Stacked BiLSTM-Adaboost Collaborative Model: Construction of a Precision Analysis Model for GABA and Vitamin B9 in the Foxtail Millet

Abstract

Amidst health-conscious consumption trends, functional foods rich in gamma-aminobutyric acid (GABA) and vitamin B9 are gaining prominence. Foxtail millet, naturally abundant in these compounds, plays vital roles in nervous system regulation, cellular metabolism, and fetal development. In this work, we introduced hyperspectral imaging for detecting these nutrients in millet, collecting spectral data from 190 samples across 19 varieties. Four "coarse-fine" wavelength screening strategies were innovatively proposed: interval spectral variables initially identified wavelengths with strong explanatory power, followed by model cluster analysis for precise optimization. A stacked BiLSTM-Adaboost model was built for efficient prediction. Results revealed significant variations in GABA and vitamin B9 content among varieties, offering crucial insights for breeding and quality assessment. The GABA prediction model achieved R2=0.5421, RMSE=13.9349, RPD=1.4778; the vitamin B9 model attained R2=0.8932, RMSE=4.6676, RPD=3.0600. This integration of technology and methodological innovation provides a novel pathway for rapid, non-destructive millet nutrient analysis, facilitating intelligent quality control, industry advancement, and functional food development.

Plant phenotyping relevance

コムギではなくフォックステイルミレットの種子中GABA・ビタミンB9という植物器官の形質を、ハイパースペクトル画像と予測モデルで非破壊推定する方法の開発・評価が研究の中心である。

Code and data availability

The paper's hyperspectral imaging data, GABA/vitamin B9 measurements, and model results are not publicly deposited; the Data Availability Statement says the dataset is available only on request from the authors. No public URL, repository, or code deposit is provided.

No evidence-backed public reproduction asset is currently recorded.