BreathAI: Transfer Learning-Based Thermal Imaging for Automated Breathing Pattern Recognition

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Kheddar, Hamza, Himeur, Yassine, Amira, Abbes
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908978232950784
author Kheddar, Hamza
Himeur, Yassine
Amira, Abbes
author_facet Kheddar, Hamza
Himeur, Yassine
Amira, Abbes
contents This study presents an Adaptive Transfer Learning and Thresholding-based Deep Learning Model (ATL-TDLM) for automated breathing pattern recognition using thermal imaging. Unlike conventional methods that rely on sound-based respiratory data, our approach leverages hierarchical deep feature extraction and adaptive multi-thresholding (AMT) to enhance feature segmentation. The model integrates knowledge distillation-based fine-tuning (KD-FT) to optimize learning transfer and contrastive representation learning (CRL) to improve inter-class separability between inhalation (INH) and exhalation (EXH) phases. The ATL-TDLM framework achieves an accuracy of 98.8%, significantly outperforming state-of-the-art models while ensuring computational efficiency. This approach has potential applications in respiratory disorder detection, including sleep apnea and asthma monitoring.
format Preprint
id arxiv_https___arxiv_org_abs_2604_17442
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle BreathAI: Transfer Learning-Based Thermal Imaging for Automated Breathing Pattern Recognition
Kheddar, Hamza
Himeur, Yassine
Amira, Abbes
Image and Video Processing
This study presents an Adaptive Transfer Learning and Thresholding-based Deep Learning Model (ATL-TDLM) for automated breathing pattern recognition using thermal imaging. Unlike conventional methods that rely on sound-based respiratory data, our approach leverages hierarchical deep feature extraction and adaptive multi-thresholding (AMT) to enhance feature segmentation. The model integrates knowledge distillation-based fine-tuning (KD-FT) to optimize learning transfer and contrastive representation learning (CRL) to improve inter-class separability between inhalation (INH) and exhalation (EXH) phases. The ATL-TDLM framework achieves an accuracy of 98.8%, significantly outperforming state-of-the-art models while ensuring computational efficiency. This approach has potential applications in respiratory disorder detection, including sleep apnea and asthma monitoring.
title BreathAI: Transfer Learning-Based Thermal Imaging for Automated Breathing Pattern Recognition
topic Image and Video Processing
url https://arxiv.org/abs/2604.17442