Respiratory Signal Research Articles

OC-0779 Reconstruction of lung tumour volumes from 4D-MRI: the use of a tumour-derived respiratory signal

OC-0779 Reconstruction of lung tumour volumes from 4D-MRI: the use of a tumour-derived respiratory signal

Integrating medical rules to assist attention for sleep apnea detection

AbstractSleep apnea is one of the most common sleep disorders. The consequences of undiagnosed sleep apnea can be very serious, increasing the risk of high blood pressure, heart disease, stroke, and Alzheimer’s disease over a long period of time. However, many people are often unaware of their condition. The gold standard for diagnosing sleep apnea is nighttime polysomnography monitoring in a specialized sleep laboratory. However, these diagnoses are expensive and the number of beds is limited, and there is insufficient monitoring in terms of time dimension. Existing methods for automated detection use no more than three physiological signals, but all other signals are also associated with the patient’s sleep. In addition, the limited amount of medical real annotation data, especially abnormal samples, lead to weak model generalization capability. The gap between model generalization capability and medical field needs still exists. In this paper, we propose a method for integrating medical interpretation rules into a long short-term memory neural network based on self-attention with multichannel respiratory signals as input. We obtain attention weights through a token-level attention mechanism and then extract key rules of medical interpretation to assist the weights, improving model generalization and reducing the dependence on data volume. Compared with the best prediction performance of existing methods, the average improvements of our method in accuracy, precision, and f1-score are 3.26%, 7.03%, and 1.78%, respectively. The algorithm tested the performance of our model on the Sleep Heart Health Study data set and found that the model outperformed existing methods and could help physicians make decisions in their practices.

Motion-resolved fat-fraction mapping with whole-heart free-running multiecho GRE and pilot tone.

To develop a free-running 3D radial whole-heart multiecho gradient echo (ME-GRE) framework for cardiac- and respiratory-motion-resolved fat fraction (FF) quantification. (NTE = 8) readouts optimized for water-fat separation and quantification were integrated within a continuous non-electrocardiogram-triggered free-breathing 3D radial GRE acquisition. Motion resolution was achieved with pilot tone (PT) navigation, and the extracted cardiac and respiratory signals were compared to those obtained with self-gating (SG). After extra-dimensional golden-angle radial sparse parallel-based image reconstruction, FF, R2 *, and B0 maps, as well as fat and water images were generated with a maximum-likelihood fitting algorithm. The framework was tested in a fat-water phantom and in 10 healthy volunteers at 1.5 T using NTE = 4 and NTE = 8 echoes. The separated images and maps were compared with a standard free-breathing electrocardiogram (ECG)-triggered acquisition. The method was validated in vivo, and physiological motion was resolved over all collected echoes. Across volunteers, PT provided respiratory and cardiac signals in agreement (r = 0.91 and r = 0.72) with SG of the first echo, and a higher correlation to the ECG (0.1% of missed triggers for PT vs. 5.9% for SG). The framework enabled pericardial fat imaging and quantification throughout the cardiac cycle, revealing a decrease in FF at end-systole by 11.4% ± 3.1% across volunteers (p < 0.0001). Motion-resolved end-diastolic 3D FF maps showed good correlation with ECG-triggered measurements (FF bias of -1.06%). A significant difference in free-running FF measured with NTE = 4 and NTE = 8 was found (p < 0.0001 in sub-cutaneous fat and p < 0.01 in pericardial fat). Free-running fat fraction mapping was validated at 1.5 T, enabling ME-GRE-based fat quantification with NTE = 8 echoes in 6:15 min.

MXene/TPU Composite Film for Humidity Sensing and Human Respiration Monitoring

AbstractHuman respiration reflects abundant physiological information and could enable non‐invasive monitoring, providing information about various biological parameters such as respiration rate and depth. The rapidly growing field of humidity sensors bring forward the requirement for good performance. In this work, by virtue of good hydrophilicity and conductivity, a MXene/thermoplastic polyurethane (TPU) composite film is prepared by coating MXene nanosheets on chitosan‐modified TPU electrospun nanofibers via electrostatic interactions, for fabricating a humidity sensor. Based on the principle that the tunnel resistance changing with water molecules influences the distance of MXene nanosheets, the MXene/TPU humidity sensor exhibits fast response (12 s), wide humidity response range (11–94% RH), low hysteresis (&lt;7% RH), and excellent repeatability. The humidity sensor can be assembled with a face mask for distinguishing different human respiration patterns and accurately monitoring respiratory signals during different physical activities, suggesting its promising applications in the fields of respiratory monitoring.

MmRH: Noncontact Vital Sign Detection With an FMCW mm-Wave Radar

Vital signs, such as respiration rate (RR) and heart rate (HR), are essential for human health status assessment. The radar could detect RR and HR in a noncontact manner by sensing the repetitive chest wall movement caused by cardiopulmonary activity, which is attractive due to the more comfortable experience and better privacy protection. However, as the chest wall displacement caused by heartbeat is much smaller than that caused by respiration, the weak heartbeat component may be easily overwhelmed by respiration harmonics, noise, and clutter, making it difficult to achieve accurate and reliable HR detection. In order to tackle this challenge, in this article, we design mmRH, a system that can accurately estimate RR as well as HR using a frequency-modulated continuous-wave (FMCW) millimeter-wave (mm-wave) radar, which consists of four functional modules. In the vital sign signal extraction module, the range bin corresponding to the body part with strong vital signs is determined to extract the vital sign signal, eliminating the clutter interference. The differential enhancement module is proposed to enhance the heartbeat component by the first-order temporal difference, reducing the impacts of respiration harmonics and noise on HR estimation. In the signal decomposition module, the respiration and heartbeat signals are separated via wavelet packet decomposition (WPD) to further suppress respiration harmonics and high-frequency noise of the heartbeat signal. In the vital sign rate reconstruction module, the sparse spectrum reconstruction (SSR) of vital sign rates is mapped to an adaptive filter and zero attracting sign exponentially forgetting least mean square (ZA-SEFLMS) algorithm is proposed to achieve high-resolution sparse spectrum for accurate and reliable RR and HR estimation. The effects of different settings, including the distance between the human subject and the radar, the sensor position, and the user heterogeneity, are investigated by extensive experiments. The results indicate that the designed mmRH could effectively suppress the respiration harmonics, noise, and clutter interference, and bring a significant improvement in RR and HR detection accuracy compared with existing methods.

Artificial Intelligence-Based Patient Selection for Deep Inspiration Breath-Hold Breast Radiotherapy from Respiratory Signals

Purpose: to predict eligibility for deep inspiration breath-hold (DIBH) radiotherapy (RT) treatment of patients with left breast cancer from analysis of respiratory signal, using Deep Bidirectional Long Short-Term Memory (BLSTM) recurrent neural networks. Methods: The respiratory traces from 36 patients who underwent DIBH RT were collected. The patients’ RT treatment plans were generated for both DIBH and free-breathing (FB) modalities. The patients were divided into two classes (patient eligible or not), based on the decrease of maximum dose to the left anterior descending (LAD) artery achieved with DIBH, compared to that achieved with FB and ΔDL. Patients with ΔDL &gt; median value of ΔDL within the patient cohort were assumed to be those selected for DIBH. A BLSTM-RNN was trained for classification of patients eligible for DIBH by analysis of their respiratory signals, as acquired during acquisition of the pre-treatment computed tomography (CT), for selecting the window for DIBH. The dataset was split into training (60%) and test groups (40%), and the hyper-parameters, including the number of hidden layers, the optimizer, the learning rate, and the number of epochs, were selected for optimising model performance. The BLSTM included 2 layers of 100 neural units, each followed by a dropout layer with 20% dropout, and was trained in 35 epochs using the Adam optimizer, with an initial learning rate of 0.0003. Results: The system achieved accuracy, specificity, and sensitivity of, F1 score and area under the receiving operating characteristic curve (AUC) of 71.4%, 66.7%, 80.1%, 72.4%, and 69.4% in the test dataset, respectively. Conclusions: The proposed BLSTM-RNN classified patients in the test set eligible for DIBH with good accuracy. These results look promising for building an accurate and robust decision system to provide automated assistance to the radiotherapy team in assigning patients to DIBH.

Real-time respiratory motion prediction using photonic reservoir computing

Respiration induced motion is a well-recognized challenge in many clinical practices including upper body imaging, lung tumor motion tracking and radiation therapy. In this work, we present a recurrent neural network algorithm that was implemented in a photonic delay-line reservoir computer (RC) for real-time respiratory motion prediction. The respiratory motion signals are quasi-periodic waveforms subject to a variety of non-linear distortions. In this work, we demonstrated for the first time that RC can be effective in predicting short to medium range of respiratory motions within practical timescales. A double-sliding window technology is explored to enable the real-time establishment of an individually trained model for each patient and the real-time processing of live-streamed respiratory motion data. A breathing dataset from a total of 76 patients with breathing speeds ranging from 3 to 20 breaths per minute (BPM) is studied. Motion prediction of look-ahead times of 66.6, 166.6, and 333 ms are investigated. With a 333 ms look-ahead time, the real-time RC model achieves an average normalized mean square error (NMSE) of 0.025, an average mean absolute error (MAE) of 0.34 mm, an average root mean square error (RMSE) of 0.45 mm, an average therapeutic beam efficiency (TBE) of 94.14% for an absolute error (AE) < 1 mm, and 99.89% for AE < 3 mm. This study demonstrates that real-time RC is an efficient computing framework for high precision respiratory motion prediction.

An investigation of effects of a non-repetitive preferred music on physiological responses amongst a group of chronic tinnitus patients

A kind of specific music, which was non-repetitive and fulfilled the individualized preferences of tinnitus patients, has been shown to be effective in relieving tinnitus in short term. Currently, the effects of the specific music on physiological responses of tinnitus patients were unknown, which was the main objective of this study. Thirty patients suffering from chronic tinnitus were involved in this study for a three-week specific music intervention. During this period, tinnitus patients received four measurements, each of which simultaneously collected electrocardiogram (ECG), photoplethysmogram (PPG), respiration (RSP), and surface electromyogram (EMG) signals to measure physiological responses, such as activities of autonomic nervous system (ANS) and mental stress. The effects of the specific music were evaluated by differences of physiological responses in different stages (before, during, and after specific music intervention). Significant differences (p ≤ 0.001) of physiological parameters among the three stages were observed. Results indicated that the specific music might be helpful to increase HRV, enhance activities of parasympathetic nervous system, and mitigate the mental stress of tinnitus patients.

Mattress-Based Non-Influencing Sleep Apnea Monitoring System.

A mattress-type non-influencing sleep apnea monitoring system was designed to detect sleep apnea-hypopnea syndrome (SAHS). The pressure signals generated during sleep on the mattress were collected, and ballistocardiogram (BCG) and respiratory signals were extracted from the original signals. In the experiment, wavelet transform (WT) was used to reduce noise and decompose and reconstruct the signal to eliminate the influence of interference noise, which can directly and accurately separate the BCG signal and respiratory signal. In feature extraction, based on the five features commonly used in SAHS, an innovative respiratory waveform similarity feature was proposed in this work for the first time. In the SAHS detection, the binomial logistic regression was used to determine the sleep apnea symptoms in the signal segment. Simulation and experimental results showed that the device, algorithm, and system designed in this work were effective methods to detect, diagnose, and assist the diagnosis of SAHS.

Signal Preprocessing for Heartbeat Detection Using Continuous-Wave Doppler Radar

A signal preprocessing method is proposed for effectively monitoring heartbeats from signals obtained using a continuous-wave (CW) Doppler radar. The proposed method consists of windowing with a size optimized for heartbeat detection, detrending to minimize the effect of dc drift, and data reconstruction to increase the fast Fourier transform (FFT) resolution. A window size of 3 s effectively extracts the heartbeat alone from the received signals of the CW radar, which comprised respiration, motion, and heartbeat signals. The effect of the low-frequency noise caused by the internal heat generation of hardware components is reduced by the detrending technique, and the low resolution of the FFT, due to a small number of samples, is improved through reconstruction by combining data from three windows. The signal-to-noise ratio of the measured waveform was improved to 5.83 dB using the proposed method in a 5.8-GHz CW radar. For the data acquired for 60 s, the proposed method showed a heartbeat detection accuracy of 95.8% by using a signal processing time 13% lower than that of the conventional method.

A Real-Time Respiration Monitoring System Using WiFi Sensing Based on the Concentric Circle Model.

This paper proposes and experimentally validates a novel concentric circle (CC) model for indoor WiFi sensing. By setting the transmitter and receiver together, the perception model becomes concentric circles with equal spacing, eliminating the blind zone and unequal radial sensitivity problems of the Fresnel zone (FZ) model. Then a human respiratory monitoring system is developed based on this model, which executes the following steps: (1) Principal component analysis (PCA) is applied to the channel state information ratio (CSIR) as a preprocessing to extract the components related to human activities. (2) Human presence and respiratory signal detection are adopted to improve monitoring accuracy. (3) The Doppler respiratory frequency is extracted to calculate the respiratory rate. Experimental results show that the CC model achieves high accuracy in velocity measurement with an error of less than 0.4 cm/s. The respiration monitoring system can accurately monitor human respiration with an error of less than 0.7 bpm within 6 m.

Self-x based closed loop wearable IoT for real-time detection and resolution of sleep apnea

This study introduces the Sleepcare Kit and S-Pillow (SKP), a wearable body-area network concept designed to address the uncomfortable and imprecise measurements associated with previous sleep apnea diagnosis and treatment methods that depend on external servers or human experts. SKP measures respiration, acceleration, and SpO2 signals to accurately identify risky sleep apnea in real-time. In addition, it incorporates a S-Pillow that can deterministically adjust the position of the patient’s head to resolve sleep apnea symptoms. SKP operates in a self-x based closed loop to optimally manage sleep apnea risk in a patient’s home environment without external help. Experimental results shows a 1/6 reduction in control frequency and a 1/3 reduction in apnea time compared to previous studies related to pillow use.

Estimation of Respiratory Rate during Biking with a Single Sensor Functional Near-Infrared Spectroscopy (fNIRS) System

The employment of wearable systems for continuous monitoring of vital signs is increasing. However, due to substantial susceptibility of conventional bio-signals recorded by wearable systems to motion artifacts, estimation of the respiratory rate (RR) during physical activities is a challenging task. Alternatively, functional Near-Infrared Spectroscopy (fNIRS) can be used, which has been proven less vulnerable to the subject's movements. This paper proposes a fusion-based method for estimating RR during bicycling from fNIRS signals recorded by a wearable system. Firstly, five respiratory modulations are extracted, based on amplitude, frequency, and intensity of the oxygenated hemoglobin concentration (O2Hb) signal. Secondly, the dominant frequency of each modulation is computed using the fast Fourier transform. Finally, dominant frequencies of all modulations are fused, based on averaging, to estimate RR. The performance of the proposed method was validated on 22 young healthy subjects, whose respiratory and fNIRS signals were simultaneously recorded during a bicycling task, and compared against a zero delay Fourier domain band-pass filter. The comparison between results obtained by the proposed method and band-pass filtering indicated the superiority of the former, with a lower mean absolute error (3.66 vs. 11.06 breaths per minute, p<0.05). The proposed fusion strategy also outperformed RR estimations based on the analysis of individual modulation. This study orients towards the practical limitations of traditional bio-signals for RR estimation during physical activities.

Determining Obstruction in Endotracheal Tubes Using Physical Respiratory Signals

This study proposes a method for determining obstruction of the endotracheal tube (ET) and its degree and location. Respiratory signals were acquired using a three sensor (microphone, pressure, and flow) integrated sensor connector. Obstruction classification involved pre-processing and feature extraction. During pre-processing, one cycle of the respiratory signal was extracted using respiratory cycle extraction and phase segmentation. The signal was then divided into three phases: (i) inspiratory phase, (ii) expiratory phase, and (iii) between both the phases, where the intrapulmonary pressure increased, decreased, and remained constant, respectively. In the feature extraction process, the results were quantified using absolute value average and texture analyses. Artificial ET tubes were fabricated to simulate the presence of foreign substances in the ET tube; they had different degrees of obstruction (0%, 20%, 40%, and 50%) and obstruction positions (Sections 1, 2, and 3). The experiment was performed by connecting the sensor connector and artificial ET tube between the ventilator and test lung. Respiratory signals were obtained in 10 cases by cross connecting the artificial ET tubes. The degree and location of obstruction were classified according to the average absolute value and texture analyses of the flow data. The obstruction can be determined through the texture analysis results using the combined microphone and flow sensor data. The proposed method is simple in configuration, can be readily used in existing setups, and can be operated regardless of surrounding noise.

Increased respiratory modulation of cardiovascular control reflects improved blood pressure regulation in pregnancy.

Hypertensive pregnancy disorders put the maternal-fetal dyad at risk and are one of the leading causes of morbidity and mortality during pregnancy. Multiple efforts have been made to understand the physiological mechanisms behind changes in blood pressure. Still, to date, no study has focused on analyzing the dynamics of the interactions between the systems involved in blood pressure control. In this work, we aim to address this question by evaluating the phase coherence between different signals using wavelet phase coherence. Electrocardiogram, continuous blood pressure, electrocardiogram-derived respiration, and muscle sympathetic nerve activity signals were obtained from ten normotensive pregnant women, ten normotensive non-pregnant women, and ten pregnant women with preeclampsia during rest and cold pressor test. At rest, normotensive pregnant women showed higher phase coherence in the high-frequency band (0.15-0.4Hz) between muscle sympathetic nerve activity and the RR interval, blood pressure, and respiration compared to non-pregnant normotensive women. Although normotensive pregnant women showed no phase coherence differences with respect to hypertensive pregnant women at rest, higher phase coherence between the same pairs of variables was found during the cold pressor test. These results suggest that, in addition to the increased sympathetic tone of normotensive pregnant women widely described in the existing literature, there is an increase in cardiac parasympathetic modulation and respiratory-driven modulation of muscle sympathetic nerve activity and blood pressure that could compensate sympathetic increase and make blood pressure control more efficient to maintain it in normal ranges. Moreover, blunted modulation could prevent its buffer effect and produce an increase in blood pressure levels, as observed in the hypertensive women in this study. This initial exploration of cardiorespiratory coupling in pregnancy opens the opportunity to follow up on more in-depth analyses and determine causal influences.

A Machine-Learning-Algorithm-Assisted Intelligent System for Real-Time Wireless Respiratory Monitoring

Respiratory signals are basic indicators of human life and health that are used as effective biomarkers to detect respiratory diseases in clinics, including cardiopulmonary function, breathing disorders, and breathing system infections. Therefore, it is necessary to continuously measure respiratory signals. However, there is still a lack of effective portable electronic devices designed to meet the needs of daily respiratory monitoring. This study presents an intelligent, portable, and wireless respiratory monitoring system for real-time evaluation of human respiratory behaviors. The system consists of a triboelectric respiratory sensor; circuit board hardware for data acquisition, preprocessing, and wireless transmission; a machine learning algorithm for enhancing recognition accuracy; and a mobile terminal app. The triboelectric sensor—fabricated by the screen-printing method—is lightweight, non-invasive, and biocompatible. It provides a clear response to the frequency and intensity of respiratory airflow. The portable circuit board is reusable and cost-effective. The decision tree model algorithm is used to identify the respiratory signals with an average accuracy of 97.2%. The real-time signal and statistical results can be uploaded to a server network and displayed on various mobile terminals for body health warnings and advice. This work promotes the development of wearable health monitoring systems.

Design and Testing of a Smart Facemask for Respiratory Monitoring during Cycling Exercise

Given the importance of respiratory frequency (fR) as a valid marker of physical effort, there is a growing interest in developing wearable devices measuring fR in applied exercise settings. Biosensors measuring chest wall movements are attracting attention as they can be integrated into textiles, but their susceptibility to motion artefacts may limit their use in some sporting activities. Hence, there is a need to exploit sensors with signals minimally affected by motion artefacts. We present the design and testing of a smart facemask embedding a temperature biosensor for fR monitoring during cycling exercise. After laboratory bench tests, the proposed solution was tested on cyclists during a ramp incremental frequency test (RIFT) and high-intensity interval training (HIIT), both indoors and outdoors. A reference flowmeter was used to validate the fR extracted from the temperature respiratory signal. The smart facemask showed good performance, both at a breath-by-breath level (MAPE = 2.56% and 1.64% during RIFT and HIIT, respectively) and on 30 s average fR values (MAPE = 0.37% and 0.23% during RIFT and HIIT, respectively). Both accuracy and precision (MOD ± LOAs) were generally superior to those of other devices validated during exercise. These findings have important implications for exercise testing and management in different populations.

Focused navigation for respiratory-motion-corrected free-running radial 4D flow MRI.

To validate a respiratory motion correction method called focused navigation (fNAV) for free-running radial whole-heart 4D flow MRI. Using fNAV, respiratory signals derived from radial readouts are converted into three orthogonal displacements, which are then used to correct respiratory motion in 4D flow datasets. Hundred 4D flow acquisitions were simulated with non-rigid respiratory motion and used for validation. The difference between generated and fNAV displacement coefficients was calculated. Vessel area and flow measurements from 4D flow reconstructions with (fNAV) and without (uncorrected) motion correction were compared to the motion-free ground-truth. In 25 patients, the same measurements were compared between fNAV 4D flow, 2D flow, navigator-gated Cartesian 4D flow, and uncorrected 4D flow datasets. For simulated data, the average difference between generated and fNAV displacement coefficients was 0.04 0.32 mm and 0.31 0.35 mm in the x and y directions, respectively. In the z direction, this difference was region-dependent (0.02 0.51 mm up to 5.85 3.41 mm). For all measurements (vessel area, net volume, and peak flow), the average difference from ground truth was higher for uncorrected 4D flow datasets (0.32 0.11 cm2 , 11.1 3.5mL, and 22.3 6.0mL/s) than for fNAV 4D flow datasets (0.10 0.03 cm2 , 2.6 0.7mL, and 5.1 .9mL/s, p < 0.05). In vivo, average vessel area measurements were 4.92 2.95 cm2 , 5.06 2.64 cm2 , 4.87 2.57 cm2 , 4.87 2.69 cm2 , for 2D flow and fNAV, navigator-gated and uncorrected 4D flow datasets, respectively. In the ascending aorta, all 4D flow datasets except for the fNAV reconstruction had significantly different vessel area measurements from 2D flow. Overall, 2D flow datasets demonstrated the strongest correlation to fNAV 4D flow for both net volume (r2 =0.92) and peak flow (r2 =0.94), followed by navigator-gated 4D flow (r2 =0.83 and r2 =0.86, respectively), and uncorrected 4D flow (r2 =0.69 and r2 =0.86, respectively). fNAV corrected respiratory motion in vitro and in vivo, resulting in fNAV 4D flow measurements that are comparable to those derived from 2D flow and navigator-gated Cartesian 4D flow datasets, with improvements over those from uncorrected 4D flow.

Real Time Implementation of Robust Sound based Respiratory Disease Classification using Spectrogram and Deep Convolutional Neural Networks

Respiratory diseases become burden to affect health of the people and five lung related diseases namely COPD, Asthma, Tuberculosis, Lower respiratory tract infection and Lung cancer are leading causes of death worldwide. X-ray or CT scan images of lungs of patients are analysed for prediction of any lung related respiratory diseases clinically. Respiratory sounds also can be analysed to diagnose the respiratory illness prevailing among humans. Sound based respiratory disease classification against healthy subjects is done by extracting spectrogram from the respiratory sound signal and Convolutional neural network (CNN) templates are created by applying the extracted features on the layered CNN architecture. Test sound is classified to be associated with respiratory disease or healthy subjects by applying the testing procedure on the test feature frames of spectrogram. Evaluation of the respiratory disease binary classification is performed by considering 80% and 20% of the extracted spectrogram features for training and testing. An automated system is developed to classify the respiratory diseases namely upper respiratory tract infection (URTI), pneumonia, bronchitis, bronchiectasis, and coronary obstructive pulmonary disease (COPD) against healthy subjects from breathing &amp; wheezing sounds. Decision level fusion of spectrogram, Melspectrogram and Gammatone gram features with CNN for modelling &amp; classification is done and the system has deliberated the accuracy of 98%. Combination of Gammatone gram and CNN has provided very good results for binary classification of pulmonary diseases against healthy subjects. This system is realized in real time by using Raspberry Pi hardware and this system provides the validation error of 14%. This automated system would be useful for COVID testing using breathing sounds if respiratory sound database with breathing sound recordings from COVID patients would be available.

4DCT is long overdue for improvement.

4DCT is long overdue for improvement.