PulseDB: A large, cleaned dataset based on MIMIC-III and VitalDB for benchmarking cuff-less blood pressure estimation methods

Frontiers in Digital Health, Journal Year: 2023, Volume and Issue: 4

Published: Feb. 8, 2023

There has been a growing interest in developing cuff-less blood pressure (BP) estimation methods to enable continuous BP monitoring from electrocardiogram (ECG) and/or photoplethysmogram (PPG) signals. The majority of these have evaluated using publicly-available datasets, however, there exist significant discrepancies across studies with respect the size, number subjects, and applied pre-processing steps for data that is eventually used training testing models. Such differences make conducting performance comparison models largely unfair, mask generalization capability various methods. To fill this important gap, paper presents “PulseDB,” largest cleaned dataset date, benchmarking also fulfills requirements standardized protocols. PulseDB contains 1) 5,245,454 high-quality 10 -s segments ECG, PPG, arterial (ABP) waveforms 5,361 subjects retrieved MIMIC-III waveform database matched subset VitalDB database; 2) subjects’ identification demographic information, can be utilized as additional input features improve models, or evaluate generalizability unseen subjects; 3) positions characteristic points ECG/PPG signals, making directly usable deep learning minimal pre-processing. Additionally, dataset, we conduct first study provide insights about gap between calibration-based calibration-free approaches evaluating We expect PulseDB, user-friendly, large, comprehensive multi-functional reliable source evaluation

Language: Английский

---

Challenges and prospects of visual contactless physiological monitoring in clinical study

npj Digital Medicine, Journal Year: 2023, Volume and Issue: 6(1)

Published: Dec. 15, 2023

The monitoring of physiological parameters is a crucial topic in promoting human health and an indispensable approach for assessing status diagnosing diseases. Particularly, it holds significant value patients who require long-term or with underlying cardiovascular disease. To this end, Visual Contactless Physiological Monitoring (VCPM) capable using videos recorded by consumer camera to monitor blood volume pulse (BVP) signal, heart rate (HR), respiratory (RR), oxygen saturation (SpO2) pressure (BP). Recently, deep learning-based pipelines have attracted numerous scholars achieved unprecedented development. Although VCPM still emerging digital medical technology presents many challenges opportunities, has the potential revolutionize clinical medicine, health, telemedicine as well other areas. viable solution that can be integrated into these systems measuring vital during video consultation, owing its merits contactless measurement, cost-effectiveness, user-friendly passive sole requirement off-the-shelf camera. In fact, studies technologies been rocketing recently, particularly AI-based approaches, but few are employed settings. Here we provide comprehensive overview applications, challenges, prospects from perspective settings AI first time. thorough exploration analysis scenarios will profound guidance research development

Language: Английский

---

Arterial stiffness assessment using PPG feature extraction and significance testing in an in vitro cardiovascular system

Scientific Reports, Journal Year: 2024, Volume and Issue: 14(1)

Published: Jan. 23, 2024

Abstract Cardiovascular diseases (CVDs) remain the leading cause of global mortality, therefore understanding arterial stiffness is essential to developing innovative technologies detect, monitor and treat them. The ubiquitous spread photoplethysmography (PPG), a completely non-invasive blood-volume sensing technology suitable for all ages, highlights immense potential assessment in wider healthcare setting outside specialist clinics, example during routine visits General Practitioner or even at home with use mobile wearable health devices. This study employs custom-manufactured vitro cardiovascular system vessels varying test hypothesis that PPG signals may be used detect assess level under controlled conditions. Analysis various morphological features demonstrated significant (p < 0.05) correlations vessel stiffness. Particularly, area related were closely linked red signals, while infrared most correlated pulse-width. demonstrates utility custom investigations work towards using PPG, valuable tool applications clinical healthcare, devices beyond.

Language: Английский

---

Cuffless Blood Pressure in clinical practice: challenges, opportunities and current limits.

Blood Pressure, Journal Year: 2024, Volume and Issue: 33(1)

Published: Jan. 21, 2024

Background: Cuffless blood pressure measurement technologies have attracted significant attention for their potential to transform cardiovascular monitoring.

Language: Английский

---

Recent Advances in Non-Invasive Blood Pressure Monitoring and Prediction Using a Machine Learning Approach

Sensors, Journal Year: 2022, Volume and Issue: 22(16), P. 6195 - 6195

Published: Aug. 18, 2022

Blood pressure (BP) monitoring can be performed either invasively via arterial catheterization or non-invasively through a cuff sphygmomanometer. However, for conscious individuals, traditional cuff-based BP devices are often uncomfortable, intermittent, and impractical frequent measurements. Continuous non-invasive (NIBP) is currently gaining attention in the human health area due to its promising potentials assessing status of an individual, enabled by machine learning (ML), various purposes such as early prediction disease intervention treatment. This review presents development measuring tool called sphygmomanometer brief, summarizes state-of-the-art NIBP sensors, identifies extended works on continuous using commercial devices. Moreover, predictive techniques including pulse arrival time, transit wave velocity, ML elaborated basis bio-signals acquisition from these sensors. Additionally, different values (systolic BP, diastolic mean pressure) models adopted several reported studies compared terms international validation standards developed Advancement Medical Instrumentation (AAMI) British Hypertension Society (BHS) clinically-approved monitors. Finally, challenges possible solutions implementation realization technology addressed.

Language: Английский

---

A Deep Learning Approach for Atrial Fibrillation Classification Using Multi-Feature Time Series Data from ECG and PPG

Diagnostics, Journal Year: 2023, Volume and Issue: 13(14), P. 2442 - 2442

Published: July 21, 2023

Atrial fibrillation is a prevalent cardiac arrhythmia that poses significant health risks to patients. The use of non-invasive methods for AF detection, such as Electrocardiogram and Photoplethysmogram, has gained attention due their accessibility ease use. However, there are challenges associated with ECG-based the significance PPG signals in this context been increasingly recognized. limitations ECG untapped potential taken into account work attempts classify non-AF using time series data deep learning. In work, we emploted hybrid neural network comprising 1D CNN BiLSTM task classification. We addressed under-researched area applying learning transmissive by proposing novel approach. Our approach involved integrating multi-featured training models model achieved an accuracy 95% on test identifying atrial fibrillation, showcasing its strong performance reliable predictive capabilities. Furthermore, evaluated our additional metrics. precision classification was measured at 0.88, indicating ability accurately identify true positive cases AF. recall, or sensitivity, 0.85, illustrating model's capacity detect high proportion actual cases. Additionally, F1 score, which combines both calculated 0.84, highlighting overall effectiveness classifying

Language: Английский

---

Ultralow-Power Photoplethysmography (PPG) Sensors: A Methodological Review

IEEE Sensors Journal, Journal Year: 2023, Volume and Issue: 23(15), P. 16467 - 16480

Published: June 14, 2023

Photoplethysmography (PPG) sensors are used to accurately, instantaneously, and noninvasively measure vital signs provide a real-time indication of overall physical health long-term well-being. Achieving continuous monitoring is an important requirement increase user safety diagnostic accuracy. PPG need light-emitting diode (LED) with sufficient output power detect the signal, which consumes tens milliwatts. On other hand, low ac/dc ratios < 0.1%–4%, ambient light, motion artifacts, semiconductor noise greatly affect signal-to-noise ratio (SNR), dynamic range (DR), signal quality. Specialized circuit blocks needed cancel these interferences, further increasing consumption. Several ultralow-power designs, techniques, sampling schemes have been proposed in literature extend sensors' lifetime. This article reviews, analyzes, critiques solutions designers comprehensive design considerations for achieving ultralow consumption while required SNR DR sensor design.

Language: Английский

---

Intelligent cardiovascular disease diagnosis using deep learning enhanced neural network with ant colony optimization

Scientific Reports, Journal Year: 2024, Volume and Issue: 14(1)

Published: Sept. 18, 2024

Language: Английский

---

Sensitivity and Specificity of Wearables for Atrial Fibrillation in Elderly Populations: A Systematic Review

Current Cardiology Reports, Journal Year: 2023, Volume and Issue: 25(7), P. 761 - 779

Published: May 24, 2023

Language: Английский

---

Leveraging Regression Analysis to Predict Overlapping Symptoms of Cardiovascular Diseases

IEEE Access, Journal Year: 2023, Volume and Issue: 11, P. 60254 - 60266

Published: Jan. 1, 2023

In medical informatics, deep learning-based models are being used to predict and diagnose cardiovascular diseases (CVDs). These can detect clinical signs, recognize phenotypes, pick treatment methods for complicated illnesses. One approach predicting CVDs is collect a large dataset of patient records use it train learning model. This study investigated early prediction using regression analysis on 2621 from UAE hospitals, including age, symptoms, CVD information. We propose long short-term memory-based neural network by leveraging the analysis. It be seen that accuracy level increased when they were simulated in pairs one disease with another due overlapping symptoms. The study's results suggest coronary heart has been predicted an 71.5% level, 84.4% Dyspnea; measured combination three conditions was 86.7%, Dyspnea, Chest Pain, Cyanosis, up 88.9%. Weakness, Fatigue, Emptysis showed value 89.8%. our proposed work, combinations Weakness Emptysis, discomfort pressure chest have shown ideal 90.6%, Fever, 91%. show effectiveness method several evaluation benchmarks.

Language: Английский

---