Physiological measurement最新文献

Objective. Air trapping is a major symptom of respiratory diseases like chronic obstructive pulmonary disease and asthma, and has always been a significant problem in treating patients using mechanical ventilation. If not handled timely, it can pose risk of severe respiratory dysfunction and potential life-threatening complications. Currently, the assessment of air trapping for ventilated patients largely relies on clinical experience of medical staffs.Approach. We introduced an interpretable dual-channel one-dimensional convolutional neural network (DC-1DCNN) with a simple structure, which enables fast inference. This model is designed to classify whether a respiratory waveform is indicative of air trapping. A global average pooling layer was integrated into the DC-1DCNN model to highlight the segments of the respiratory waveform that the model focused on when making a classification. An air trapping index (ATI) was introduced to quantify the condition of air trapping in the ventilated patients and to evaluate the effectiveness of bronchodilator nebulized treatments.Main results. The results demonstrate a satisfactory accuracy of 96.4% in identifying air trapping breath cycles, with highlighted critical sections in breath cycles that match the understanding of clinical experts for air trapping. The efficacy of bronchodilators can be well assessed by the ATI.Significance. The findings suggest that the proposed DC-1DCNN can help detect air trapping in real-time, and help the clinicians better monitor the airway condition of the ventilated patients.

Objective.Interventional therapy represents a primary treatment modality for moderate to severe coronary atherosclerosis. However, potential complications following stent implantation can pose significant risks to patients. This study aims to explore the relationship between aberrant hemodynamic patterns and the incidence of post-stent implantation complications.Approach.By creating models of three distinct types of coronary artery stents and utilizing clinical fractional flow reserve data, this research employs fluid-structure interaction analyses to simulate the hemodynamic alterations and vascular wall responses post-stent implantation.Main results.It is indicated that implantation of stents can induce complex hemodynamic modifications in the vicinity of the stent, particularly at the juncture where the stent contacts the vascular wall. While the hemodynamic profiles of the three stent types exhibit general consistency, distinctions in local hemodynamics arise from the varied pore densities inherent to each stent design. Notably, the B-type stent, characterized by their moderate pore density, demonstrates comparatively stable hemodynamics relative to the other stent types. Additionally, stent implantation impacts areas of the vascular wall not covered by the stent, with notable hemodynamic changes also manifesting in these regions.Significance.The implantation of stents has a significant impact on the hemodynamics inside the blood vessels. Specifically, abnormal hemodynamic changes near the stents can inflict damage to the blood vessel wall, thus accelerating the occurrence of complications. Moreover, the hemodynamic changes elicited by different stents vary significantly, and it has been observed that stents with moderate grid spacing exhibit superior performance in mitigating adverse hemodynamic effects.

Objective. Pain assessment in preterm infants is often based on subjective observations, which may lack objectivity and are labor-intensive. Non-invasive EEG can serve as an objective assessment tool. However, no specific EEG feature within a particular frequency band and brain region has been reported for pain detection in the objective pain assessment of preterm infants. This study quantified electroencephalography (EEG) responses to procedural pain during a puncture in preterm infants, specifically analyzing three EEG parameters.Approach. Fifty-seven EEG datasets from forty-two preterm infants were analyzed across eight EEG channels. The differences between the upper and lower margins (UM-LM) of amplitude-integrated EEG (aEEG), as well as the five frequency bands (low delta, high delta, theta, alpha, and beta) of frequency power and time-frequency power, were used to characterize the response of the brain to pain during specific periods: before, during, and after the puncture.Main results. The Fp1 and Fp2 exhibited the most significant differences in the UM-LM aEEG differences between before vs during (Fp1:p= 0.0060, Fp2:p= 0.0031), before vs after (p< 0.0001), and during vs after (Fp1:p= 0.0427, Fp2:p= 0.025) the puncture. The C3 and C4 responded significantly to pain during the puncture in the frequency and time-frequency power, notably the time-frequency power in the low delta, which showed the most significant differences between the periods before vs during (p< 0.0001), before vs after (p< 0.0001), and during vs after (p= 0.0002) the puncture.Significance. The central brain region responds significantly to procedural pain in preterm infants, which is prominently detected in the low delta of time-frequency power. These findings support the use of EEG application as an objective and non-invasive method to identify and detect pain in nonverbal populations, focusing on specific critical channels and frequency bands.

Objective. Operative vaginal delivery (OVD) is a major obstetrical issue in developed countries. In this study, we analyzed simultaneous maternal (MHR) and fetal heart rate (FHR) variabilities, as markers of maternal and fetal autonomous nervous systems activity, in relation with OVD.Approach. A set of 44 simultaneous MHR and FHR recordings were obtained from distinct singleton term pregnancies in normal (n= 27) and OVD (n= 17) in the last two hours of labor (H₁and H₂), and were analyzed using linear and nonlinear indices of heart rate variability analysis. Interaction between MHR and FHR was assessed through their differences and cross-sample entropy analysis.Main results. With progression of labor, there was an increase in most MHR and FHR linear indices, a decrease of entropy indices and an increase of MHR and FHR synchrony/regularity, whereas the sympatho-vagal balance (LF/HF) increased in the mother but decreased in the fetus. Mean MHR, predominance of low frequencies (LF_norm) and LF/HF were significantly higher in the OVD group, while the opposite occurred with the high frequencies (HF) and entropy. The synchronization/regularity between MHR and FHR was significantly higher in OVD. A sensitivity and specificity of 94.1% and 70.4%, respectively, were achieved in the classification of OVD cases using an univariate linear discriminant.Significance. Maternal-FHR variability analysis adds information regarding intrapartum physiology and maternal-fetal behavior and might be considered in prediction models.

Objective. This study aims to introduce a novel generative pre-trained transformer (GPT)-based foundation model specifically tailored to photoplethysmography (PPG) signals, enabling effective adaptation to various downstream biomedical tasks.Approach. We adapted the standard GPT architecture to handle the continuous characteristics of PPG signals, leveraging extensive pre-training on a large dataset comprising over 200 million 30 s PPG samples, followed by supervised fine-tuning strategies for task-specific optimization.Main results. Our approach achieves performance comparable to or exceeding current state-of-the-art methods on various downstream tasks, notably atrial fibrillation detection, and demonstrates a unique generative capability, such as effective signal denoising, inherently available without additional fine-tuning.Significance. The successful adaptation of GPT to PPG signals underscores the potential of generative transformer frameworks as versatile foundation models in biomedical signal processing, highlighting their dual role in predictive and generative tasks.

Objective.To develop and validate a machine learning framework for the classification of distinct seizure onset patterns using intracranial EEG (iEEG) recordings in a non-human primate (NHP) model of penicillin-induced seizures.Approach.iEEG data were collected from six NHPs, comprising 1, 496 frontal and 549 temporal lobe seizures. Seizure onset patterns were manually categorized into five types: Sharp Activity (5-15 Hz), Low Amplitude Fast Activity (15-30 Hz), Delta Brush (1-3 Hz with bursts), High Amplitude Spike (2-5 Hz), and Polyspike. A Random Forest classifier was trained using features extracted from optimized seizure onset segments. Feature selection and seizure segment length optimization were performed using nested cross-validation to enhance classification accuracy and generalizability.Main results.The classifier achieved strong performance with F1-scores exceeding 79% for Sharp Activity, Low Amplitude Fast Activity, and High Amplitude Spike patterns. When validated on an independent temporal lobe seizure dataset, the model demonstrated robust generalizability, achieving precision and sensitivity above 80% for Sharp Activity and High Amplitude Spike.Significance.These findings demonstrate that the suggested spectral and dynamic features can effectively distinguish seizure onset patterns and generalize in distinct brain regions. Although there are limitations due to use of manual annotations and the sample size of certain categories, the proposed approach provides a framework for automatic classification of seizure onset patterns. Further, the framework has a potential use for epilepsy research and clinical applications in future.

Objective.Monitoring the volume status of patients remains a fundamentally unresolved issue in the perioperative setting but also in intensive care medicine. Electrical impedance tomography (EIT) enables noninvasive and radiation-free functional imaging of impedance changes of a cross-sectional plane of the thorax.Approach.We hypothesized that cardiac-induced impedance fluctuations in the EIT signal contain information about the global volume status. Therefore, we recorded EIT signals from 24 healthy volunteers before and up to 100 min after ingestion of 750 ml liquid following more than nine hours fluid restriction. We isolated a characteristic cardiac-induced impedance profile from the global impedance curve and compared its characteristics to measures of non-invasive hemodynamic monitoring and the diameter of the vena cava inferior (VCI) determined via ultrasonography.Main results.The diameter of the VCI, mean arterial pressure and stroke volume index, did not change systematically after liquid ingestion. Heart rate was increased 20 min after, and heart rate variability was increased immediately after liquid ingestion. The amplitude, the area under the curve and gradients of its rise and decline of the characteristic cardiac-induced impedance profile changed 60 min after liquid ingestion.Significance.Cardiac-induced pulsatile signals in EIT recordings changed characteristically after liquid ingestion, loosely corresponding with changes in heart rate and heart rate variability. These signals may contain valuable information about the general volume status and may be utilized for enhanced monitoring of a patient's volume status.

Objective.Radiofrequency (RF) catheter ablation is a standard treatment for patients with cardiac arrhythmias, providing an efficient, minimally invasive solution. However, the ablation efficiency remains suboptimal due to numerous contributed factors that are overlooked in the literature and not monitored during the procedure. This paper explores the effect of catheter-to-tissue contact angles on lesion formations and the feasibility of the multichannel bioimpedance method in characterising the angles to inform cardiologists.Approach.Two silico simulations based on a realistic human model were built to: (1) simulate lesion formations with different catheter-to-tissue angles under varying conditions of powers and convection cooling, and (2) simulate multichannel bioimpedances measured at each catheter's location and angle. 13 locations were picked in all four chambers with 3 contact conditions (catheter lies along the muscle (0° and 180°), in perpendicular to the muscle (90°) and in middle angles (45° and 135°)). 64 electrodes divided into 4 bands were placed on the thorax for multichannel bioimpedances (3-terminal) measured between the catheter's second electrode E2 (I+,V+), and each pair of adjacent surface electrodes (I-,V-). ANOVA and Tukey's Honestly Significant Difference (HSD) tests were used to evaluate the contact angle's effect on the lesion formations and the bioimpedance's capability in distinguishing between angles.Main results.The results showed that 0° and 180° configurations generated significantly different lesions from other angles. The multichannel bioimpedances could recognise 0°/180° from other angles and correlated moderately to lesion sizes at low ablation power.Significance.This paper concludes that catheter-to-tissue angles can influence the lesion outcomes significantly and the multichannel bioimpedance is able to detect the angles that matter.

Objective. Functional network connectivity (FNC) estimated from resting-state functional magnetic resonance imaging showed great information about the neural mechanism in different brain disorders. But previous research has mainly focused on standard statistical learning approaches to find FNC features separating patients from control. While machine learning models can improve classification accuracy, they often lack interpretability, making it difficult to understand how they arrive at their decisions.Approach. Explainable machine learning helps address this issue by identifying which features contribute most to the model's predictions. In this study, we introduce a novel framework leveraging SHapley Additive exPlanations (SHAPs) to identify crucial FNC features distinguishing between two distinct population classes.Main results. Initially, we validate our approach using synthetic data. Subsequently, applying our framework, we ascertain FNC biomarkers distinguishing between, controls and schizophrenia (SZ) patients with accuracy of 81.04% as well as middle aged adults and old aged adults with accuracy 71.38%, respectively, employing random forest, XGBoost, and CATBoost models.Significance. Our analysis underscores the pivotal role of the cognitive control network (CCN), subcortical network (SCN), and somatomotor network in discerning individuals with SZ from controls. In addition, our platform found CCN and SCN as the most important networks separating young adults from older.