[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-85839-en":3,"doc-seo-85839-105":30,"detail-sidebar-cat-0-en-105":91},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":20,"is_deleted":4,"is_public":21,"is_downloadable":21,"audit_status":21,"page_count":22,"language":23,"language_code":24,"site_id":25,"html_lang":24,"table_of_contents":26,"faqs":27,"seo_title":13,"seo_description":14,"update_tm":28,"read_time":29},85839,8796095461610,"Oliver","https://ap-avatar.wpscdn.com/davatar_276721f389ce27ea32af1340a28f341c",8,"Research & Report","Generative Augmentation of Raman Spectra for Glioma Classification","Access to sufficiently large biomedical datasets remains a major obstacle for machine learning in Raman spectroscopy-based diagnostics. For glioma analysis, cohorts are typically small and heterogeneous, with acquisition-specific variability and strong within-sample correlations. This study evaluates deep generative augmentation using a β-CVAE that generates class-conditioned synthetic Raman spectra from 58 tumor samples across binary IDH-status and 6-class methylation subtype tasks.","arXiv :2607 . 10 196v 1 [ cs .LG] 11 Jul 2026  \nGenerative Augmentation of Raman Spectra for Glioma Classification  \nAndrei Ius, an∗1, Iulian Vasile∗1, Daria Voiculescu 1 , Ion Petre 1,2 , Andrei P˘aun3,4 , Bogdan Oancea 1,5 , and Mihaela P˘aun 1,5  \n1 National Institute of Research and Development for Biological Sciences, Romania  \n2 Department of Mathematics and Statistics, University of Turku, Finland  \n3 Research Institute for Artificial Intelligence “Mihai  \nDr˘ag˘anescu”, Romanian Academy  \n4 Faculty of Mathematics and Computer Science, University of Bucharest, Romania  \n5 Faculty of Business and Administration, University of  \nBucharest, Romania  \nAbstract  \nAccess to sufficiently large biomedical datasets remains a major obstacle for machine learning in Raman spectroscopy-based diagnostics. In particular, for glioma analysis, datasets are typically small and heterogeneous, affected by acquisition-specific variability. This work investigates the utility of deep generative augmentation in such a small-cohort setting. We analyze glioma biopsy spectra acquired from 58 tumor samples and consider both binary IDH-status classification and 6-class methylation subtype classification problems. To address the limited size and imbalance of the dataset, we develop a  \n∗ These authors contributed equally to this work.  \nconditional variational autoencoder (β-CVAE) capable of generating class-conditioned synthetic Raman spectra. The generated data are evaluated in Train-on-Synthetic, Test-on-Real (TS/TR) and Trainon-Synthetic+Real, Test-on-Real (TSR/TR) settings under a strict patient-isolated cross-validation protocol. Models trained exclusively on synthetic data underperform models trained on real spectra, indicating a substantial domain gap between synthetic and real distributions. However, augmenting the real training data with synthetic spectra consistently improves classification performance across multiple models. These findings indicate that, even with a limited number of independent patient samples, generative models can capture sufficient structure to provide useful regularization for downstream classifiers. We also investigate a reconstruction-based inference strategy, termed Classification by Reconstruction (CbR), in which class prediction is based on reconstruction error under different class conditions.  \nOverall, the results support the use of deep generative augmentation as a practical strategy for improving machine learning robustness in Raman spectroscopy applications characterized by limited biomedical datasets.  \n1 Introduction  \nRaman spectroscopy (RS) has emerged as a promising tool for biomedical diagnostics due to its ability to provide rapid, label-free, and non-destructive biochemical characterization of tissue [11] . In oncology, RS has attracted particular interest as a potential support tool for tumor grading and molecular characterization, where subtle biochemical differences between tissue types may be reflected in spectral signatures [4] . In the context of gliomas, accurate molecular characterization is essential for treatment planning and prognosis, motivating the development of automated RS-based classification pipelines [12, 15] .  \nDespite this potential, the development of reliable machine learning models for biomedical diagnostics remains limited by the availability of sufficiently large and representative datasets [3] . In biomedical Raman spectroscopy, this limitation is notably severe because studies often involve only tens of patients, substantial acquisition variability, and strong correlations between spectra originating from the same biopsy sample [15] . A typical Raman spectrum may contain more than one thousand spectral bands, while the number  \nof independent patient samples remains relatively small [15] . This highdimensional, low-sample system substantially increases the risk of overfitting and reduces the robustness and generalizability of discriminative models [3, 16] .  \nThe probl","cbCaic4AhqsTtNjO","https://ap.wps.com/l/cbCaic4AhqsTtNjO","pdf",2079854,2,1,38,"English","en",105,"# Abstract\n# Introduction\n# Background","[{\"question\":\"Why is generative augmentation useful for Raman-based glioma classification?\",\"answer\":\"Biomedical Raman datasets are often small and heterogeneous, with acquisition-specific variability that can cause overfitting and domain gaps. The study uses generative augmentation to improve robustness by regularizing downstream classifiers with synthetic class-conditioned spectra.\"},{\"question\":\"How is the synthetic data generated and conditioned in this work?\",\"answer\":\"The method trains a conditional variational autoencoder (β-CVAE) to generate class-conditioned synthetic Raman spectra, enabling augmentation aligned with target labels for classification.\"},{\"question\":\"What evaluation protocols are used to measure the impact of synthetic augmentation?\",\"answer\":\"Models are evaluated under Train-on-Synthetic, Test-on-Real (TS/TR) and Train-on-Synthetic+Real, Test-on-Real (TSR/TR) settings using a strict patient-isolated cross-validation protocol to avoid data leakage.\"}]",1784206631,96,{"code":4,"msg":31,"data":32},"ok",{"site_id":25,"language":24,"slug":33,"title":13,"keywords":34,"description":14,"schema_data":35,"social_meta":86,"head_meta":88,"extra_data":90,"updated_unix":28},"generative-augmentation-of-raman-spectra-for-glioma-classification","",{"@graph":36,"@context":85},[37,53,68],{"@type":38,"itemListElement":39},"BreadcrumbList",[40,44,47,50],{"item":41,"name":42,"@type":43,"position":21},"https://docshare.wps.com","Home","ListItem",{"item":45,"name":46,"@type":43,"position":20},"https://docshare.wps.com/document/","Document",{"item":48,"name":12,"@type":43,"position":49},"https://docshare.wps.com/document/research-report/",3,{"item":51,"name":13,"@type":43,"position":52},"https://docshare.wps.com/document/generative-augmentation-of-raman-spectra-for-glioma-classification/85839/",4,{"url":51,"name":13,"@type":54,"author":55,"headline":13,"publisher":57,"fileFormat":60,"inLanguage":24,"description":14,"dateModified":61,"datePublished":62,"encodingFormat":60,"isAccessibleForFree":63,"interactionStatistic":64},"DigitalDocument",{"name":9,"@type":56},"Person",{"url":41,"name":58,"@type":59},"DocShare","Organization","application/pdf","2026-07-24","2026-07-16",true,{"@type":65,"interactionType":66,"userInteractionCount":20},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71,77,81],{"name":72,"@type":73,"acceptedAnswer":74},"Why is generative augmentation useful for Raman-based glioma classification?","Question",{"text":75,"@type":76},"Biomedical Raman datasets are often small and heterogeneous, with acquisition-specific variability that can cause overfitting and domain gaps. The study uses generative augmentation to improve robustness by regularizing downstream classifiers with synthetic class-conditioned spectra.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"How is the synthetic data generated and conditioned in this work?",{"text":80,"@type":76},"The method trains a conditional variational autoencoder (β-CVAE) to generate class-conditioned synthetic Raman spectra, enabling augmentation aligned with target labels for classification.",{"name":82,"@type":73,"acceptedAnswer":83},"What evaluation protocols are used to measure the impact of synthetic augmentation?",{"text":84,"@type":76},"Models are evaluated under Train-on-Synthetic, Test-on-Real (TS/TR) and Train-on-Synthetic+Real, Test-on-Real (TSR/TR) settings using a strict patient-isolated cross-validation protocol to avoid data leakage.","https://schema.org",{"og:url":51,"og:type":87,"og:title":13,"og:site_name":58,"og:description":14},"article",{"robots":89,"canonical":51},"index,follow",{"doc_id":7,"site_id":25},{"code":4,"msg":5,"data":92},[93,97,101,105,110,115,120,123,128,131,135],{"id":21,"doc_module":4,"doc_module_name":46,"category_name":94,"show_sort_weight":95,"slug":96},"Story & Novel",90,"story-novel",{"id":20,"doc_module":4,"doc_module_name":46,"category_name":98,"show_sort_weight":99,"slug":100},"Literature",80,"literature",{"id":52,"doc_module":4,"doc_module_name":46,"category_name":102,"show_sort_weight":103,"slug":104},"Exam",70,"exam",{"id":106,"doc_module":4,"doc_module_name":46,"category_name":107,"show_sort_weight":108,"slug":109},5,"Comic",60,"comic",{"id":111,"doc_module":4,"doc_module_name":46,"category_name":112,"show_sort_weight":113,"slug":114},6,"Technology",50,"technology",{"id":116,"doc_module":4,"doc_module_name":46,"category_name":117,"show_sort_weight":118,"slug":119},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":121,"slug":122},30,"research-report",{"id":124,"doc_module":4,"doc_module_name":46,"category_name":125,"show_sort_weight":126,"slug":127},9,"Religion & Spirituality",20,"religion-spirituality",{"id":126,"doc_module":4,"doc_module_name":46,"category_name":129,"show_sort_weight":126,"slug":130},"World Cup","world-cup",{"id":132,"doc_module":4,"doc_module_name":46,"category_name":133,"show_sort_weight":132,"slug":134},10,"Lifestyle","lifestyle",{"id":136,"doc_module":4,"doc_module_name":46,"category_name":137,"show_sort_weight":106,"slug":138},19,"General","general"]