{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,23]],"date-time":"2026-09-23T05:05:51Z","timestamp":1790139951446,"version":"4.0.1"},"reference-count":42,"publisher":"MDPI AG","issue":"18","license":[{"start":{"date-parts":[[2026,9,19]],"date-time":"2026-09-19T00:00:00Z","timestamp":1789776000000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"DOI":"10.13039\/501100000780","name":"European Union","doi-asserted-by":"publisher","award":["101132308"],"award-info":[{"award-number":["101132308"]}],"id":[{"id":"10.13039\/501100000780","id-type":"DOI","asserted-by":"publisher"}]},{"award":["101132308"],"award-info":[{"award-number":["101132308"]}],"id":[{"id":"https:\/\/ror.org\/019w4f821","id-type":"ROR","asserted-by":"publisher"}]},{"name":"Ram\u00f3n y Cajal Fellowship","award":["RYC2023-044886-I"],"award-info":[{"award-number":["RYC2023-044886-I"]}]},{"name":"MICIU\/AEI"},{"name":"FSE+"},{"name":"Consejo Superior de Investigaciones Cient\u00edficas","award":["202650E113"],"award-info":[{"award-number":["202650E113"]}],"id":[{"id":"https:\/\/ror.org\/02gfc7t72","id-type":"ROR","asserted-by":"crossref"}]},{"award":["202650E113"],"award-info":[{"award-number":["202650E113"]}],"id":[{"id":"https:\/\/ror.org\/02gfc7t72","id-type":"ROR","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Sensors"],"abstract":"<jats:p>Cultural heritage (CH) sites are continuously exposed to pressures that can lead to deterioration. Continuous monitoring and anomaly detection (AD) enable early damage detection for preventive conservation. We propose a late-fusion AD framework combining NST-Net, a nested spatiotemporal neural network, with a semantic module encoding expert conservation rules, and evaluate it on five real-world CH monitoring datasets. NST-Net outperforms six state-of-the-art baselines on most sites, achieving, on average, 42% higher Average Precision than the best baseline per site under a synthetic evaluation protocol emphasizing sharp, short-duration anomalies. Dataset length serves as an important performance factor: the combined architecture and preprocessing pipeline particularly benefit longer campaigns, whereas they add little value on shorter ones. NST-Net also requires approximately 13 times less peak memory than baseline detectors, at a higher but still millisecond-scale latency. Further analysis shows that performance depends strongly on anomaly density, type, and severity. The semantic module passes label-free sanity checks and complements NST-Net effectively (complementarity index &gt; 0.87), identifying conservation-relevant events missed by the deep model. These findings support late-fusion statistical\u2013semantic frameworks for AD in CH monitoring.<\/jats:p>","DOI":"10.3390\/s26185931","type":"journal-article","created":{"date-parts":[[2026,9,22]],"date-time":"2026-09-22T10:21:48Z","timestamp":1790072508000},"page":"5931","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["Nested Spatiotemporal Anomaly Detection with Semantic Augmentation: A Case Study in Heritage Conservation"],"prefix":"10.3390","volume":"26","author":[{"ORCID":"https:\/\/orcid.org\/0009-0007-1558-8184","authenticated-orcid":false,"given":"Lidia","family":"Abad","sequence":"first","affiliation":[{"name":"Institute of Physical and Information Technologies \u201cLeonardo Torres Quevedo\u201d (ITEFI), Consejo Superior de Investigaciones Cient\u00edficas (CSIC), C\/Serrano 144, 28006 Madrid, Spain"}],"role":[{"vocabulary":"crossref","role":"author"},{"role":"conceptualization","vocabulary":"credit"},{"role":"methodology","vocabulary":"credit"},{"role":"software","vocabulary":"credit"},{"role":"validation","vocabulary":"credit"},{"role":"investigation","vocabulary":"credit"},{"role":"data-curation","vocabulary":"credit"},{"role":"formal-analysis","vocabulary":"credit"},{"role":"writing-original-draft","vocabulary":"credit"},{"role":"writing-review-editing","vocabulary":"credit"},{"role":"visualization","vocabulary":"credit"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-7517-0894","authenticated-orcid":false,"given":"Fernando","family":"Ramonet","sequence":"additional","affiliation":[{"name":"Institute of Physical and Information Technologies \u201cLeonardo Torres Quevedo\u201d (ITEFI), Consejo Superior de Investigaciones Cient\u00edficas (CSIC), C\/Serrano 144, 28006 Madrid, Spain"}],"role":[{"vocabulary":"crossref","role":"author"},{"role":"conceptualization","vocabulary":"credit"},{"role":"methodology","vocabulary":"credit"},{"role":"validation","vocabulary":"credit"},{"role":"formal-analysis","vocabulary":"credit"},{"role":"investigation","vocabulary":"credit"},{"role":"writing-original-draft","vocabulary":"credit"},{"role":"writing-review-editing","vocabulary":"credit"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-6075-593X","authenticated-orcid":false,"given":"Javier","family":"Ortega","sequence":"additional","affiliation":[{"name":"Institute of Physical and Information Technologies \u201cLeonardo Torres Quevedo\u201d (ITEFI), Consejo Superior de Investigaciones Cient\u00edficas (CSIC), C\/Serrano 144, 28006 Madrid, Spain"}],"role":[{"vocabulary":"crossref","role":"author"},{"role":"conceptualization","vocabulary":"credit"},{"role":"methodology","vocabulary":"credit"},{"role":"validation","vocabulary":"credit"},{"role":"investigation","vocabulary":"credit"},{"role":"writing-review-editing","vocabulary":"credit"},{"role":"writing-original-draft","vocabulary":"credit"},{"role":"formal-analysis","vocabulary":"credit"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-2415-471X","authenticated-orcid":false,"given":"Jos\u00e9 Javier","family":"Anaya","sequence":"additional","affiliation":[{"name":"Institute of Physical and Information Technologies \u201cLeonardo Torres Quevedo\u201d (ITEFI), Consejo Superior de Investigaciones Cient\u00edficas (CSIC), C\/Serrano 144, 28006 Madrid, Spain"}],"role":[{"vocabulary":"crossref","role":"author"},{"role":"conceptualization","vocabulary":"credit"},{"role":"methodology","vocabulary":"credit"},{"role":"validation","vocabulary":"credit"},{"role":"resources","vocabulary":"credit"},{"role":"investigation","vocabulary":"credit"},{"role":"formal-analysis","vocabulary":"credit"},{"role":"writing-original-draft","vocabulary":"credit"},{"role":"writing-review-editing","vocabulary":"credit"},{"role":"supervision","vocabulary":"credit"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-4069-015X","authenticated-orcid":false,"given":"Sof\u00eda","family":"Aparicio","sequence":"additional","affiliation":[{"name":"Institute of Physical and Information Technologies \u201cLeonardo Torres Quevedo\u201d (ITEFI), Consejo Superior de Investigaciones Cient\u00edficas (CSIC), C\/Serrano 144, 28006 Madrid, Spain"}],"role":[{"vocabulary":"crossref","role":"author"},{"role":"conceptualization","vocabulary":"credit"},{"role":"methodology","vocabulary":"credit"},{"role":"validation","vocabulary":"credit"},{"role":"formal-analysis","vocabulary":"credit"},{"role":"investigation","vocabulary":"credit"},{"role":"resources","vocabulary":"credit"},{"role":"writing-original-draft","vocabulary":"credit"},{"role":"writing-review-editing","vocabulary":"credit"},{"role":"supervision","vocabulary":"credit"},{"role":"project-administration","vocabulary":"credit"},{"role":"funding-acquisition","vocabulary":"credit"}]}],"member":"1968","published-online":{"date-parts":[[2026,9,19]]},"reference":[{"key":"ref_1","doi-asserted-by":"crossref","first-page":"172","DOI":"10.1186\/s40494-022-00802-6","article-title":"Practices and challenges of cultural heritage conservation in historical and religious heritage sites: Evidence from North Shoa Zone, Amhara Region, Ethiopia","volume":"10","author":"Mekonnen","year":"2022","journal-title":"npj Herit. Sci."},{"key":"ref_2","doi-asserted-by":"crossref","unstructured":"Fahmy, A., and Thamarat, M. (2025). How Community Engagement Approach Enhances Heritage Conservation: Two Case Studies on Sustainable Urban Development in Historic Cairo. Sustainability, 17.","DOI":"10.3390\/su17104565"},{"key":"ref_3","doi-asserted-by":"crossref","first-page":"111906","DOI":"10.1016\/j.jobe.2025.111906","article-title":"Navigating Complexity in Sustainable Conservation: A Multi-Criteria Decision Making of Architectural Heritage in Urbanizing China","volume":"102","author":"Cao","year":"2025","journal-title":"J. Build. Eng."},{"key":"ref_4","doi-asserted-by":"crossref","unstructured":"Ghini, E., Marra, M., Palermo, M., Gasparini, G., Silvestri, S., and Trombetti, T. (2025). Physically-Based Data-Driven Approach for Anomaly Detection in Long-Term Continuous Monitoring Systems: Application to a Case Study. Struct. Health Monit.","DOI":"10.12783\/shm2025\/37552"},{"key":"ref_5","doi-asserted-by":"crossref","first-page":"536","DOI":"10.1016\/j.culher.2024.01.005","article-title":"Artificial intelligence-assisted visual inspection for cultural heritage: State-of-the-art review","volume":"66","author":"Mishra","year":"2024","journal-title":"J. Cult. Herit."},{"key":"ref_6","doi-asserted-by":"crossref","first-page":"012005","DOI":"10.1088\/1742-6596\/1131\/1\/012005","article-title":"Computational methods for wireless structural health monitoring of cultural heritages","volume":"1131","author":"Bertolli","year":"2018","journal-title":"Proc. J. Phys. Conf. Ser."},{"key":"ref_7","doi-asserted-by":"crossref","unstructured":"Wang, F., Jiang, Y., Zhang, R., Wei, A., Xie, J., and Pang, X. (2025). A Survey of Deep Anomaly Detection in Multivariate Time Series: Taxonomy, Applications, and Directions. Sensors, 25.","DOI":"10.3390\/s25010190"},{"key":"ref_8","first-page":"7194","article-title":"Towards a Rigorous Evaluation of Time-Series Anomaly Detection","volume":"36","author":"Kim","year":"2022","journal-title":"Proc. AAAI Conf. Artif. Intell."},{"key":"ref_9","unstructured":"Abad, L., Ramonet, F., Ortega, J., Gonz\u00e1lez, M., Anaya, J.J., and Aparicio, S. (2026, January 7\u20139). Bridging Anomaly Detection and Expert Knowledge in Cultural Heritage Conservation. Proceedings of the 2nd IEEE International Conference on Cyber Humanities (IEEE-CH), Venice, Italy. Available online: https:\/\/www.ieee-ch.org\/conference-program\/."},{"key":"ref_10","unstructured":"European Commission (2025, December 30). ARGUS: Non-Destructive, Scalable, Smart Monitoring of Remote Cultural Treasures. CORDIS Project 101132308. Available online: https:\/\/cordis.europa.eu\/project\/id\/101132308."},{"key":"ref_11","unstructured":"Carrara, F., Falchi, F., Girardi, M., Messina, N., Padovani, C., and Pellegrini, D. (2022). Deep learning for structural health monitoring: An application to heritage structures. arXiv."},{"key":"ref_12","doi-asserted-by":"crossref","unstructured":"Wu, Y., Dong, Y., Shan, Z., Meng, X., He, Y., Jia, P., and Lu, D. (2024). Enhancing anomaly detection for cultural heritage via long short-term memory with attention mechanism. Electronics, 13.","DOI":"10.3390\/electronics13071254"},{"key":"ref_13","doi-asserted-by":"crossref","unstructured":"Su, Y., Zhao, Y., Niu, C., Liu, R., Sun, W., and Pei, D. (2019). Robust Anomaly Detection for Multivariate Time Series through Stochastic Recurrent Neural Network. KDD \u201919: Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining, Association for Computing Machinery.","DOI":"10.1145\/3292500.3330672"},{"key":"ref_14","unstructured":"Xu, J., Wu, H., Wang, J., and Long, M. (2022). Anomaly Transformer: Time Series Anomaly Detection with Association Discrepancy. arXiv."},{"key":"ref_15","doi-asserted-by":"crossref","unstructured":"Tuli, S., Casale, G., and Jennings, N.R. (2022). TranAD: Deep Transformer Networks for Anomaly Detection in Multivariate Time Series Data. arXiv.","DOI":"10.14778\/3514061.3514067"},{"key":"ref_16","doi-asserted-by":"crossref","unstructured":"Alves, B., Pinho, A.J., and Gouveia, S. (2026). Unified Taxonomy for Multivariate Time Series Anomaly Detection Using Deep Learning. arXiv.","DOI":"10.1007\/s10462-026-11660-0"},{"key":"ref_17","first-page":"5953","article-title":"Multivariate time-series anomaly detection via temporal convolutional and graph attention networks","volume":"44","author":"He","year":"2023","journal-title":"J. Intell. Fuzzy Syst. Appl. Eng. Technol."},{"key":"ref_18","doi-asserted-by":"crossref","unstructured":"Zhao, H., Wang, Y., Duan, J., Huang, C., Cao, D., Tong, Y., Xu, B., Bai, J., Tong, J., and Zhang, Q. (2020). Multivariate Time-Series Anomaly Detection via Graph Attention Network. Proceedings of the 2020 IEEE International Conference on Data Mining (ICDM), IEEE.","DOI":"10.1109\/ICDM50108.2020.00093"},{"key":"ref_19","doi-asserted-by":"crossref","unstructured":"Zhang, H., Wang, Y., and Han, Q. (2024). USTG:Multivariate Time Series Anomaly Detection via Unsupervised Spatial-Temporal Graph Learning. Proceedings of the 2024 11th International Conference on Behavioural and Social Computing (BESC), IEEE.","DOI":"10.1109\/BESC64747.2024.10780512"},{"key":"ref_20","doi-asserted-by":"crossref","first-page":"11802","DOI":"10.1109\/TNNLS.2023.3325667","article-title":"Correlation-Aware Spatial\u2013Temporal Graph Learning for Multivariate Time-Series Anomaly Detection","volume":"35","author":"Zheng","year":"2024","journal-title":"IEEE Trans. Neural Netw. Learn. Syst."},{"key":"ref_21","doi-asserted-by":"crossref","unstructured":"Zhang, Y., Kong, L., Antwi-Afari, M.F., and Zhang, Q. (2024). An Integrated Method Using a Convolutional Autoencoder, Thresholding Techniques, and a Residual Network for Anomaly Detection on Heritage Roof Surfaces. Buildings, 14.","DOI":"10.3390\/buildings14092828"},{"key":"ref_22","doi-asserted-by":"crossref","unstructured":"Rodr\u00edguez, D., L\u00f3pez, A., Jim\u00e9nez, J.R., Garrido del Solo, A.J., Feito-Higueruela, F., and Jurado, J.M. (2024). SAM-Based Detection of Structural Anomalies in 3D Models for Preserving Cultural Heritage. Proceedings of the 19th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications, SciTePress.","DOI":"10.5220\/0012464000003660"},{"key":"ref_23","doi-asserted-by":"crossref","unstructured":"Bhattacharya, D., Mukherjee, S., Kamanchi, C., Ekambaram, V., Jati, A., and Dayama, P. (2024). Towards Unbiased Evaluation of Time-series Anomaly Detector. arXiv.","DOI":"10.1109\/ICASSP49660.2025.10890568"},{"key":"ref_24","doi-asserted-by":"crossref","unstructured":"Atzmueller, M., Bohne, T., and Windler, P. (2024). Knowledge-Augmented Explainable and Interpretable Learning for Anomaly Detection and Diagnosis. arXiv.","DOI":"10.3233\/FAIA250234"},{"key":"ref_25","unstructured":"Ahmed, F., and Courville, A. (2019). Detecting semantic anomalies. arXiv."},{"key":"ref_26","doi-asserted-by":"crossref","unstructured":"Caspary, J., Rebmann, A., and van der Aa, H. (2023). Does This Make Sense? Machine Learning-Based Detection of Semantic Anomalies in Business Processes. Proceedings of the Business Process Management, Springer.","DOI":"10.1007\/978-3-031-41620-0_10"},{"key":"ref_27","doi-asserted-by":"crossref","first-page":"80","DOI":"10.1179\/2047058412Y.0000000073","article-title":"Damage functions in heritage science","volume":"58","author":"Thickett","year":"2013","journal-title":"Stud. Conserv."},{"key":"ref_28","doi-asserted-by":"crossref","first-page":"101681","DOI":"10.1016\/j.iot.2025.101681","article-title":"An interpretable deep learning framework for intrusion detection in industrial Internet of Things","volume":"33","author":"Ahmad","year":"2025","journal-title":"Internet Things"},{"key":"ref_29","unstructured":"Kim, T., Kim, J., Tae, Y., Park, C., Choi, J.H., and Choo, J. (2022, January 25\u201329). Reversible Instance Normalization for Accurate Time-Series Forecasting against Distribution Shift. Proceedings of the International Conference on Learning Representations, Virtual Event."},{"key":"ref_30","unstructured":"Clevert, D.A., Unterthiner, T., and Hochreiter, S. (2015). Fast and accurate deep network learning by exponential linear units (elus). arXiv."},{"key":"ref_31","unstructured":"Zhang, B., and Sennrich, R. (2019, January 8\u201314). Root Mean Square Layer Normalization. Proceedings of the Advances in Neural Information Processing Systems 32 (NeurIPS 2019), Vancouver, BC, Canada."},{"key":"ref_32","unstructured":"Glorot, X., and Bengio, Y. (2010). Understanding the difficulty of training deep feedforward neural networks. Proceedings of the Thirteenth International Conference on Artificial Intelligence and Statistics, PMLR (Proceedings of Machine Learning Research)."},{"key":"ref_33","unstructured":"Kingma, D.P., and Ba, J. (2014). Adam: A method for stochastic optimization. arXiv."},{"key":"ref_34","doi-asserted-by":"crossref","unstructured":"Welsing, A., Nienk\u00f6tter, A., and Jiang, X. (2021, January 21\u201322). Exponential Weighted Moving Average of Time Series in Arbitrary Spaces with Application to Strings. Proceedings of the Structural, Syntactic, and Statistical Pattern Recognition: Joint IAPR International Workshops, S+SSPR 2020, Padua, Italy.","DOI":"10.1007\/978-3-030-73973-7_5"},{"key":"ref_35","doi-asserted-by":"crossref","first-page":"764","DOI":"10.1016\/j.jesp.2013.03.013","article-title":"Detecting outliers: Do not use standard deviation around the mean, use absolute deviation around the median","volume":"49","author":"Leys","year":"2013","journal-title":"J. Exp. Soc. Psychol."},{"key":"ref_36","doi-asserted-by":"crossref","first-page":"273","DOI":"10.1037\/met0000079","article-title":"Comparing the Pearson and Spearman correlation coefficients across distributions and sample sizes: A tutorial using simulations and empirical data","volume":"21","author":"Gosling","year":"2016","journal-title":"Psychol. Methods"},{"key":"ref_37","doi-asserted-by":"crossref","unstructured":"Liu, F.T., Ting, K.M., and Zhou, Z.H. (2008). Isolation Forest. Proceedings of the IEEE International Conference on Data Mining (ICDM), IEEE.","DOI":"10.1109\/ICDM.2008.17"},{"key":"ref_38","unstructured":"Audibert, J., Michiardi, P., Guyard, F., Marti, S., and Zuluaga, M.A. (2020). USAD: UnSupervised Anomaly Detection on Multivariate Time Series. KDD \u201920: Proceedings of the 26th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining, Association for Computing Machinery."},{"key":"ref_39","doi-asserted-by":"crossref","first-page":"4027","DOI":"10.1609\/aaai.v35i5.16523","article-title":"Graph neural network-based anomaly detection in multivariate time series","volume":"Volume 35","author":"Deng","year":"2021","journal-title":"Proceedings of the AAAI Conference on Artificial Intelligence"},{"key":"ref_40","unstructured":"Wu, H., Hu, T., Liu, Y., Zhou, H., Wang, J., and Long, M. (2023). TimesNet: Temporal 2D-Variation Modeling for General Time Series Analysis. arXiv."},{"key":"ref_41","doi-asserted-by":"crossref","first-page":"405","DOI":"10.1146\/annurev-statistics-030718-104938","article-title":"Statistical aspects of Wasserstein distances","volume":"6","author":"Panaretos","year":"2019","journal-title":"Annu. Rev. Stat. Its Appl."},{"key":"ref_42","doi-asserted-by":"crossref","first-page":"74","DOI":"10.1016\/j.paid.2016.06.069","article-title":"Effect size guidelines for individual differences researchers","volume":"102","author":"Gignac","year":"2016","journal-title":"Personal. Individ. Differ."}],"container-title":["Sensors"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/1424-8220\/26\/18\/5931\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,9,23]],"date-time":"2026-09-23T04:32:06Z","timestamp":1790137926000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/1424-8220\/26\/18\/5931"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,9,19]]},"references-count":42,"journal-issue":{"issue":"18","published-online":{"date-parts":[[2026,9]]}},"alternative-id":["s26185931"],"URL":"https:\/\/doi.org\/10.3390\/s26185931","relation":{},"ISSN":["1424-8220"],"issn-type":[{"value":"1424-8220","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,9,19]]}}}