Hybrid Machine Learning-Assisted Raman Spectroscopy with Generative Feature Augmentation for Pharmaceutical Identification

基于生成式特征增强的混合机器学习辅助拉曼光谱药物识别技术

Abstract: 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.

摘要： 快速且可靠地识别药物残留对于保障公共卫生、确保食品安全以及实现实用的拉曼光谱筛查至关重要。在本研究中，我们提出了 HyMLRaman，这是一个混合拉曼光谱框架，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、布洛芬和扑热息痛。

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.

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进行编码，其中 EfficientNet-B3 产生了最有效的表示。随后，利用所得的 1536 维嵌入向量，通过分层 10 折交叉验证来训练下游分类器，包括 SVM、KNN、逻辑回归、随机森林、XGBoost 和 ANN。

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.

混合配置 EfficientNet-B3—SVM 实现了最强的基准性能，准确率达到 96.31%，宏观 F1 分数达到 96.36%，优于独立的 CNN 基准模型。为了解决数据受限的情况，我们在经 PCA 降维后的 EfficientNet-B3 潜在空间中引入了一种生成模型——基于 DDPM 的特征增强技术。

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.

小样本消融实验结果表明，DDPM 增强提供了选择性的益处，特别是在训练样本减少的情况下对 KNN 分类器效果显著，且其效果因分类器而异。最后，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的可行性。这些结果表明，HyMLRaman 为快速拉曼药物筛查提供了一条实用且可解释的途径。