MétaCan
Menu
Back to cohort
Record W4408171278 · doi:10.1038/s41598-025-92396-9

Hybrid neural network method for damage localization in structural health monitoring

2025· article· en· W4408171278 on OpenAlexaff
Fatahlla Moreh, Yusuf Hasan, Zarghaam Haider Rizvi, Sven Tomforde, Frank Wuttke

Bibliographic record

VenueScientific Reports · 2025
Typearticle
Languageen
FieldEngineering
TopicInfrastructure Maintenance and Monitoring
Canadian institutionsUniversity of Waterloo
FundersChristian-Albrechts-Universität zu Kiel
KeywordsComputer scienceRecurrent neural networkArtificial intelligenceDomain (mathematical analysis)Deep learningComplement (music)SegmentationReduction (mathematics)Machine learningFeature engineeringArtificial neural network

Abstract

fetched live from OpenAlex

The detection of cracks in large structures is of critical importance, as such damage can result not only in significant financial costs but also pose serious risks to public safety. Many existing methods for crack detection rely on deep learning algorithms or traditional approaches that typically use image data. In this study, however, we explore an innovative approach based on numerical data, which is characterized by greater cost efficiency and offers intriguing research implications. This study emphasizes the evaluation of hybrid RNN-CNN models in comparison to the pure CNN models previously utilized in related research. Our proposed model incorporates a single RNN layer, complemented by essential supporting layers, which contributes to a reduction in complexity and a decrease in the number of parameters. This design choice results in a more streamlined and efficient architecture. Our experimental results reveal an accuracy of 78.9%, which, while slightly lower than the performance of conventional CNN models, underscores the potential of RNN layers in crack detection tasks. Importantly, this work demonstrates that integrating additional RNN layers can effectively enhance crack detection capabilities, particularly given the significance of preserving spatial information for accurate crack segmentation. These findings open avenues for further exploration and optimization of RNN-based methods in structural damage analysis, suggesting that the strategic use of RNNs can complement CNN models to achieve robust performance in this domain.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.

How this classification was reachedexpand

Full frame distilled prediction

Teacher imitation

Not calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.

metaresearch head score (Codex)0.001
metaresearch head score (Gemma)0.000
Version: codex-gemma-dda1882f352aValidation status: machine_predicted_unvalidated
Candidate categoriesnone
Consensus categoriesnone
DomainCandidate signal: none · Consensus signal: none
Study designCandidate signal: Simulation or modeling · Consensus signal: Simulation or modeling
GenreCandidate signal: Empirical · Consensus signal: Empirical
Teacher disagreement score0.519
Threshold uncertainty score0.554

Codex and Gemma teacher scores by category

CategoryCodexGemma
Metaresearch0.0010.000
Meta-epidemiology (narrow)0.0000.000
Meta-epidemiology (broad)0.0000.000
Bibliometrics0.0000.000
Science and technology studies0.0000.000
Scholarly communication0.0000.000
Open science0.0000.000
Research integrity0.0000.000
Insufficient payload (model declined to judge)0.0000.000

Machine scores (provisional)

The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.

Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.

Opus teacher head0.009
GPT teacher head0.288
Teacher spread0.280 · how far apart the two teachers sit on this one work
Validation statusscore_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it

Classification

machine, unvalidated

Machine predicted; a candidate call from one teacher head, not a consensus.

The models applied no category: nothing in the taxonomy fit this work.
Study designSimulation or modeling
Domainnot available
GenreEmpirical

How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".

Quick stats

Citations6
Published2025
Admission routes1
Has abstractyes

Explore more

Same venueScientific ReportsSame topicInfrastructure Maintenance and MonitoringFrench-language works237,207