How Not to Make the Joint Extended Kalman Filter Fail with Unstructured Mechanistic Models.

The unstructured mechanistic model (UMM) allows for modeling the macro-scale of a phenomenon without known mechanisms. This is extremely useful in biomanufacturing because using the UMM for the joint estimation of states and parameters with an extended Kalman filter (JEKF) can enable the real-time monitoring of bioprocesses with unknown mechanisms. However, the UMM commonly used in biomanufacturing contains ordinary differential equations (ODEs) with unshared parameters, weak variables, and weak terms. When such a UMM is coupled with an initial state error covariance matrix P(t=0) and a process error covariance matrix Q with uncorrelated elements, along with just one measured state variable, the joint extended Kalman filter (JEKF) fails to estimate the unshared parameters and state simultaneously. This is because the Kalman gain corresponding to the unshared parameter remains constant and equal to zero. In this work, we formally describe this failure case, present the proof of JEKF failure, and propose an approach called SANTO to side-step this failure case. The SANTO approach consists of adding a quantity to the state error covariance between the measured state variable and unshared parameter in the initial P(t = 0) of the matrix Ricatti differential equation to compute the predicted error covariance matrix of the state and prevent the Kalman gain from being zero. Our empirical evaluations using synthetic and real datasets reveal significant improvements: SANTO achieved a reduction in root-mean-square percentage error (RMSPE) of up to approximately 17% compared to the classical JEKF, indicating a substantial enhancement in estimation accuracy.

Analysis of the effect of CPAP on hemodynamics using clinical data and a theoretical model: CPAP therapy decreases cardiac output mechanically but increases it via afterload reduction.

BACKGROUND: Noninvasive positive pressure ventilation (NIPPV) has been established as an effective treatment for heart failure. Positive airway pressure such as continuous positive airway pressure (CPAP) increases cardiac output (CO) in some patients but decreases it in others. However, the mechanism behind such unpredictable responses remains undetermined. METHODS AND RESULTS: We measured hemodynamic parameters of 38 cases using Swan-Ganz catheter before and after CPAP in chronic heart failure status. In those whose CO increased by CPAP, pulmonary vascular resistance (PVR) was significantly decreased and SpO2 significantly increased, but the other parameters were not changed. On the other hand, PVR was not changed, but systemic vascular resistance (SVR) was increased in those whose CO decreased by CPAP. To explain this phenomenon, we simulated the cardiovascular system using a cardiac model of time-varying elastance. In this model, it was indicated that CPAP decreases CO irrespective of cardiac function or filling status under constant PVR condition. However, when reduction of PVR by CPAP was taken into account, an increase in CO was expected especially in the hypervolemic and low right ventricle (RV) systolic function cases. CONCLUSIONS: CPAP would increase CO only where PVR can be reduced by CPAP therapy, especially in the case with hypervolemia and/or low RV systolic function. Understanding the underlying mechanism should help identify the patients for whom NIPPV would be effective.

rAAV Manufacturing: The Challenges of Soft Sensing during Upstream Processing.

Recombinant adeno-associated virus (rAAV) is the most effective viral vector technology for directly translating the genomic revolution into medicinal therapies. However, the manufacturing of rAAV viral vectors remains challenging in the upstream processing with low rAAV yield in large-scale production and high cost, limiting the generalization of rAAV-based treatments. This situation can be improved by real-time monitoring of critical process parameters (CPP) that affect critical quality attributes (CQA). To achieve this aim, soft sensing combined with predictive modeling is an important strategy that can be used for optimizing the upstream process of rAAV production by monitoring critical process variables in real time. However, the development of soft sensors for rAAV production as a fast and low-cost monitoring approach is not an easy task. This review article describes four challenges and critically discusses the possible solutions that can enable the application of soft sensors for rAAV production monitoring. The challenges from a data scientist's perspective are (i) a predictor variable (soft-sensor inputs) set without AAV viral titer, (ii) multi-step forecasting, (iii) multiple process phases, and (iv) soft-sensor development composed of the mechanistic model.

A Real-Time Respiration Monitoring and Classification System Using a Depth Camera and Radars.

Respiration rate (RR) and respiration patterns (RP) are considered early indicators of physiological conditions and cardiorespiratory diseases. In this study, we addressed the problem of contactless estimation of RR and classification of RP of one person or two persons in a confined space under realistic conditions. We used three impulse radio ultrawideband (IR-UWB) radars and a 3D depth camera (Kinect) to avoid any blind spot in the room and to ensure that at least one of the radars covers the monitored subjects. This article proposes a subject localization and radar selection algorithm using a Kinect camera to allow the measurement of the respiration of multiple people placed at random locations. Several different experiments were conducted to verify the algorithms proposed in this work. The mean absolute error (MAE) between the estimated RR and reference RR of one-subject and two-subjects RR estimation are 0.61±0.53 breaths/min and 0.68±0.24 breaths/min, respectively. A respiratory pattern classification algorithm combining feature-based random forest classifier and pattern discrimination algorithm was developed to classify different respiration patterns including eupnea, Cheyne-Stokes respiration, Kussmaul respiration and apnea. The overall classification accuracy of 90% was achieved on a test dataset. Finally, a real-time system showing RR and RP classification on a graphical user interface (GUI) was implemented for monitoring two subjects.

Monitoring Respiratory Motion during VMAT Treatment Delivery Using Ultra-Wideband Radar.

The goal of this paper is to evaluate the potential of a low-cost, ultra-wideband radar system for detecting and monitoring respiratory motion during radiation therapy treatment delivery. Radar signals from breathing motion patterns simulated using a respiratory motion phantom were captured during volumetric modulated arc therapy (VMAT) delivery. Gantry motion causes strong interference affecting the quality of the extracted respiration motion signal. We developed an artificial neural network (ANN) model for recovering the breathing motion patterns. Next, automated classification into four classes of breathing amplitudes is performed, including no breathing, breath hold, free breathing and deep inspiration. Breathing motion patterns extracted from the radar signal are in excellent agreement with the reference data recorded by the respiratory motion phantom. The classification accuracy of simulated deep inspiration breath hold breathing was 94% under the worst case interference from gantry motion and linac operation. Ultra-wideband radar systems can achieve accurate breathing rate estimation in real-time during dynamic radiation delivery. This technology serves as a viable alternative to motion detection and respiratory gating systems based on surface detection, and is well-suited to dynamic radiation treatment techniques. Novelties of this work include detection of the breathing signal using radar during strong interference from simultaneous gantry motion, and using ANN to perform adaptive signal processing to recover breathing signal from large interference signals in real time.

Contactless Measurement of Vital Signs Using Thermal and RGB Cameras: A Study of COVID 19-Related Health Monitoring.

In this study, a contactless vital signs monitoring system was proposed, which can measure body temperature (BT), heart rate (HR) and respiration rate (RR) for people with and without face masks using a thermal and an RGB camera. The convolution neural network (CNN) based face detector was applied and three regions of interest (ROIs) were located based on facial landmarks for vital sign estimation. Ten healthy subjects from a variety of ethnic backgrounds with skin colors from pale white to darker brown participated in several different experiments. The absolute error (AE) between the estimated HR using the proposed method and the reference HR from all experiments is 2.70±2.28 beats/min (mean ± std), and the AE between the estimated RR and the reference RR from all experiments is 1.47±1.33 breaths/min (mean ± std) at a distance of 0.6-1.2 m.

Novel Cuffless Blood Pressure Estimation Method Using a Bayesian Hierarchical Model.

Continuous blood pressure (BP) monitoring is important for the prevention and early diagnosis of cardiovascular diseases. Cuffless BP estimation using pulse arrival time (PAT) via a mathematical model which enables continuous BP measurement has recently become a popular research topic. In this study, simultaneous biomedical signals from ten healthy subjects were acquired by electrocardiogram (ECG) and photoplethysmogram (PPG) sensors and the continuous reference BP data were collected by a cuff-based Finometer PRO BP monitor. A hierarchical model was applied to estimate the parameters of a nonlinear model which in turn is used to estimate systolic blood pressure (SBP) using PAT with few calibration measurements. The mean absolute difference (MAD) between the estimated SBP and reference SBP is 4.35±1.43 mmHg using the proposed hierarchical model with three calibration measurements and is 4.36±1.17 mmHg with a single calibration measurement.

Detection of Respiratory Signal Based on Depth Camera Body Tracking.

A novel respiratory signal detection system capable of simultaneously tracking the position of the subject and detecting his or her respiratory signal is described. The monitoring system consists of depth camera with ultra wide band radar device. Both sensors are connected through a mini computer, which performs data acquisition and storage. In this paper, we propose a method to locate the position of the subject where he or she is lying in the bed covered with blanket. Mask R-CNN is used to help segment upper-body's silhouette and give out the center point distance. The distance between the camera and the subject is then converted into a range bin of the radar and the breath-like signal is extracted from that range- bin. Additional contribution of this paper is that we developed a classifier to classify the whether the extracted signal in the selected range bin is indeed a breathing signal or not.

A Joint Localization Assisted Respiratory Rate Estimation using IR-UWB Radars.

Respiratory rate (RR) is one of the vital signs which is commonly measured by contact-based methods, such as using a breathing belt. Recently, significant research has been conducted related to contactless RR monitoring - however, the majority of experiments are performed in situations when the subject is oriented towards the radar. In this research, we are interested in monitoring the breathing of subjects who can be anywhere in the room. A system of three impulse radio ultrawideband (IR-UWB) radars is used to cover the whole room. A Kinect camera that can track subjects' joints 3D coordinates was employed to localize the subjects. The results of RR monitoring using IR-UWB radars and Kinect camera show good performance in single/multiple subject(s) tracking and RR estimation.

Continuous Tracking of Changes in Systolic Blood Pressure using BCG and ECG.

Blood pressure (BP) is an important physiological marker of human health. It is commonly measured by a cuff-based monitor via either auscultatory or oscillometric methods. Recently, significant research has been conducted to mathematically estimate BP from pulse transit time (PTT) to enable cuffless and continuous BP measurement. In this research, a new time reference, RJ interval, which is the time delay between electrocardiogram (ECG) R peak and ballistocardiogram (BCG) J peak was evaluated to determine if it can be used as a surrogate of PTT in cuffless BP estimation. Biomedical signals from ten healthy subjects were acquired by BCG, ECG and PPG sensors and the continuous reference BP data were collected by a cuff-based Finometer PRO BP monitor. An exponential model was employed to estimate systolic blood pressure (SBP) using RJ interval and PTT. RJ intervals extracted from ECG and BCG were shown to be useful in evaluating trends of SBP and can be the surrogate of PTT in cuffless SBP estimation.

Classification of Human Posture from Radar Returns Using Ultra-Wideband Radar.

There is a great need for new technology that helps ensure the well-being of senior citizens who have compromised health and are at an elevated risk of injury due to falls. Being able to detect posture and postural changes may be helpful in prediction and prevention of impending falls. Ultra-Wideband (UWB) radar is an attractive means for patient monitoring because it is inexpensive, capable of penetrating obstacles, privacy preserving and it consumes little power. In this paper, classification of postures, namely sitting, standing and lying is presented using stand-off sensing using UWB radar in an indoor environment. It is found that using location specific classifiers, overall accuracy can be improved. In this paper, a decision tree classifier capable of achieving 85% overall accuracy is proposed. This classifier uses 33 features from 10 second data sample segments.

Detecting Cardiac Activity by Capacitive Electrodes from a Single Point on the Wrist.

Ballistocardiography (BCG) is the measurement of body movement by forces associated with heart contraction that can be used for monitoring cardiac activity. It has already been measured by force sensor and accelerometer. In this research, we developed a capacitive wristband that provides a method for single point, continuous BCG measurement, which has the potential to become a new type of sensor for wearable health care. The aim of this paper is to validate that the signal detected by capacitive electrodes is actually the BCG signal. Signals from four healthy subjects were acquired by a capacitive wristband together with Electrocardiogram (ECG). The capacitive signal was validated by both morphology matching analysis and wave occurrence time matching analysis to show that it is indeed BCG signal. JJ intervals extracted from BCG were shown to have potential to be surrogate of ECG RR series in heart rate variability analysis.

An integrated blood pressure measurement system for suppression of motion artifacts.

Accuracy in blood pressure (BP) estimation is essential for proper diagnosis and management of hypertension. Motion artifacts are considered external sources of inaccuracy and can be due to sudden arm motion, muscle tremor, shivering, and transport vehicle vibrations. In the proposed work, a new algorithmic stage is integrated in a non-invasive BP monitor. This stage suppresses the effect of the motion artifact and adjusts the pressure estimation before displaying it to users. The proposed stage is based on a 3-axis accelerometer signal, which helps in the accurate detection of the motion artifact. Both transient motion artifacts and artifact due to vibrations are suppressed using algorithms based on Empirical Mode Decomposition (EMD). Measurements with human subjects show that the proposed algorithms considerably improved the accuracy of the blood pressure estimates in comparison with the commonly-used conventional oscillometric algorithm that does not include an EMD-based stage for artifact suppression, and allowed the estimates to meet the requirements of the international ANSI/AAMI/ISO standard.

Characteristic Ratio-Independent Arterial Stiffness-Based Blood Pressure Estimation.

Noninvasive blood pressure (BP) measurement is an important tool for managing hypertension and cardiovascular disease. However, automated noninvasive BP measurement devices, which are usually based on the oscillometric method, do not always provide accurate estimation of BP. It has been found that change in arterial stiffness (AS) is an underlying mechanism of disagreement between an oscillometric BP monitor and a sphygmomanometer. This problem is addressed by incorporating parameters related to AS in the algorithm for BP measurement. Pulse transit time (PTT) is first used to estimate AS parameters, which are fixed into a model of the oscillometric envelope. This model can then be used to perform curve fitting to the measured signal using only four parameters: systolic BP, diastolic BP, mean BP, and lumen area at zero transmural pressure. The proposed technique is independent of the experimentally determined characteristic ratios that are commonly used in existing oscillometric methods. The accuracy of the proposed technique was evaluated by comparing with the same model without incorporation of AS, and with reference BP device measurements. The new method achieved standard deviation of error less than 8 mmHg and mean error less than 5 mmHg. The results show consistency with ANSI/AAMI SP-10 standard for noninvasive BP measurement techniques.

Event Recognition for Contactless Activity Monitoring Using Phase-Modulated Continuous Wave Radar.

OBJECTIVES: The use of remote sensing technologies such as radar is gaining popularity as a technique for contactless detection of physiological signals and analysis of human motion. This paper presents a methodology for classifying different events in a collection of phase modulated continuous wave radar returns. The primary application of interest is to monitor inmates where the presence of human vital signs amidst different, interferences needs to be identified. METHODS: A comprehensive set of features is derived through time and frequency domain analyses of the radar returns. The Bhattacharyya distance is used to preselect the features with highest class separability as the possible candidate features for use in the classification process. The uncorrelated linear discriminant analysis is performed to decorrelate, denoise, and reduce the dimension of the candidate feature set. Linear and quadratic Bayesian classifiers are designed to distinguish breathing, different human motions, and nonhuman motions. The performance of these classifiers is evaluated on a pilot dataset of radar returns that contained different events including breathing, stopped breathing, simple human motions, and movement of fan and water. RESULTS: Our proposed pattern classification system achieved accuracies of up to 93% in stationary subject detection, 90% in stop-breathing detection, and 86% in interference detection. CONCLUSION: Our proposed radar pattern recognition system was able to accurately distinguish the predefined events amidst interferences. SIGNIFICANCE: Besides inmate monitoring and suicide attempt detection, this paper can be extended to other radar applications such as home-based monitoring of elderly people, apnea detection, and home occupancy detection.

Bayesian fusion algorithm for improved oscillometric blood pressure estimation.

A variety of oscillometric algorithms have been recently proposed in the literature for estimation of blood pressure (BP). However, these algorithms possess specific strengths and weaknesses that should be taken into account before selecting the most appropriate one. In this paper, we propose a fusion method to exploit the advantages of the oscillometric algorithms and circumvent their limitations. The proposed fusion method is based on the computation of the weighted arithmetic mean of the oscillometric algorithms estimates, and the weights are obtained using a Bayesian approach by minimizing the mean square error. The proposed approach is used to fuse four different oscillometric blood pressure estimation algorithms. The performance of the proposed method is evaluated on a pilot dataset of 150 oscillometric recordings from 10 subjects. It is found that the mean error and standard deviation of error are reduced relative to the individual estimation algorithms by up to 7 mmHg and 3 mmHg in estimation of systolic pressure, respectively, and by up to 2 mmHg and 3 mmHg in estimation of diastolic pressure, respectively.

Predicting Modulation in Corticomotor Excitability and in Transcallosal Inhibition in Response to Anodal Transcranial Direct Current Stimulation.

INTRODUCTION: Responses to neuromodulatory protocols based either on transcranial direct current stimulation (tDCS) or transcranial magnetic stimulation (TMS) are known to be highly variable between individuals. In this study, we examined whether variability of responses to anodal tDCS (a-tDCS) could be predicted from individual differences in the ability to recruit early or late indirect waves (I-waves), as reflected in latency differences of motor evoked potentials (MEPs) evoked by TMS of different coil orientation. METHODS: Participants (n = 20) first underwent TMS to measure latency of MEPs elicited at different coil orientations (i.e., PA, posterior-anterior; AP, anterior-posterior; LM, latero-medial). Then, participants underwent a-tDCS (20 min @ 2 mA) targeting the primary motor cortex of the contralateral preferred hand (right, n = 18). Individual responses to a-tDCS were determined by monitoring changes in MEP amplitude at rest and in the duration of the contralateral silent period (cSP) and ipsilateral silent period (iSP) during contraction; the latter providing an index of the latency and duration of transcallosal inhibition (LTI and DTI). RESULTS: Consistent with previous reports, individual responses to a-tDCS were highly variable when expressed in terms of changes in MEP amplitude or in cSP duration with ~50% of the participants showing either little or no modulation. In contrast, individual variations in measures of transcallosal inhibition were less variable, allowing detection of significant after-effects. The reduced LTI and prolonged DTI observed post-tDCS were indicative of an enhanced excitability of the transcallosal pathway in the stimulated hemisphere. In terms of predictions, AP-LM latency differences proved to be good predictors of responses to a-tDCS when considering MEP modulation. CONCLUSION: The present results corroborate the predictive value of latency differences derived from TMS to determine who is likely to express "canonical" responses to a-tDCS in terms of MEP modulation. The results also provide novel suggestive evidencethat a-tDCS can modulate the excitability of the transcallosal pathway of the stimulated hemisphere.

Extraction of Cole parameters from the electrical bioimpedance spectrum using stochastic optimization algorithms.

Fitting the measured bioimpedance spectroscopy (BIS) data to the Cole model and then extracting the Cole parameters is a common practice in BIS applications. The extracted Cole parameters then can be analysed as descriptors of tissue electrical properties. To have a better evaluation of physiological or pathological properties of biological tissue, accurate extraction of Cole parameters is of great importance. This paper proposes an improved Cole parameter extraction based on bacterial foraging optimization (BFO) algorithm. We employed simulated datasets to test the performance of the BFO fitting method regarding parameter extraction accuracy and noise sensitivity, and we compared the results with those of a least squares (LS) fitting method. The BFO method showed better robustness to the noise and higher accuracy in terms of extracted parameters. In addition, we applied our method to experimental data where bioimpedance measurements were obtained from forearm in three different positions of the arm. The goal of the experiment was to explore how robust Cole parameters are in classifying position of the arm for different people, and measured at different times. The extracted Cole parameters obtained by LS and BFO methods were applied to different classifiers. Two other evolutionary algorithms, GA and PSO were also used for comparison purpose. We showed that when the classifiers are fed with the extracted feature sets by BFO fitting method, higher accuracy is obtained both when applying on training data and test data.

Using bioimpedance spectroscopy parameters as real-time feedback during tDCS.

An exploratory analysis is carried out to investigate the feasibility of using BioImpedance Spectroscopy (BIS) parameters, measured on scalp, as real-time feedback during Transcranial Direct Current Stimulation (tDCS). TDCS is shown to be a potential treatment for neurological disorders. However, this technique is not considered as a reliable clinical treatment, due to the lack of a measurable indicator of treatment efficacy. Although the voltage that is applied on the head is very simple to measure during a tDCS session, changes of voltage are difficult to interpret in terms of variables that affect clinical outcome. BIS parameters are considered as potential feedback parameters, because: 1) they are shown to be associated with the DC voltage applied on the head, 2) they are interpretable in terms of conductive and capacitive properties of head tissues, 3) physical interpretation of BIS measurements makes them prone to be adjusted by clinically controllable variables, 4) BIS parameters are measurable in a cost-effective and safe way and do not interfere with DC stimulation. This research indicates that a quadratic regression model can predict the DC voltage between anode and cathode based on parameters extracted from BIS measurements. These parameters are extracted by fitting the measured BIS spectra to an equivalent electrical circuit model. The effect of clinical tDCS variables on BIS parameters needs to be investigated in future works. This work suggests that BIS is a potential method to be used for monitoring a tDCS session in order to adjust, tailor, or personalize tDCS treatment protocols.

Lasting depression in corticomotor excitability associated with local scalp cooling.

In this study, we investigated the effect of local scalp cooling on corticomotor excitability with transcranial magnetic simulation (TMS). Participants (healthy male adults, n=12) were first assessed with TMS to derive baseline measure of excitability from motor evoked potentials (MEPs) using the right first dorsal interosseous as the target muscle. Then, local cooling was induced on the right hemi-scalp (upper frontal region ∼ 15 cm(2)) by means of a cold wrap. The cooling was maintained for 10-15 min to get a decrease of at least 10°C from baseline temperature. In the post-cooling period, both scalp temperature and MEPs were reassessed at specific time intervals (i.e., T0, T10, T20 and T30 min). Scalp surface temperatures dropped on average by 12.5°C from baseline at T0 (p<0.001) with partial recovery at T10 (p<0.05) and full recovery at T20. Parallel analysis of post-cooling variations in MEP amplitude revealed significant reductions relative to baseline at T0, T10 and T20. No concurrent change in MEP latency was observed. A secondary control experiment was performed in a subset of participants (n=5) to account for the mild discomfort associated with the wrapping procedure without the cooling agent. Results showed no effect on any of the dependent variables (temperature, MEP amplitude and latency). To our knowledge, this report provides the first neurophysiological evidence linking changes in scalp temperature with lasting changes in corticomotor excitability.