Hiring in life sciences? Share your open positions with our professional community. Read more Close

Advertisement

Hyperspectral imaging and artificial intelligence for frozen-thawed meat assessment: Mechanistic basis, model robustness, and industrial translation.

Created on 19 Nov 2026

Authors

Yiqi Fang, Yifan Lv, Jing Yang, Ming Huang, Kunjie Chen, Jingxin Sun, Jichao Huang

Published in

Food research international (Ottawa, Ont.). Volume 246. Pages 121076. Dec 31, 2026. Epub Nov 02, 2026.

Abstract

Freezing and thawing disrupt muscle microstructure, redistribute water, and promote pigment, protein, and lipid deterioration, while subsequent post-thaw storage can permit microbial spoilage, producing coupled optical changes that can be captured by hyperspectral imaging (HSI). This review critically evaluates how HSI combined with artificial intelligence and chemometrics has been used to authenticate frozen-thawed meat and assess deterioration, with emphasis on physicochemical mechanisms, validation strength, and industrial requirements. A targeted narrative search covered primary spectral-spatial imaging studies of fresh, frozen, frozen-stored, and frozen-thawed meat, supplemented by adjacent-domain evidence on model transfer and deployment. The literature shows a progression from binary authentication to freeze-thaw-history identification, water-state and oxidation prediction, multi-task learning, integrated quality indices, and pixel-level visualization. These advances confirm that HSI can characterize the combined effects of hydration, pigment chemistry, tissue scattering, oxidation, and exudation under controlled conditions. However, model development has advanced faster than evidence of generalizability. Many studies rely on limited biological populations and internal data splits; large numbers of pixels or patches do not replace independent specimens, and selected wavelengths or explanation maps do not establish physicochemical causality. Cross-batch, temporal, cross-instrument, and cross-laboratory evaluations remain uncommon. Full-spectrum HSI is particularly useful for identifying candidate spectral regions, but their value for reduced-band sensing depends on whether they remain stable across independent biological and instrumental conditions. For practical use, external transfer, long-term stability, throughput, and decision-specific error provide more informative evidence than further improvements obtained within a single calibration dataset.

PMID:
43051777
Bibliographic data and abstract were imported from PubMed on 19 Nov 2026.

Read full publication at:
Please sign in to see all details.

Advertisement

Stats

  • Community rating n/a 0 votes
  • Reviewers' rating n/a 0 votes
  • Your rating

1-terrible, 9-excellent. How would you rate this publication? Sign in in to submit your rating.

  • Recommendations n/a n/a positive of 0 vote(s)
  • Views 1
  • Comments 0

Recommended by

  • No recommendations yet.

Post a comment

You need to be signed in to post comments. You can sign in here.

Comments

There are no comments yet.

Advertisement