

Citations

4
A survey of Emotional Artificial Intelligence and crimes: detection, prediction, challenges and future direction DOI

Tala Talaei Khoei,

Aditi Singh

Journal of Computational Social Science, Journal Year: 2024, Volume and Issue: 7(3), P. 2359 - 2402

Published: July 17, 2024

Language: Английский

Citations

4
Hygrothermal modeling in mass timber constructions: Recent advances and machine learning prospects DOI Creative Commons
Sina Akhavan Shams, Hua Ge, Lin Wang

et al.

Journal of Building Engineering, Journal Year: 2024, Volume and Issue: 96, P. 110500 - 110500

Published: Aug. 23, 2024

Language: Английский

Citations

4
Advanced Deep Learning Techniques for Battery Thermal Management in New Energy Vehicles DOI Creative Commons

Shaotong Qi,

Yubo Cheng,

Zhiyuan Li

et al.

Energies, Journal Year: 2024, Volume and Issue: 17(16), P. 4132 - 4132

Published: Aug. 19, 2024

In the current era of energy conservation and emission reduction, development electric other new vehicles is booming. With their various attributes, lithium batteries have become ideal power source for vehicles. However, lithium-ion are highly sensitive to temperature changes. Excessive temperatures, either high or low, can lead abnormal operation batteries, posing a threat safety entire vehicle. Therefore, developing reliable efficient Battery Thermal Management System (BTMS) that monitor battery status prevent thermal runaway becoming increasingly important. recent years, deep learning has gradually widely applied in fields as an method, it also been some extent BTMS. this work, we discuss basic principles related optimization elaborate on algorithmic principles, frameworks, applications advanced methods We several emerging algorithms proposed feasibility BTMS applications. Finally, obstacles faced by potential directions development, proposing ideas progress. This paper aims analyze technologies commonly used provide insights into combination technology trams assist

Language: Английский

Citations

4
Combining Transfer Learning and Ensemble Algorithms for Improved Citrus Leaf Disease Classification DOI Creative Commons
Hongyan Zhu,

Dani Wang,

Yuzhen Wei

et al.

Agriculture, Journal Year: 2024, Volume and Issue: 14(9), P. 1549 - 1549

Published: Sept. 7, 2024

Accurate categorization and timely control of leaf diseases are crucial for citrus growth. We proposed the Multi-Models Fusion Network (MMFN) detection based on model fusion transfer learning. Compared to traditional methods, algorithm (integrating learning Alexnet, VGG, Resnet) we can address issues limited categories, slow processing speed, low recognition accuracy. By constructing efficient deep models training optimizing them with a large dataset images, ensured broad applicability accuracy disease detection, achieving high-precision classification. Herein, various algorithms, including original Resnet, versions Resnet34 (Pre_Resnet34) Resnet50 (Pre_Resnet50) were also discussed compared. The results demonstrated that MMFN achieved an average 99.72% in distinguishing between diseased healthy leaves. Additionally, attained 98.68% classification multiple (citrus huanglongbing (HLB), greasy spot canker), insect pests miner), deficiency (zinc deficiency). These findings conclusively illustrate networks combining integration algorithms automatically extract image features, enhance automation recognition, demonstrate significant potential application value classification, potentially drive development smart agriculture.

Language: Английский

Citations

4
The application of machine learning in 3D/4D printed stimuli-responsive hydrogels DOI
Onome Ejeromedoghene, Moses Kumi,

Ephraim Akor

et al.

Advances in Colloid and Interface Science, Journal Year: 2024, Volume and Issue: 336, P. 103360 - 103360

Published: Nov. 27, 2024

Language: Английский

Citations

4
Detecting and assessing weak adhesion in structural single lap joints using a machine learning pipeline with lamb waves data DOI
Gabriel M. F. Ramalho, António M. Lopes, Lucas F. M. da Silva

et al.

Neural Computing and Applications, Journal Year: 2025, Volume and Issue: unknown

Published: Jan. 4, 2025

Language: Английский

Citations

0
A Review on Medical Image Segmentation: Datasets, Technical Models, Challenges and Solutions DOI Open Access
Hong‐Seng Gan, Muhammad Hanif Ramlee, Zimu Wang

et al.

Wiley Interdisciplinary Reviews Data Mining and Knowledge Discovery, Journal Year: 2025, Volume and Issue: 15(1)

Published: Jan. 20, 2025

ABSTRACT Medical image segmentation is prerequisite in computer‐aided diagnosis. As the field experiences tremendous paradigm changes since introduction of foundation models, technicality deep medical model no longer a privilege limited to computer science researchers. A comprehensive educational resource suitable for researchers broad, different backgrounds such as biomedical and medicine, needed. This review strategically covers evolving trends that happens fundamental components emerging multimodal datasets, updates on learning libraries, classical‐to‐contemporary development models latest challenges with focus enhancing interpretability generalizability model. Last, conclusion section highlights future worth further attention investigations.

Language: Английский

Citations

0