EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Molecular Remodeling of Milk Fat Globules Induced by Centrifugation: Insights from Deep Learning-Based Detection of Milk Adulteration

Grzegorz Gwardys, Grzegorz Grodkowski iD, Piotr Kostusiak iD, Wojciech Mendelowski, Jan Slósarz, Michał Satława, Bartłomiej Śmietanka, Krzysztof Gwardys, Marcin Gołębiewski iD, Kamila Puppel iD

International Journal of Molecular Sciences · 10.12.2025 · 26 (24) · 11919

DOI: 10.3390/ijms262411919 ↗

Kurz gefasst

Fachartikel in der Zeitschrift International Journal of Molecular Sciences, online erschienen am 10.12.2025. Verfasst von Grzegorz Gwardys und 9 weiteren Autor:innen. Beteiligte Einrichtung: Promity sp. z o.o., ul. Wiejska 14/25, 00-490 Warsaw, Poland. Hauptthemen laut MEDLINE: Deep Learning, Lipid Droplets, Milk, Glycolipids. Die Arbeit wurde bisher 5-mal zitiert (Crossref) und nennt 35 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: NCBR, Poland, European Regional Development Fund.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Milk adulteration through centrifugation, which artificially reduces the somatic cell count (SCC), represents a significant challenge to food authenticity and public health. This fraudulent practice alters the native molecular architecture of milk, masking inflammatory conditions such as subclinical mastitis and distorting product quality. Conventional analytical and microscopic techniques remain insufficiently sensitive to detect the subtle physicochemical changes associated with centrifugation, highlighting the need for molecular-level, data-driven diagnostics. The dataset included 128 paired raw milk samples and approximately 25,000 bright-field micrographs acquired across multiple microscopes, of which 95% were confirmed to be of high quality. In this study, advanced machine learning (ML) and deep learning (DL) approaches were applied to identify centrifugation-induced alterations in raw milk microstructure. Bright-field micrographs (pixel size 0.27 µm) of paired unprocessed and centrifuged samples were obtained under standardized optical conditions and analyzed using convolutional neural networks (ResNet-18/50, Inception-v3, Xception, NasNet-Mobile) and hybrid attention architectures (MaxViT, CoAtNet). Model performance was evaluated using the harmonic average of recalls across five micrographs per sample (HAR5). Human microscopy experts (n = 4) achieved only 18% classification accuracy—below the random baseline (25%)—confirming that centrifugation-induced modifications are not visually discernible. In contrast, DL architectures reached up to 97% accuracy (HAR5, Xception), successfully identifying subtle molecular cues. Class activation and sensitivity analyses indicated that models focused not on milk fat globule (MFG) boundaries but on high-frequency nanoscale variations related to the reorganization of casein micelles and solid non-fat fractions. The findings strongly suggest that centrifugation adulteration constitutes a molecular reorganization event rather than a morphological alteration. The integration of optical microscopy with AI-driven molecular analytics establishes deep learning as a precise and objective tool for detecting fraudulent milk processing and improving food integrity diagnostics.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Deep LearningLipid DropletsMilkGlycolipidsGlycoproteinsFood ContaminationAnimalsCentrifugationHumansNeural Networks, Computer

Stichwörter der Autor:innen

artificial intelligencecasein micellescentrifugationdeep learningfood adulterationmachine learningmicroscopymilksomatic cell count

Substanzen

milk fat globuleGlycolipidsGlycoproteins

Autor:innen und Institutionen

  • Promity sp. z o.o., ul. Wiejska 14/25, 00-490 Warsaw, Poland.
  • Institute of Animal Science, Warsaw University of Life Sciences, Ciszewskiego 8, 02-786 Warsaw, Poland.

ORCID-Profile: Grzegorz Grodkowski, Piotr Kostusiak, Marcin Gołębiewski, Kamila Puppel

Förderung

  • NCBR, Poland
  • European Regional Development Fund

Zitieren

APA

Gwardys, G., Grodkowski, G., Kostusiak, P., Mendelowski, W., Slósarz, J., Satława, M., Śmietanka, B., Gwardys, K., Gołębiewski, M., & Puppel, K. (2025). Molecular Remodeling of Milk Fat Globules Induced by Centrifugation: Insights from Deep Learning-Based Detection of Milk Adulteration. International Journal of Molecular Sciences, 26(24), 11919. https://doi.org/10.3390/ijms262411919

Vancouver

Gwardys G, Grodkowski G, Kostusiak P, Mendelowski W, Slósarz J, Satława M, et al. Molecular Remodeling of Milk Fat Globules Induced by Centrifugation: Insights from Deep Learning-Based Detection of Milk Adulteration. International Journal of Molecular Sciences. 2025;26(24):11919. doi:10.3390/ijms262411919

BibTeX
@article{Gwardys2025,
  author = {Gwardys, Grzegorz and Grodkowski, Grzegorz and Kostusiak, Piotr and Mendelowski, Wojciech and Slósarz, Jan and Satława, Michał and Śmietanka, Bartłomiej and Gwardys, Krzysztof and Gołębiewski, Marcin and Puppel, Kamila},
  title = {Molecular Remodeling of Milk Fat Globules Induced by Centrifugation: Insights from Deep Learning-Based Detection of Milk Adulteration},
  journal = {International Journal of Molecular Sciences},
  year = {2025},
  volume = {26},
  number = {24},
  pages = {11919},
  doi = {10.3390/ijms262411919}
}
Kostenlos starten →