EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Self-Supervised Stacked Masked Denoising Autoencoder (S2MDAE) for Brain MRI Denoising and Feature Learning

Rui Li iD, Puyu Zhang iD, Chenglin Wen, Guoxi Sun

Sensors · 2026

DOI: 10.3390/s26175388 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Sensors, 2026. Verfasst von Rui Li und 3 weiteren Autor:innen. Beteiligte Einrichtung: College of Electronic Information Engineering, Guangdong University of Petrochemical Technology, Maoming 525000, China. Hauptthemen laut MEDLINE: Magnetic Resonance Imaging, Brain, Image Processing, Computer-Assisted. Die Arbeit nennt 26 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Key Research and Development Program, Guangdong University of Petrochemical Technology, National Key Research and Development Program of China.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Medical-image annotation is costly and limits the use of fully supervised learning. This study proposes the Self-Supervised Stacked Masked Denoising Autoencoder (S2MDAE) for four-class brain MRI classification. During pre-training, three convolutional encoder–decoder blocks perform layer-wise masked reconstruction under a 75% mask ratio and Gaussian corruption; the pretrained encoder stack is then fine-tuned for classification. On the primary dataset, S2MDAE achieved 87.207% Accuracy, while ablation studies supported the contributions of pre-training, masking, noise injection, and layer-wise reconstruction. Representation analysis further showed increased class separability in deeper blocks after fine-tuning. Without retraining, external evaluation on a duplicate-screened independent dataset achieved 67.254% Accuracy and 66.009% Macro-F1, indicating partial transfer under domain shift. These findings support the task-specific value of the proposed framework while showing that broader clinical generalization requires further validation.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Magnetic Resonance ImagingBrainImage Processing, Computer-AssistedAutoencoderHumansAlgorithmsSignal-To-Noise RatioConvolutional Neural Networks

Stichwörter der Autor:innen

brain MRIdenoising autoencoderfeature learningmasked image modelingself-supervised learning

Autor:innen und Institutionen

  • College of Electronic Information Engineering, Guangdong University of Petrochemical Technology, Maoming 525000, China.
  • Provincial Key Laboratory of Petrochemical Equipment Fault Diagnosis, Guangdong University of Petrochemical Technology, Maoming 525000, China.

ORCID-Profile: Rui Li, Puyu Zhang

Förderung

  • National Key Research and Development Program
  • Guangdong University of Petrochemical Technology
  • National Key Research and Development Program of China

Zitieren

APA

Li, R., Zhang, P., Wen, C., & Sun, G. (2026). Self-Supervised Stacked Masked Denoising Autoencoder (S2MDAE) for Brain MRI Denoising and Feature Learning. Sensors. https://doi.org/10.3390/s26175388

Vancouver

Li R, Zhang P, Wen C, Sun G. Self-Supervised Stacked Masked Denoising Autoencoder (S2MDAE) for Brain MRI Denoising and Feature Learning. Sensors. 2026. doi:10.3390/s26175388

BibTeX
@article{Li2026,
  author = {Li, Rui and Zhang, Puyu and Wen, Chenglin and Sun, Guoxi},
  title = {Self-Supervised Stacked Masked Denoising Autoencoder (S2MDAE) for Brain MRI Denoising and Feature Learning},
  journal = {Sensors},
  year = {2026},
  doi = {10.3390/s26175388}
}
Kostenlos starten →