EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Interpretable Multiscale Convolutional Neural Network for Classification and Feature Visualization of Weak Raman Spectra of Biomolecules at Cell Membranes

Che-Lun Chin iD, Chia-En Chang, Ling Chao iD

ACS Sensors · 25.04.2025 · 10 (4) · 2652-2666

DOI: 10.1021/acssensors.4c03260 ↗

Kurz gefasst

Fachartikel in der Zeitschrift ACS Sensors, online erschienen am 04.04.2025. Verfasst von Che-Lun Chin und 2 weiteren Autor:innen. Beteiligte Einrichtung: Department of Chemical Engineering, National Taiwan University, No. 1, Sec. 4, Roosevelt Rd., Taipei 10617, Taiwan. Hauptthemen laut MEDLINE: Spectrum Analysis, Raman, Neural Networks, Computer, Cell Membrane. Die Arbeit wurde bisher 30-mal zitiert (Crossref) und nennt 80 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Taiwan University, National Science and Technology Council.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Raman spectroscopy in biological applications faces challenges due to complex spectra, characterized by peaks of varying widths and significant biological background noise. Convolutional neural networks (CNNs) are widely used for spectrum classification due to their ability to capture local peak features. In this study, we introduce a multiscale CNN designed to detect weak biomolecule signals and differentiate spectra with features that cannot be statistically distinguished. The approach is further enhanced by a new visualization technique tailored for multiscale spectral analysis, providing clear insights into classification results. Using the classification of cholera toxin B subunit (CTB)-treated versus untreated cell membrane samples, whose spectra cannot be statistically differentiated, the optimized multiscale CNN achieved superior performance compared to traditional machine learning methods and existing multiscale CNNs, with accuracy (99.22%), sensitivity (99.27%), specificity (99.16%), and precision (99.20%). Our new visualization method, based on gradients of activation maps with respect to class scores, generates saliency scores that capture sample variations, with decision-making relying on consistently identified peak features. By visualizing the effects of different kernel sizes, Grad-AM highlights features at varying scales, aligning closely with spectral features and enhancing CNN interpretability in complex biomolecular analysis. These advancements demonstrate the potential of our method to improve spectral analysis and reveal previously hidden peaks in complex biological environments.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Spectrum Analysis, RamanNeural Networks, ComputerCell MembraneCholera ToxinHumansConvolutional Neural Networks

Stichwörter der Autor:innen

Raman spectroscopybiomolecular spectraconvolutional neural networks (CNN)interpretablemultiscale

Substanzen

Cholera Toxin

Autor:innen und Institutionen

  • Department of Chemical Engineering, National Taiwan University, No. 1, Sec. 4, Roosevelt Rd., Taipei 10617, Taiwan.

ORCID-Profile: Che-Lun Chin, Ling Chao

Förderung

  • National Taiwan University
  • National Science and Technology Council

Zitieren

APA

Chin, C. L., Chang, C. E., & Chao, L. (2025). Interpretable Multiscale Convolutional Neural Network for Classification and Feature Visualization of Weak Raman Spectra of Biomolecules at Cell Membranes. ACS Sensors, 10(4), 2652-2666. https://doi.org/10.1021/acssensors.4c03260

Vancouver

Chin CL, Chang CE, Chao L. Interpretable Multiscale Convolutional Neural Network for Classification and Feature Visualization of Weak Raman Spectra of Biomolecules at Cell Membranes. ACS Sensors. 2025;10(4):2652-2666. doi:10.1021/acssensors.4c03260

BibTeX
@article{Chin2025,
  author = {Chin, Che-Lun and Chang, Chia-En and Chao, Ling},
  title = {Interpretable Multiscale Convolutional Neural Network for Classification and Feature Visualization of Weak Raman Spectra of Biomolecules at Cell Membranes},
  journal = {ACS Sensors},
  year = {2025},
  volume = {10},
  number = {4},
  pages = {2652--2666},
  doi = {10.1021/acssensors.4c03260}
}
Kostenlos starten →