Multimodal data generative fusion method for complex system health condition estimation

Abstract For the health management of complex systems, the high value of such systems often necessitates multimodal monitoring data, including video surveillance, internal sensors, empirical formulas, and even digital twins. Therefore, it is essential to design an effective intelligent fusion method...

Full description

Saved in:
Bibliographic Details
Main Authors: Weijie Kang, Xianyang Zhang, Jiarui Zhang, Xudan Chen, Honglan Huang, Bing He, Weiwei Qin, Haizhen Zhu
Format: Article
Language:English
Published: Nature Portfolio 2025-06-01
Series:Scientific Reports
Subjects:
Online Access:https://doi.org/10.1038/s41598-025-04985-3
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850136567349248000
author Weijie Kang
Xianyang Zhang
Jiarui Zhang
Xudan Chen
Honglan Huang
Bing He
Weiwei Qin
Haizhen Zhu
author_facet Weijie Kang
Xianyang Zhang
Jiarui Zhang
Xudan Chen
Honglan Huang
Bing He
Weiwei Qin
Haizhen Zhu
author_sort Weijie Kang
collection DOAJ
description Abstract For the health management of complex systems, the high value of such systems often necessitates multimodal monitoring data, including video surveillance, internal sensors, empirical formulas, and even digital twins. Therefore, it is essential to design an effective intelligent fusion method for multimodal data. Firstly, a global monotonicity calculation method and a time series data augmentation technique are developed to address the inconsistencies arising from varying temporal lengths across different modalities. Secondly, in response to the need for efficient time series fusion, we propose a fast sequential learning network architecture along with a time series generative data structure. Finally, we introduce a many-to-many transfer training approach that culminates in the formation of a Multi-source Generative Adversarial Network (Ms-GAN). Numerical experiments and monitoring datasets are employed to validate the effectiveness of this multimodal generative fusion method. Notably, Ms-GAN enhances traditional GANs—typically limited to learning single data distributions—by enabling multimodal data fusion capabilities. This advancement holds significant promise for applications in various fields such as multimedia processing and medical diagnosis.
format Article
id doaj-art-b71b9aeceecf4711a4c0e43e9d526638
institution OA Journals
issn 2045-2322
language English
publishDate 2025-06-01
publisher Nature Portfolio
record_format Article
series Scientific Reports
spelling doaj-art-b71b9aeceecf4711a4c0e43e9d5266382025-08-20T02:31:04ZengNature PortfolioScientific Reports2045-23222025-06-0115111710.1038/s41598-025-04985-3Multimodal data generative fusion method for complex system health condition estimationWeijie Kang0Xianyang Zhang1Jiarui Zhang2Xudan Chen3Honglan Huang4Bing He5Weiwei Qin6Haizhen Zhu7Rocket Force University of EngineeringRocket Force University of EngineeringNorthwest Institute of Nuclear TechnologyRocket Force University of EngineeringRocket Force University of EngineeringRocket Force University of EngineeringRocket Force University of EngineeringAir Force Engineering UniversityAbstract For the health management of complex systems, the high value of such systems often necessitates multimodal monitoring data, including video surveillance, internal sensors, empirical formulas, and even digital twins. Therefore, it is essential to design an effective intelligent fusion method for multimodal data. Firstly, a global monotonicity calculation method and a time series data augmentation technique are developed to address the inconsistencies arising from varying temporal lengths across different modalities. Secondly, in response to the need for efficient time series fusion, we propose a fast sequential learning network architecture along with a time series generative data structure. Finally, we introduce a many-to-many transfer training approach that culminates in the formation of a Multi-source Generative Adversarial Network (Ms-GAN). Numerical experiments and monitoring datasets are employed to validate the effectiveness of this multimodal generative fusion method. Notably, Ms-GAN enhances traditional GANs—typically limited to learning single data distributions—by enabling multimodal data fusion capabilities. This advancement holds significant promise for applications in various fields such as multimedia processing and medical diagnosis.https://doi.org/10.1038/s41598-025-04985-3Multimodal informationHealth managementHealth condition EstimationGenerative adversarial networksKernel canonical correlation analysis
spellingShingle Weijie Kang
Xianyang Zhang
Jiarui Zhang
Xudan Chen
Honglan Huang
Bing He
Weiwei Qin
Haizhen Zhu
Multimodal data generative fusion method for complex system health condition estimation
Scientific Reports
Multimodal information
Health management
Health condition Estimation
Generative adversarial networks
Kernel canonical correlation analysis
title Multimodal data generative fusion method for complex system health condition estimation
title_full Multimodal data generative fusion method for complex system health condition estimation
title_fullStr Multimodal data generative fusion method for complex system health condition estimation
title_full_unstemmed Multimodal data generative fusion method for complex system health condition estimation
title_short Multimodal data generative fusion method for complex system health condition estimation
title_sort multimodal data generative fusion method for complex system health condition estimation
topic Multimodal information
Health management
Health condition Estimation
Generative adversarial networks
Kernel canonical correlation analysis
url https://doi.org/10.1038/s41598-025-04985-3
work_keys_str_mv AT weijiekang multimodaldatagenerativefusionmethodforcomplexsystemhealthconditionestimation
AT xianyangzhang multimodaldatagenerativefusionmethodforcomplexsystemhealthconditionestimation
AT jiaruizhang multimodaldatagenerativefusionmethodforcomplexsystemhealthconditionestimation
AT xudanchen multimodaldatagenerativefusionmethodforcomplexsystemhealthconditionestimation
AT honglanhuang multimodaldatagenerativefusionmethodforcomplexsystemhealthconditionestimation
AT binghe multimodaldatagenerativefusionmethodforcomplexsystemhealthconditionestimation
AT weiweiqin multimodaldatagenerativefusionmethodforcomplexsystemhealthconditionestimation
AT haizhenzhu multimodaldatagenerativefusionmethodforcomplexsystemhealthconditionestimation