Search research articles

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by

Same author

Automated Brain Tumor Detection Using Convolutional Neural Network.

Biotechnology and applied biochemistry·2025

Same author

Hesperidin Attenuates Chronic Stress-Induced Depression via 5-HT2A-Linked Modulation of Neurochemical, Oxidative, and Inflammatory Pathways: Experimental and In Silico Evidence.

Neurochemical research·2025

Same author

Assessment of Garbage Enzyme as a Bioremediation Method for the Wastewater Treatment.

Biotechnology and applied biochemistry·2025

Same author

A review on efforts for improvement in medicinally important chemical constituents in<i>Aconitum</i> through biotechnological interventions.

3 Biotech·2023

Same author

Role of matrix metalloproteinase in wound healing.

American journal of translational research·2022

Same author

Comparative Analysis of Waste Materials for Their Potential Utilization in Green Concrete Applications.

Materials (Basel, Switzerland)·2022

Same journal

Development and Validation of an ELISA Method for Quantification of HuB01 in Fermentation Products.

Biotechnology and applied biochemistry·2026

Same journal

Optimization and Characterization of Poly-γ-Glutamic Acid Production From a Locally Isolated Bacillus cereus ARM24.

Biotechnology and applied biochemistry·2026

Same journal

Isolation and Characterization of a Novel Microbacterium takaoensis Lipase for Nutritional and Textural Improvement of Pizza Dough.

Biotechnology and applied biochemistry·2026

Same journal

Targeting Amastigote and Trypomastigote Phases: Multi-Epitope Vaccine Strategy Against Trypanosoma cruzi.

Biotechnology and applied biochemistry·2026

Same journal

Novel Inhibitors Targeting Class B1 Metallo-β-Lactamases in Carbapenem-Resistant Acinetobacter baumannii.

Biotechnology and applied biochemistry·2026

Same journal

Formulation of PLA/Chitosan Copolymeric Nanocarriers Enhanced the Sustained and Targeted Drug Delivery of Cinnamic Acid: In Vitro Drug Optimization and In Vivo Biodistribution in Rat Model.

Biotechnology and applied biochemistry·2026

See all related articles

ABOUT JoVE

Overview Leadership Blog JoVE Help Center

AUTHORS

Publishing Process Editorial Board Scope & Policies Peer Review FAQ Submit

LIBRARIANS

Testimonials Subscriptions Access Resources Library Advisory Board FAQ

RESEARCH

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments Archive

EDUCATION

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual Faculty Resource Center Faculty Site

Terms & Conditions of Use

Search research articles

Related Experiment Video

Updated: Jan 14, 2026

Author Spotlight: Addressing Technical and Subjective Challenges in Measuring Classroom Attention

Author Spotlight: Addressing Technical and Subjective Challenges in Measuring Classroom Attention

Published on: December 15, 2023

A Novel Domain Adaptation Framework for Wearable Human Activity Recognition Using Multi-Sensor Feature Alignment.

Prawar Chaudhary¹, Chintan Singh², Roobal Chaudhary³

¹School of Basic and Applied Sciences, K. R. Mangalam University, Gurugram, Haryana, India.

Biotechnology and Applied Biochemistry

|January 13, 2026

Summary

This summary is machine-generated.

This study introduces the Multi-Sensor Adaptive Feature Alignment Network (MSAFAN) to improve wearable human activity recognition (HAR) across different users and sensors. The new model enhances accuracy and generalization while reducing computational costs for edge AI applications.

Keywords:

MSAFAN human activity recognition modeling multi‐sensor signal

More Related Videos

Evaluation of a Smartphone-based Human Activity Recognition System in a Daily Living Environment

Evaluation of a Smartphone-based Human Activity Recognition System in a Daily Living Environment

Published on: December 11, 2015

Design and Analysis for Fall Detection System Simplification

Design and Analysis for Fall Detection System Simplification

Published on: April 6, 2020

Related Experiment Videos

Last Updated: Jan 14, 2026

Author Spotlight: Addressing Technical and Subjective Challenges in Measuring Classroom Attention

Author Spotlight: Addressing Technical and Subjective Challenges in Measuring Classroom Attention

Published on: December 15, 2023

Evaluation of a Smartphone-based Human Activity Recognition System in a Daily Living Environment

Evaluation of a Smartphone-based Human Activity Recognition System in a Daily Living Environment

Published on: December 11, 2015

Design and Analysis for Fall Detection System Simplification

Design and Analysis for Fall Detection System Simplification

Published on: April 6, 2020

Area of Science:

Wearable computing
Artificial Intelligence
Biomedical Engineering

Background:

Human Activity Recognition (HAR) models face performance degradation due to domain shifts across users and sensor placements.
Existing Unsupervised Domain Adaptation (UDA) models struggle with robust cross-sensor generalization in wearable HAR.

Purpose of the Study:

To develop a novel adaptive network, the Multi-Sensor Adaptive Feature Alignment Network (MSAFAN), for robust cross-sensor generalization in wearable HAR.
To address domain shifts and improve the performance and efficiency of HAR models in real-world wearable applications.

Main Methods:

Integration of Sensor-Specific Normalization Layer (SSNL) for sensor-wise adaptation.
Application of Hybrid Polynomial Feature Transformation (HPFT) and Conditional Alignment Loss (CAL) for feature alignment.
Utilization of Entropy-Guided Pseudo-Labeling (EGPL) for enhanced class-wise adaptation and generalization.

Main Results:

MSAFAN demonstrated significant improvements, increasing macro-F1 score by 8.4% and accuracy by 10.3% across four benchmark datasets.
The framework achieved a 26% reduction in computational cost compared to state-of-the-art UDA models.
Stable convergence, efficient adaptation, and scalable performance were observed.

Conclusions:

MSAFAN offers a robust solution for cross-sensor generalization in wearable HAR, effectively mitigating domain shift challenges.
The proposed framework's efficiency and scalability make it suitable for real-time deployment in edge AI and wearable computing.
This research advances the field of wearable HAR by providing a more adaptable and computationally efficient model.