This paper demonstrates drug classification using a smartphone-based Raman spectrometer. A novel "spectral barcode," a 2D Raman spectral intensity map, is generated and analyzed using a convolutional neural network (CNN). The system achieves 99.0% accuracy in classifying 11 major drug components, even distinguishing between brands. The combination of spectral barcodes with RGB imaging further enhances brand identification accuracy.
Publisher
Nature Communications
Published On
Aug 29, 2023
Authors
Un Jeong Kim, Suyeon Lee, Hyochul Kim, Yeongeun Roh, Seungju Han, Hojung Kim, Yeonsang Park, Seokin Kim, Myung Jin Chung, Hyungbin Son, Hyuck Choo
Tags
Raman spectrometer
drug classification
spectral barcode
convolutional neural network
brand identification
Related Publications
Explore these studies to deepen your understanding of the subject.