发表机构
University of Science, Ho Chi Minh City; Vietnam National University, Ho Chi Minh City; University of Health Sciences (UHS), Viet Nam National University Ho Chi Minh City; Frontier Research Institute for Interdisciplinary Sciences, Tohoku University（胡志明市理科大学; 越南国立大学胡志明市分校; 越南国立大学胡志明市分校健康科学大学; 东北大学跨学科前沿研究所）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
本研究提出HyMLRaman混合拉曼光谱框架，结合深度特征提取与生成式增强，实现六种药物的高精度识别，准确率达96.31%。
AI 中文摘要
快速可靠地识别药物残留对于保障公众健康、确保食品安全以及实现基于拉曼光谱的实用筛查具有重要意义。在本研究中，我们提出了HyMLRaman，一种混合拉曼光谱框架，该框架结合了深度光谱特征提取、生成模型和经典机器学习分类器，用于识别六种药物化合物，包括阿莫西林、氯霉素、环丙沙星、四环素、布洛芬和对乙酰氨基酚。拉曼光谱被转换为光谱图像，并使用多种深度神经网络骨干网络进行编码，其中EfficientNet-B3产生了最有效的表示。随后，得到的1536维嵌入被用于训练下游分类器，包括SVM、KNN、逻辑回归、随机森林、XGBoost和ANN，采用分层10折交叉验证。混合EfficientNet-B3-SVM配置达到了最强的基线性能，准确率达到96.31%，宏F1分数为96.36%，优于独立的CNN基线。为了解决数据有限的条件，在PCA降维的EfficientNet-B3潜在空间中引入了生成模型，即基于DDPM的特征增强。低数据消融实验结果表明，DDPM增强提供了选择性益处，特别是对于训练分数减少的KNN，并且其效果仍然依赖于分类器。最后，一个应用层面的拉曼药物分析仪展示了将训练好的模型嵌入交互式拉曼分析工作流程的可行性。这些结果表明，HyMLRaman为快速基于拉曼光谱的药物筛查提供了一条实用且可解释的途径。
英文摘要
Rapid and reliable identification of pharmaceutical residues is important for safeguarding public health, ensuring food safety, and enabling practical Raman-based screening. In this study, we propose HyMLRaman, a hybrid Raman spectroscopy framework that combines deep spectral feature extraction, generative models, and classical machine-learning classifiers to identify six pharmaceutical compounds, including amoxicillin, chloramphenicol, ciprofloxacin, tetracycline, ibuprofen, and paracetamol. Raman spectra are converted into spectral images and encoded with several deep neural-network backbones, among which EfficientNet-B3 yields the most effective representation. The resulting 1536-dimensional embeddings are then used to train downstream classifiers, including SVM, KNN, logistic regression, random forest, XGBoost, and ANN, using stratified 10-fold cross-validation. The hybrid EfficientNet-B3--SVM configuration achieves the strongest baseline performance, reaching 96.31% accuracy and a macro-F1 score of 96.36%, outperforming the standalone CNN baseline. To address limited-data conditions, a generative model, a DDPM-based feature augmentation, is introduced in a PCA-reduced EfficientNet-B3 latent space. The low-data ablation results show that DDPM augmentation provides selective benefits, particularly for KNN with reduced training fractions, and that its effect remains classifier-dependent. Finally, an application-level Raman Pharmaceutical Analyzer demonstrates the feasibility of embedding the trained model into an interactive Raman analysis workflow. These results suggest that HyMLRaman provides a practical and interpretable route for rapid Raman-based pharmaceutical screening.
Comments12 pages, 7 figures, 2 tables
