{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,21]],"date-time":"2026-09-21T18:12:55Z","timestamp":1790014375645,"version":"4.0.1"},"reference-count":37,"publisher":"Public Library of Science (PLoS)","issue":"9","license":[{"start":{"date-parts":[[2026,9,21]],"date-time":"2026-09-21T00:00:00Z","timestamp":1789948800000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"name":"University of Transport Technology","award":["DTTD2023-07"],"award-info":[{"award-number":["DTTD2023-07"]}],"id":[{"id":"https:\/\/ror.org\/04dp8y877","id-type":"ROR","asserted-by":"crossref"}]}],"content-domain":{"domain":["www.plosone.org"],"crossmark-restriction":false},"short-container-title":["PLoS One"],"abstract":"<jats:p>\n                    Vibration-based structural health monitoring (SHM) detects damage by identifying subtle changes in a structure\u2019s dynamic response. Deep learning models have shown strong potential for automating this classification task. However, most existing architectures share a key limitation: they either process signals purely in the time domain, or apply a global frequency-domain transform that discards information about when each frequency component occurs. This is a serious drawback, because structural damage typically manifests as short, localised events confined to specific frequency bands. This paper proposes SpectralTCN, a temporal convolutional network augmented with a Wavelet Gating Module (WGM) that performs learnable, data-dependent multi-resolution filtering within each convolutional block. The WGM decomposes intermediate features via the Discrete Wavelet Transform (DWT) into physically interpretable sub-bands corresponding to distinct structural vibration modes, applies input-adaptive sigmoid gates independently at each decomposition level, and reconstructs the filtered signal via the Inverse DWT (IDWT) with a learnable residual connection initialised to zero. Unlike FFT-based approaches, the DWT simultaneously preserves both time and frequency information, enabling the network to detect both the timing and the spectral location of damage-induced anomalies. Combined with Generalized Mean (GeM) pooling and large-kernel causal depthwise convolutions, SpectralTCN is evaluated on two benchmark datasets: the Z24 Bridge benchmark and a finite element model (FEM)-derived dataset of the My Thuan cable-stayed bridge. Experiments against 10 baseline models and 4 ablation variants, evaluated via stratified 5-fold cross-validation, demonstrate the effectiveness and generalisation capability of the proposed approach: SpectralTCN attains the highest mean accuracy on both benchmarks (92.2% on Z24 and 92.1% on My Thuan) and outperforms the strongest baseline in the 5-fold cross-validation protocol. In addition, the proposed architecture operates on short, streaming acceleration windows with a purely convolutional backbone of moderate computational cost, making SpectralTCN suitable for\n                    <jats:bold>near-real-time, online damage detection<\/jats:bold>\n                    in continuous bridge monitoring.\n                  <\/jats:p>","DOI":"10.1371\/journal.pone.0358224","type":"journal-article","created":{"date-parts":[[2026,9,21]],"date-time":"2026-09-21T17:46:02Z","timestamp":1790012762000},"page":"e0358224","update-policy":"https:\/\/doi.org\/10.1371\/journal.pone.corrections_policy","source":"Crossref","is-referenced-by-count":0,"title":["SpectralTCN: A multi-resolution wavelet gating network for vibration-based structural damage detection"],"prefix":"10.1371","volume":"21","author":[{"given":"Quan","family":"Pham Hong","sequence":"first","affiliation":[{"name":"University of Transport Technology","place":["Ha Noi, Viet Nam"],"department":["Faculty of Civil Engineering"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0009-7596-1520","authenticated-orcid":true,"given":"Trung","family":"Vu Manh","sequence":"additional","affiliation":[{"name":"University of Transport and Communications","place":["Hanoi, Vietnam"],"department":["Application of Artificial Intelligence for Structural Health Monitoring (AI-SHM) Research Group (R2026-UTC-01), Faculty of Civil and Environmental Engineering"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Bich","family":"Nguyen Thach","sequence":"additional","affiliation":[{"name":"University of Transport and Communications","place":["Hanoi, Vietnam"],"department":["Application of Artificial Intelligence for Structural Health Monitoring (AI-SHM) Research Group (R2026-UTC-01), Faculty of Civil and Environmental Engineering"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0008-6351-1795","authenticated-orcid":true,"given":"Le","family":"Nguyen Dan","sequence":"additional","affiliation":[{"name":"University of Transport and Communications","place":["Hanoi, Vietnam"],"department":["Application of Artificial Intelligence for Structural Health Monitoring (AI-SHM) Research Group (R2026-UTC-01), Faculty of Civil and Environmental Engineering"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0005-9255-3263","authenticated-orcid":true,"given":"Hoa","family":"Tran Ngoc","sequence":"additional","affiliation":[{"name":"University of Transport and Communications","place":["Hanoi, Vietnam"],"department":["Application of Artificial Intelligence for Structural Health Monitoring (AI-SHM) Research Group (R2026-UTC-01), Faculty of Civil and Environmental Engineering"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"340","published-online":{"date-parts":[[2026,9,21]]},"reference":[{"issue":"1851","key":"pone.0358224.ref001","first-page":"303","article-title":"An introduction to structural health monitoring","volume":"365","author":"CR Farrar","year":"2007","journal-title":"Philos Trans A Math Phys Eng Sci"},{"issue":"2","key":"pone.0358224.ref002","doi-asserted-by":"crossref","first-page":"91","DOI":"10.1177\/058310249803000201","article-title":"A summary review of vibration-based damage identification methods","volume":"30","author":"SW Doebling","year":"1998","journal-title":"Shock Vib Dig"},{"key":"pone.0358224.ref003","doi-asserted-by":"crossref","first-page":"174","DOI":"10.28991\/CEJ-SP2024-010-010","article-title":"Impact of vertical vibration on group piles during earthquake loading: experimental findings","volume":"10","author":"BJ Noman","year":"2024","journal-title":"Civ Eng J"},{"key":"pone.0358224.ref004","first-page":"1","article-title":"Swarm intelligence-based technique to enhance performance of ANN in structural damage detection","volume":"73","author":"VL Ho","year":"2022","journal-title":"Transp Commun Sci J"},{"key":"pone.0358224.ref005","doi-asserted-by":"crossref","first-page":"154","DOI":"10.1016\/j.jsv.2016.10.043","article-title":"Real-time vibration-based structural damage detection using one-dimensional convolutional neural networks","volume":"388","author":"O Abdeljaber","year":"2017","journal-title":"J Sound Vib"},{"key":"pone.0358224.ref006","doi-asserted-by":"crossref","first-page":"436","DOI":"10.1016\/j.istruc.2021.10.088","article-title":"Vibration-based multiclass damage detection and localization using long short-term memory networks","volume":"35","author":"S Sony","year":"2022","journal-title":"Structures"},{"key":"pone.0358224.ref007","unstructured":"Vaswani A, Shazeer N, Parmar N, Uszkoreit J, Jones L, Gomez AN, et al. Attention is all you need. 31st Conference on Neural Information Processing Systems (NIPS 2017); 2017. p. 5998\u20136008. https:\/\/doi.org\/10.48550\/arXiv.1706.03762"},{"key":"pone.0358224.ref008","article-title":"An empirical evaluation of generic convolutional and recurrent networks for sequence modeling.","author":"S Bai","year":"2018"},{"key":"pone.0358224.ref009","unstructured":"Luo D, Wang X. ModernTCN: A modern pure convolutional structure for general time series analysis. International Conference on Learning Representations (ICLR); 2024."},{"key":"pone.0358224.ref010","doi-asserted-by":"crossref","unstructured":"Lee-Thorp J, Ainslie J, Eckstein I, Onta\u00f1\u00f3n S. FNet: mixing tokens with fourier transforms. In: Carpuat M, de Marneffe M-C, Meza Ruiz IV, editors. Proceedings of the 2022 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies. Seattle, United States: Association for Computational Linguistics; 2022. p. 4296\u2013313. https:\/\/doi.org\/10.18653\/v1\/2022.naacl-main.319","DOI":"10.18653\/v1\/2022.naacl-main.319"},{"key":"pone.0358224.ref011","first-page":"13001","article-title":"Global filter networks for image classification","volume":"34","author":"Y Rao","year":"2021","journal-title":"Adv Neural Inf Process Syst"},{"key":"pone.0358224.ref012","doi-asserted-by":"crossref","first-page":"76656","DOI":"10.52202\/075280-3349","article-title":"Frequency-domain MLPs are more effective learners in time series forecasting","volume":"36","author":"K Yi","year":"2023","journal-title":"Adv Neural Inf Process Syst"},{"key":"pone.0358224.ref013","doi-asserted-by":"crossref","first-page":"107077","DOI":"10.1016\/j.ymssp.2020.107077","article-title":"A review of vibration-based damage detection in civil structures: From traditional methods to Machine Learning and Deep Learning applications","volume":"147","author":"O Avci","year":"2021","journal-title":"Mech Syst Signal Process"},{"issue":"7","key":"pone.0358224.ref014","doi-asserted-by":"crossref","first-page":"1655","DOI":"10.1109\/TPAMI.2018.2846566","article-title":"Fine-tuning CNN image retrieval with no human annotation","volume":"41","author":"F Radenovic","year":"2019","journal-title":"IEEE Trans Pattern Anal Mach Intell"},{"issue":"1","key":"pone.0358224.ref015","doi-asserted-by":"crossref","first-page":"127","DOI":"10.1006\/mssp.2002.1548","article-title":"Description of Z24 benchmark","volume":"17","author":"J Maeck","year":"2003","journal-title":"Mech Syst Signal Process"},{"key":"pone.0358224.ref016","doi-asserted-by":"crossref","first-page":"485","DOI":"10.47869\/tcsj.77.4.11","article-title":"Vibration-based damage detection in cable-stayed bridges using a novel 1D-ConvNeXt-LSTM network","volume":"77","author":"NH Xuan","year":"2026","journal-title":"Transp Commun Sci J"},{"key":"pone.0358224.ref017","first-page":"1281","article-title":"Detection of damage in steel truss bridges using a hybrid 1DCNN\u2013BIGRU model and time-series data augmentation techniques","volume":"76","author":"DNL Minh","year":"2025","journal-title":"Transp Commun Sci J"},{"key":"pone.0358224.ref018","first-page":"2226","article-title":"Damage detection in structural health monitoring using BiLSTM-1DCNN hybrid network: a case study on a large-scale steel truss bridge","volume":"42","author":"TNC Nguyen","year":"2025","journal-title":"Eng Comput"},{"key":"pone.0358224.ref019","doi-asserted-by":"crossref","unstructured":"Manh T-V, Ngoc H-T, Duc M-T, Phuc L-B, Duc L-N. An effective damage detection approach for a truss bridge using a hybrid deep learning model. In: Thanh N-V, Nguyen-Ngoc L, Bui-Tien T, et al., editors. Proceedings of the 5th International Conference on Sustainability in Civil Engineering-Volume 2. Singapore: Springer Nature; 2025. p. 91\u2013101. https:\/\/doi.org\/10.1007\/978-981-96-5206-8_10","DOI":"10.1007\/978-981-96-5206-8_10"},{"key":"pone.0358224.ref020","article-title":"Enhanced structural health monitoring using symbolic aggregate approximation with hybrid deep learning models","volume":"31","author":"QP Hong","year":"2026","journal-title":"Mat\u00e9ria (Rio J)"},{"key":"pone.0358224.ref021","doi-asserted-by":"crossref","first-page":"1936","DOI":"10.1007\/s10618-020-00710-y","article-title":"InceptionTime: finding AlexNet for time series classification","volume":"34","author":"FH Ismail","year":"2020","journal-title":"Data Min Knowl Discov"},{"key":"pone.0358224.ref022","doi-asserted-by":"crossref","first-page":"1959","DOI":"10.1007\/s42417-021-00343-5","article-title":"A data-driven approach to structural health monitoring of bridge structures based on the discrete model and FFT-deep learning","volume":"9","author":"TQ Nguyen","year":"2021","journal-title":"J Vib Eng Technol"},{"key":"pone.0358224.ref023","doi-asserted-by":"crossref","first-page":"4209","DOI":"10.1007\/s42417-023-01116-y","article-title":"A novel approach for structural damage detection using multi-headed stacked deep ensemble learning","volume":"12","author":"A Asghari","year":"2024","journal-title":"J Vib Eng Technol"},{"key":"pone.0358224.ref024","doi-asserted-by":"crossref","first-page":"3986","DOI":"10.28991\/CEJ-2024-010-12-011","article-title":"Sensor layout design for structural health monitoring","volume":"10","author":"J Kyung","year":"2024","journal-title":"Civ Eng J"},{"issue":"2","key":"pone.0358224.ref025","doi-asserted-by":"crossref","first-page":"398","DOI":"10.28991\/HIJ-2025-06-02-03","article-title":"UAV-based structural health monitoring using a two-stage CNN model with lighthouse localization in GNSS-denied environments","volume":"6","author":"TSC Chu","year":"2025","journal-title":"HighTech Innov J"},{"issue":"9","key":"pone.0358224.ref026","article-title":"Structural damage identification using single-point vibration data processing","volume":"20","author":"H-Y Chu","year":"2025","journal-title":"PLoS One"},{"key":"pone.0358224.ref027","doi-asserted-by":"crossref","unstructured":"Liu Z, Mao H, Wu CY, Feichtenhofer C, Darrell T, Xie S. A ConvNet for the 2020s. Proceedings-2022 IEEE\/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2022. IEEE Computer Society; 2022. p. 11966\u201376. https:\/\/doi.org\/10.48550\/arXiv.2201.03545","DOI":"10.1109\/CVPR52688.2022.01167"},{"key":"pone.0358224.ref028","article-title":"Mamba: linear-time sequence modeling with selective state spaces.","author":"A Gu","year":"2023"},{"key":"pone.0358224.ref029","doi-asserted-by":"crossref","first-page":"1647","DOI":"10.1007\/s42417-022-00471-6","article-title":"Damage detection in rectangular laminated composite plate structures using a combination of wavelet transforms and artificial neural networks","volume":"10","author":"M Saadatmorad","year":"2022","journal-title":"J Vib Eng Technol"},{"issue":"8","key":"pone.0358224.ref030","doi-asserted-by":"crossref","DOI":"10.1371\/journal.pone.0290265","article-title":"A wavelet based data coupling method for spatial damage detection in beam-type structures","volume":"18","author":"J Zhao","year":"2023","journal-title":"PLoS One"},{"key":"pone.0358224.ref031","doi-asserted-by":"crossref","first-page":"646","DOI":"10.1007\/978-3-319-46493-0_39","article-title":"Deep networks with stochastic depth.","author":"G Huang","year":"2016","journal-title":"Computer Vision \u2013 ECCV 2016. Cham: Springer International Publishing"},{"issue":"4","key":"pone.0358224.ref032","first-page":"487","article-title":"An efficient approach for model updating of a large-scale cable-stayed bridge using ambient vibration measurements combined with a hybrid metaheuristic search algorithm","volume":"25","author":"TN Hoa","year":"2020","journal-title":"Smart Struct Syst"},{"key":"pone.0358224.ref033","article-title":"Time series classification from scratch with deep neural networks: a strong baseline.","author":"Z Wang","year":"2016"},{"key":"pone.0358224.ref034","doi-asserted-by":"crossref","unstructured":"Cho K, van Merri\u00ebnboer B, Gulcehre C, Bahdanau D, Bougares F, Schwenk H, et al. Learning phrase representations using RNN encoder\u2013decoder for statistical machine translation. Proceedings of the 2014 Conference on Empirical Methods in Natural Language Processing (EMNLP); 2014. p. 1724\u201334. https:\/\/doi.org\/10.3115\/v1\/D14-1179","DOI":"10.3115\/v1\/D14-1179"},{"issue":"1","key":"pone.0358224.ref035","doi-asserted-by":"crossref","first-page":"447","DOI":"10.1177\/14759217241239041","article-title":"An integrated deep neural network model combining 1D CNN and LSTM for structural health monitoring utilizing multisensor time-series data","volume":"24","author":"M Ahmadzadeh","year":"2024","journal-title":"Struct Health Monit"},{"key":"pone.0358224.ref036","article-title":"A time series is worth 64 words: long-term forecasting with transformers.","author":"Y Nie","year":"2022"},{"key":"pone.0358224.ref037","article-title":"TSMixer: an All-MLP architecture for time series forecasting.","author":"S-A Chen","year":"2023"}],"container-title":["PLOS One"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pone.0358224","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,9,21]],"date-time":"2026-09-21T17:46:15Z","timestamp":1790012775000},"score":1,"resource":{"primary":{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pone.0358224"}},"subtitle":[],"editor":[{"given":"Shih-Lin","family":"Lin","sequence":"first","affiliation":[{"name":"National Changhua University of Education","place":["TAIWAN"]}],"role":[{"vocabulary":"crossref","role":"editor"}]}],"short-title":[],"issued":{"date-parts":[[2026,9,21]]},"references-count":37,"journal-issue":{"issue":"9","published-online":{"date-parts":[[2026,9,21]]}},"URL":"https:\/\/doi.org\/10.1371\/journal.pone.0358224","relation":{},"ISSN":["1932-6203"],"issn-type":[{"value":"1932-6203","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,9,21]]}}}