EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

E-BDL: Enhanced Band-Dependent Learning Framework for Augmented Radar Sensing

Fulin Cai iD, Teresa Wu, Fleming Y M Lure iD

Sensors · 2024

DOI: 10.3390/s24144620 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Sensors, 2024. Verfasst von Fulin Cai und 2 weiteren Autor:innen. Beteiligte Einrichtung: School of Computing and Augmented Intelligence, Arizona State University, Tempe, AZ 85287, USA. Hauptthemen laut MEDLINE: Radar, Deep Learning. Die Arbeit wurde bisher 5-mal zitiert (Crossref) und nennt 31 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: NIH HHS, National Science Foundation, National Institute of Aging.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Radar sensors, leveraging the Doppler effect, enable the nonintrusive capture of kinetic and physiological motions while preserving privacy. Deep learning (DL) facilitates radar sensing for healthcare applications such as gait recognition and vital-sign measurement. However, band-dependent patterns, indicating variations in patterns and power scales associated with frequencies in time–frequency representation (TFR), challenge radar sensing applications using DL. Frequency-dependent characteristics and features with lower power scales may be overlooked during representation learning. This paper proposes an Enhanced Band-Dependent Learning framework (E-BDL) comprising an adaptive sub-band filtering module, a representation learning module, and a sub-view contrastive module to fully detect band-dependent features in sub-frequency bands and leverage them for classification. Experimental validation is conducted on two radar datasets, including gait abnormality recognition for Alzheimer’s disease (AD) and AD-related dementia (ADRD) risk evaluation and vital-sign monitoring for hemodynamics scenario classification. For hemodynamics scenario classification, E-BDL-ResNet achieves competitive performance in overall accuracy and class-wise evaluations compared to recent methods. For ADRD risk evaluation, the results demonstrate E-BDL-ResNet’s superior performance across all candidate models, highlighting its potential as a clinical tool. E-BDL effectively detects salient sub-bands in TFRs, enhancing representation learning and improving the performance and interpretability of DL-based models.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

RadarDeep LearningHumansAlzheimer DiseaseGaitAlgorithmsHemodynamicsVital Signs

Stichwörter der Autor:innen

contrastive learningdeep learningradar sensingspectrogramsub-band

Autor:innen und Institutionen

  • School of Computing and Augmented Intelligence, Arizona State University, Tempe, AZ 85287, USA.
  • ASU-Mayo Center for Innovative Imaging, Arizona State University, Tempe, AZ 85287, USA.
  • MS Technologies Corporation, Rockville, MD 20580, USA.

ORCID-Profile: Fulin Cai, Fleming Y M Lure

Förderung

  • NIH HHS
  • National Science Foundation
  • National Institute of Aging

Zitieren

APA

Cai, F., Wu, T., & Lure, F. Y. M. (2024). E-BDL: Enhanced Band-Dependent Learning Framework for Augmented Radar Sensing. Sensors. https://doi.org/10.3390/s24144620

Vancouver

Cai F, Wu T, Lure FYM. E-BDL: Enhanced Band-Dependent Learning Framework for Augmented Radar Sensing. Sensors. 2024. doi:10.3390/s24144620

BibTeX
@article{Cai2024,
  author = {Cai, Fulin and Wu, Teresa and Lure, Fleming Y M},
  title = {E-BDL: Enhanced Band-Dependent Learning Framework for Augmented Radar Sensing},
  journal = {Sensors},
  year = {2024},
  doi = {10.3390/s24144620}
}
Kostenlos starten →