GPC-net: a generative-augmented predictive-corrective neural network for small-sample FTIR spectral quantification with application to subdural hematoma injury time
Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy · 2027
DOI: 10.1016/j.saa.2026.128495 ↗Kurz gefasst
Fachartikel in der Zeitschrift Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy, 2027. Verfasst von Yusen Wang und 6 weiteren Autor:innen. Beteiligte Einrichtung: School of Basic Medical Sciences, Chongqing Medical University, Chongqing 400016, China. Hauptthemen laut MEDLINE: Neural Networks, Computer, Hematoma, Subdural. Die Arbeit nennt 54 Literaturangaben.
Abstract (Originalsprache)
Quantitative analysis in biomedical Fourier-transform infrared (FTIR) spectroscopy faces a fundamental challenge known as the small sample, high dimensionality paradox. This study introduces GPC-Net, a three-stage neural framework developed for robust spectral quantification with limited data. The framework integrates a conditional variational Transformer autoencoder for generating physically realistic synthetic spectra, a Predictive Neural Network (PNN) pretrained on this augmented data to establish a global spectral-to-target mapping, and a Corrective Neural Network (CNN) trained on real calibration samples to perform sample-specific residual correction. We validated GPC-Net for estimating subdural hematoma (SDH) injury time using a rat model and human clinical samples. The model demonstrated superior predictive accuracy over established baselines such as Partial Least Squares (PLS) regression, Artificial Neural Networks (ANN), Random Forest (RF),XGBoost and One-dimensional convolutional neural network (1D-CNN). Evaluations confirmed the high fidelity of the synthetic spectra. Ablation studies established the necessity of each architectural component. SHAP interpretability analysis showed that model decisions are associated with biochemically relevant spectral regions, including the amide I/II bands, and their temporal contribution patterns align with known hematoma aging pathology. The framework also maintained stable performance under label noise. GPC-Net provides an accurate, interpretable, and consistent across the datasets evaluated methodology for small-sample FTIR quantification in biomedical applications.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- School of Basic Medical Sciences, Chongqing Medical University, Chongqing 400016, China.
- School of Basic Medical Sciences, Chongqing University of Chinese Medicine, Chongqing 402760, China.
- The Third Affiliated Hospital of Chongqing Medical University, Chongqing 401120, China.
- School of Basic Medical Sciences, Chongqing Medical University, Chongqing 400016, China. Electronic address: 102881@cqmu.edu.cn.
Zitieren
APA
Wang, Y., Zhang, L., Zhou, X., Feng, K., Qian, X., Li, J., & Wang, Q. (2027). GPC-net: a generative-augmented predictive-corrective neural network for small-sample FTIR spectral quantification with application to subdural hematoma injury time. Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy. https://doi.org/10.1016/j.saa.2026.128495
Vancouver
Wang Y, Zhang L, Zhou X, Feng K, Qian X, Li J, et al. GPC-net: a generative-augmented predictive-corrective neural network for small-sample FTIR spectral quantification with application to subdural hematoma injury time. Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy. 2027. doi:10.1016/j.saa.2026.128495
BibTeX
@article{Wang2027,
author = {Wang, Yusen and Zhang, Li and Zhou, Xiangyi and Feng, Kunyang and Qian, Xuehong and Li, Jianbo and Wang, Qi},
title = {GPC-net: a generative-augmented predictive-corrective neural network for small-sample FTIR spectral quantification with application to subdural hematoma injury time},
journal = {Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy},
year = {2027},
doi = {10.1016/j.saa.2026.128495}
}
EUVIMED