Wearable edge machine learning with synthetic photoplethysmograms

Article Computer Science, Artificial Intelligence

Implementing ultra-lightweight co-inference model in ubiquitous edge device for atrial fibrillation detection

Jiarong Chen et al.

Summary: Implementing internet of things technologies in health monitoring systems is gaining attention. However, running the model at edge is challenging due to limited computational capacities and storage resources.

EXPERT SYSTEMS WITH APPLICATIONS (2023)

Add to Collection

Article Computer Science, Information Systems

Synthetic PPG Signal Generation to Improve Coronary Artery Disease Classification: Study With Physical Model of Cardiovascular System

Oishee Mazumder et al.

Summary: This study introduces a method of generating synthetic PPG data using a physical model, combined with real data to enhance classifier performance. Results demonstrate its effectiveness in improving CAD classification and reducing the need for a large amount of real data.

IEEE JOURNAL OF BIOMEDICAL AND HEALTH INFORMATICS (2022)

Add to Collection

Article Mathematics

Security Threats and Cryptographic Protocols for Medical Wearables

Luis Hernandez-Alvarez et al.

Summary: This study focuses on the security issues of devices measuring the temperature of diabetic feet, finding that they often lack data security. The research suggests using lightweight cryptographic protocols as a solution.

MATHEMATICS (2022)

Add to Collection

Article Biophysics

Detecting beats in the photoplethysmogram: benchmarking open-source algorithms

Peter H. Charlton et al.

Summary: This study developed a framework to design and test the performance of 15 PPG beat detectors, and evaluated their performance in different use cases. The results showed that eight beat detectors performed well at rest, but had poorer performance during exercise, in neonates, and in atrial fibrillation. The study identified two beat detectors, 'MSPTD' and 'qppg', as the best performers.

PHYSIOLOGICAL MEASUREMENT (2022)

Add to Collection

Article Chemistry, Analytical

Robust PPG Peak Detection Using Dilated Convolutional Neural Networks

Kianoosh Kazemi et al.

Summary: This study proposes a noise-resilient peak detection algorithm for photoplethysmogram (PPG) signals, based on convolutional neural networks (CNNs) with dilated convolutions. Experimental results show that the proposed method can accurately detect PPG peaks in different signal-to-noise ratio (SNR) ranges.

SENSORS (2022)

Add to Collection

Article Computer Science, Information Systems

Non-Contact Atrial Fibrillation Detection From Face Videos by Learning Systolic Peaks

Zhaodong Sun et al.

Summary: We propose a non-contact approach for atrial fibrillation (AF) detection from face videos. Our proposed method accurately extracts systolic peaks from face videos and achieves good performance in classifying healthy versus AF and SR versus AF. The feasibility of non-contact AFL detection is also demonstrated.

IEEE JOURNAL OF BIOMEDICAL AND HEALTH INFORMATICS (2022)

Add to Collection

Article Multidisciplinary Sciences

Synthetic data as an enabler for machine learning applications in medicine

Jean-Francois Rajotte et al.

Summary: Synthetic data generation is a process of training a model using machine learning to generate new or synthetic data that captures patterns in a real dataset. It has the potential to protect privacy and is being increasingly explored in health and life sciences.

ISCIENCE (2022)

Add to Collection

Article Chemistry, Analytical

Deep Learning Techniques in the Classification of ECG Signals Using R-Peak Detection Based on the PTB-XL Dataset

Sandra Smigiel et al.

Summary: This study utilized a convolutional neural network combined with entropy-based features and extracted QRS complexes to explore the classification of ECG signals, demonstrating the effectiveness of this approach in ECG recognition.

SENSORS (2021)

Add to Collection

Editorial Material Biochemistry & Molecular Biology

Remove obstacles to sharing health data with researchers outside of the European Union

Heidi Beate Bentzen et al.

Summary: The COVID-19 pandemic has emphasized the importance of international collaborations and global data sharing in health research. However, legal barriers are currently hindering data sharing for non-pandemic-related research among public researchers worldwide, with potential harmful effects for citizens and patients.

NATURE MEDICINE (2021)

Add to Collection

Editorial Material Engineering, Biomedical

Synthetic data in machine learning for medicine and healthcare

Richard J. Chen et al.

Summary: The proliferation of synthetic data in artificial intelligence for medicine and healthcare has raised concerns about the vulnerabilities of the software and the challenges faced by current policies.

NATURE BIOMEDICAL ENGINEERING (2021)

Add to Collection

Article Computer Science, Theory & Methods

Machine Learning at the Network Edge: A Survey

M. G. Sarwar Murshed et al.

Summary: As resource-constrained IoT devices become ubiquitous, deploying machine learning systems on edge computing devices has become a solution to reduce latency, communication costs, and privacy concerns.

ACM COMPUTING SURVEYS (2021)

Add to Collection

Review Chemistry, Analytical