科研速览 · Science Skim继续刷下去 · Keep skimming →
◆ Spectrochimica acta. Part A, Molecular and biomolecular spectroscopy2026-09-21

Plasma ATR-FTIR spectroscopy combined with machine learning for nasopharyngeal carcinoma classification.

Rock Christian Tomas, Yohsuke Suzuki, Gerard Mathew Magno, Michael Lee, Javier Alfonso Millan, Keziah Nepomuceno, Mariella Cielo Cobarrubias, Michael Benedict Mejia, Teresa Sy-Ortin, Pia Marie Albano

一句话结论 · In one sentence

Plasma ATR-FTIR spectroscopy combined with machine learning showed moderate discrimination between NPC and clinically healthy controls, with the highest performance obtained using low-rank spectral features and a feedforward neural network. Larger independent cohorts are needed for validation.

原始摘要(英文原文)· Original abstract
BACKGROUND: Nasopharyngeal carcinoma (NPC) is often diagnosed at advanced stages, creating a need for minimally invasive approaches to support detection. This study evaluated plasma attenuated total reflectance Fourier transform infrared (ATR-FTIR) spectroscopy, combined with machine learning, to distinguish NPC from clinically healthy individuals. METHODS: Plasma samples from 51 histologically confirmed NPC cases and 51 age- and sex-matched clinically healthy controls were analyzed across 4000-600 cm-1. Differences at 21 spectral peaks were assessed using the Mann-Whitney U test. Seven machine-learning algorithms were evaluated using the full spectrum, fingerprint region, selected peaks, and low-rank spectral representations with repeated cross-validation and exploratory age- and sex-stratified analyses. RESULTS: Thirteen peaks showed significant between-group differences in median absorbance (p < 0.01). Despite substantial overlap in the original spectral distributions, model performance in the complete cohort increased when selected peaks and low-rank spectral representations were used: the best case AUCs were 0.6303 ± 0.0384 for the full spectrum, 0.6330 ± 0.0333 for the fingerprint region, 0.7066 ± 0.0452 for selected peaks, and 0.7404 ± 0.0422 for the low-rank representation. The neural network model achieved the best overall performance using low-rank features, with an accuracy (ACC) of 0.7115 ± 0.0456. Exploratory subgroup estimates varied across age- and sex-defined subsets. However, due to limited subgroup sizes and variability in the performance estimates, interpretation and generalizability was cautioned. CONCLUSION: Plasma ATR-FTIR spectroscopy combined with machine learning showed moderate discrimination between NPC and clinically healthy controls, with the highest performance obtained using low-rank spectral features and a feedforward neural network. Larger independent cohorts are needed for validation.
读原文 · Read the paper ↗

AI 追问PRO

登录后使用 AI 追问

讨论区

登录后参与讨论

相关论文 · Related

Plasma ATR-FTIR spectroscopy combined with machine learning for nasopharyngeal carcinoma classification. — 科研速览 Science Skim