Data-Driven AI Techniques in Raman Spectroscopy for Biomedical Applications: A Comprehensive Review
2026 14th International Symposium on Digital Forensics and Security (ISDFS), Massachusetts, Amerika Birleşik Devletleri, 19 - 20 Mart 2026, ss.1-6, (Tam Metin Bildiri)
- Yayın Türü: Bildiri / Tam Metin Bildiri
- Doi Numarası: 10.1109/isdfs69419.2026.11458934
- Basıldığı Şehir: Massachusetts
- Basıldığı Ülke: Amerika Birleşik Devletleri
- Sayfa Sayıları: ss.1-6
- Maltepe Üniversitesi Adresli: Evet
Özet
Raman Spectroscopy has emerged as a significant method for biological diagnosis. It emerged because of its ability to provide molecular-level information without the use of labels. The use of machine learning (ML) techniques, especially clustering and deep learning (DL), has strongly improved the analysis of Raman spectra data, thus enhancing the classification and detection of different cancer variants and microbial infections. This study highlights how supervised learning algorithms, such as Support Vector Machines (SVM) and Random Forests, and unsupervised learning algorithms, such as hierarchical, PCA, and K-means clustering, contribute to interpreting and classifying the complex spectral data more accurately. Deep learning models such as CNN is also used to examine their performance in biomarker identification and pattern recognition. This survey paper shows how diagnostic accuracy is improved by merging Raman Spectroscopy with intelligent models, highlighting some key limitations and future directions, including noisy data, high computational demands, and dependence on labeled data. It contributes to a continued development of advanced diagnostic techniques based on spectroscopy and intelligent data analysis.