Transitions in growing networks using a structural complexity approach DOI Creative Commons

A. A. Snarskiı̆

Physical review. E, Journal Year: 2024, Volume and Issue: 110(5)

Published: Nov. 21, 2024

Structure changes or transitions are common in growing networks (complex networks, graphs, etc.) and must be precisely determined. The introduced quantitative measure of the structural complexity network based on a procedure similar to renormalization process allows one reveal such changes. proposed concept accounts for difference between actual averaged structures different scales corresponds qualitative comprehension complexity. can found weighted also. complexities various types exhibiting phase were found-the deterministic infinite finite size artificial natures including percolation structures, time series cardiac rhythms mapped complex using parametric visibility graph algorithm. In all cases reaches maximum near transition point: formation giant component at threshold two-dimensional three-dimensional square lattices when cluster having fractal structure has emerged. Therefore, us detect study processes second-order networks. node serve as kind centrality index, auxiliary, generalization local clustering coefficient. Such an index provides another new ranking manner nodes. Being easily computable measure, might help features systems real world.

Language: Английский

Wearable Sensors as a Preoperative Assessment Tool: A Review DOI Creative Commons
Aron Syversen, Alexios Dosis, David Jayne

et al.

Sensors, Journal Year: 2024, Volume and Issue: 24(2), P. 482 - 482

Published: Jan. 12, 2024

Surgery is a common first-line treatment for many types of disease, including cancer. Mortality rates after general elective surgery have seen significant decreases whilst postoperative complications remain frequent occurrence. Preoperative assessment tools are used to support patient risk stratification but do not always provide precise and accessible assessment. Wearable sensors (WS) an alternative that offers continuous monitoring in non-clinical setting. They shown consistent uptake across the perioperative period there has been no review WS as preoperative tool. This paper reviews developments research application period. Accelerometers were consistently employed frequently combined with photoplethysmography or electrocardiography sensors. Pre-processing methods discussed missing data was theme; this dealt several ways, commonly by employing extraction threshold using imputation techniques. Research rarely processed raw data; commercial devices employ internal proprietary algorithms pre-calculated heart rate step count most limiting further feature extraction. A range machine learning models predict outcomes vector machines, random forests regression models. No individual model clearly outperformed others. Deep proved successful predicting exercise testing only within large sample-size studies. outlines challenges provides recommendations future develop viable

Language: Английский

Citations

13
KDPhys: An attention guided 3D to 2D knowledge distillation for real-time video-based physiological measurement DOI
Nicky Nirlipta Sahoo,

V. S. Sachidanand,

Matcha Naga Gayathri

et al.

Biomedical Signal Processing and Control, Journal Year: 2025, Volume and Issue: 107, P. 107797 - 107797

Published: March 15, 2025

Language: Английский

Citations

0
Opening the envelope: Efficient envelope-based PPG denoising algorithm DOI Creative Commons
George R.E. Bradley, P. A. Kyriacou

Biomedical Signal Processing and Control, Journal Year: 2023, Volume and Issue: 88, P. 105693 - 105693

Published: Nov. 8, 2023

Photoplethysmography (PPG) signals obtained from the skin's surface offer valuable insights into blood volume fluctuations. With rising interest in continuous non-invasive physiological monitoring, PPG has garnered significant attention. However, are often affected by various forms of noise, impeding reliable feature extraction. Robust data pre-processing approaches vital for both retrospective and real-time analysis. Existing denoising methods, including recent machine learning techniques, suffer implementation challenges, computational inefficiency, limited interpretability. Addressing this challenge, we propose a novel algorithm. The algorithm was evaluated using dataset representing approximately 81,015.99 min or 1360.27 h collected 31 patients. evaluation involved calculation analysis five key metrics: Signal-to-Noise Ratio (SNR), Variance, Total Variation (TV), Shannon entropy, Instances-per-second (IPS). Our results demonstrate notable increase SNR after denoising, indicating effective noise reduction while preserving signal content. Variance TV values showed post-denoising, suggesting smoother less variable signals, validating suppression efficacy. Additionally, entropy exhibited decrease successful enhanced regularity. nonparametric Wilcoxon signed-rank test (a = 0.05) employed to assess statistical significance observed differences these metrics before denoising. Furthermore, speed revealed EPDA's potential efficient processing large datasets applications. This comprehensive approach allows thorough understanding effectiveness data, fostering advancements monitoring promoting broader adoption PPG-based healthcare technologies.

Language: Английский

Citations

6
Pulse wave signal preprocessing based on improved threshold DOI

Hengjun Zhu,

lihao Ma

Published: July 22, 2024

Language: Английский

Citations

1
Study of Arrhythmia Classification Algorithms on Electrocardiogram Using Deep Learning DOI Creative Commons

Rezki Fauzan Arifin,

Satria Mandala

SinkrOn, Journal Year: 2023, Volume and Issue: 8(3), P. 1753 - 1760

Published: July 20, 2023

Arrhythmia is a heart disease that occurs due to disturbance in the heartbeat causes rhythm become irregular. In some cases, arrhythmias can be life-threatening if not detected immediately. The method used detect electrocardiogram (ECG) signal analysis. To avoid misdiagnosis by cardiologists and ease workload, methods are proposed classify utilizing Artificial Intelligence (AI). recent years, there has been lot of research on detection this disease. However, many such studies more likely use machine learning algorithms classification process, most accuracy results still do reach optimal levels general. Therefore, study aims using deep algorithms. There several stages performing arrhythmia detection, namely, preprocessing, feature extraction, classification. focus only stage, where Long Short-Term Memory (LSTM) algorithm proposed. After going through series experiments, performance further analyzed compare accuracy, specificity, sensitivity with other based previous research, aim obtaining an for detection. Based study, managed outperform 98.47%, 99.24%, 97.67%, respectively.

Language: Английский

Citations

1
Analysis of Electrocardiogram Dynamic Features for Arrhythmia Classification DOI Creative Commons

Yusril Ramadhan,

Satria Mandala

Jurnal Online Informatika, Journal Year: 2023, Volume and Issue: 8(2), P. 204 - 212

Published: Dec. 28, 2023

Arrhythmia is a deviation from the normal heart rate pattern. Arrhythmias are usually harmless, but they can cause problems. Some types of arrhythmias include Atrial Fibrillation (AF), Premature Contractions (PAC), and Ventricular (PVC). Many studies have been conducted to identify dynamic characteristics electrocardiogram (ECG) irregular waves in detection arrhythmias. However, accuracy obtained these less than optimal. This study aims solve problem by evaluating three main features using ECG signals: RR interval, PR QRS complex. Experiments were rigorously on features. The achieved was 98.21%, with specificity 98.65% sensitivity 97.37%.

Language: Английский

Citations

1
Assessment of Emotion Elicitation using Multimodal Physiological Sensors and Phase Synchronization DOI
Sourabh Banik, Himanshu Kumar, Nagarajan Ganapathy

et al.

IEEE Sensors Letters, Journal Year: 2024, Volume and Issue: 8(8), P. 1 - 4

Published: Aug. 1, 2024

Language: Английский

Citations

0
A Wavelet Based Hybrid Method for Time Interval Series Determining DOI
Galya Georgieva-Tsaneva, Krasimir Cheshmedzhiev, Penio Lebamovski

et al.

Published: June 14, 2024

Language: Английский

Citations

0
Transitions in growing networks using a structural complexity approach DOI Creative Commons

A. A. Snarskiı̆

Physical review. E, Journal Year: 2024, Volume and Issue: 110(5)

Published: Nov. 21, 2024

Structure changes or transitions are common in growing networks (complex networks, graphs, etc.) and must be precisely determined. The introduced quantitative measure of the structural complexity network based on a procedure similar to renormalization process allows one reveal such changes. proposed concept accounts for difference between actual averaged structures different scales corresponds qualitative comprehension complexity. can found weighted also. complexities various types exhibiting phase were found-the deterministic infinite finite size artificial natures including percolation structures, time series cardiac rhythms mapped complex using parametric visibility graph algorithm. In all cases reaches maximum near transition point: formation giant component at threshold two-dimensional three-dimensional square lattices when cluster having fractal structure has emerged. Therefore, us detect study processes second-order networks. node serve as kind centrality index, auxiliary, generalization local clustering coefficient. Such an index provides another new ranking manner nodes. Being easily computable measure, might help features systems real world.

Language: Английский

Citations

0