Pleural effusion (PE) affects millions of people and is caused by a wide variety of underlying diseases. The excess of fluid in the pleural cavity leads to shortness of breath and increased risk for infections that require timely interventions. Yet, existing clinical assessments lack the temporal resolution for continuous monitoring of PE and associated respiratory symptoms. Novel wearable technologies capable of measuring multi-frequency transthoracic bioimpedance (BioZ) continuously hold promise for PE monitoring. In this study, we evaluated the feasibility of monitoring a bilateral PE in a pig model using continuous multi-frequency transthoracic BioZ obtained with a previously-validated wearable device. A novel multi-resolution approach yielded strong and significant correlations to the infused PE at the DClbaseline, respiratory windows, and breath-by-breath resolutions. A total of 1.65 L of saline was infused bilaterally, first left then right, in steps of 275 mL. The low-frequency DC resistance yielded the strongest correlation to the PE size (R2=O.99, −14%/L, p

Recent advances in wearable sensing platforms and robust, dry electrode materials have enabled adhesive-free, continuous bioimpedance sensing [1]. However, long-term continuous monitoring of bioimpedance requires high-impedance dry-electrodes which can exacerbate input-impedance mismatch errors. We present a modified mathematical model and design methodology for designing a high-input impedance buffer to improve voltage sensing in four-electrode bioimpedance applications. The bootstrapped buffer design presented confers nearly 75% less common-mode to differential-mode conversion at low-frequencies, and 50% less error at high-frequencies based on the worst-case model.The voltage buffer was directly used and alternated with the baseline inputs inputs of a commercially available off the shelf integrated circuit (AD5940) to determine the extent of transient bioimpedance errors emerging from input impedance mismatch. In a small pilot study of two young male adults, we show the system reduces the time for the 5% convergence time of the final bioimpedance by 100.7s seconds on average. The voltage buffer typically reduces bioimpedance errors by 20% compared to the baseline inputs based on exponential decay curves of best-fit in dry-electrode applications.

Acute decompensated heart failure (ADHF) is one of the leading causes for hospitalization in the US. Patient’s weight and limb girth are monitored to assess medication efficacy and readiness for hospital discharge. Such metrics are assessed manually by a trained professional at discrete time intervals, thereby missing potentially critical information about treatment responsiveness. Full, upper, and lower-body bioimpedance analysis have shown promise in tracking changes in patient weight, but requires bulky systems to place electrodes across different limbs. In this work, we evaluate the use of convenient ankle limb bioimpedance for tracking fluid status changes in a proof-of-concept dataset obtained in 9 hospitalized patients with ADHF. Continuous bioimpedance measurements were recorded while the patients received diuretic treatment for more than 2 days. Biomarkers derived from the low-frequency bioimpedance measurements, ratio of low-to-high bioimpedance measurements and limb girth were significantly different from hospital admission to discharge. Bioimpedance measures correlated to changes in weight (Pearson’s r = − 0.63, subject-specific), and limb girth (Pearson’s r = − 0.83 and r = − 0.87, subject-specific and global, respectively). Such bioimpedance measures captured from the ankle in a wearable form-factor and in a continuous manner provide better insight to the patient’s acute response to medication. Such a tool can be further used to better schedule medications and alter dosage in the clinic and be sent home with patients to detect fluid accumulation early.

Stress leads to widespread peripheral effects, including manifestations of respiratory distress. The combination of respiratory monitoring and non-invasive Peripheral Nerve Stimulation (PNS) represents a promising technological approach to quantify and reduce the physiological manifestations of stress. Such technologies require non-invasive, reliable respiratory information with adequate signal quality during various conditions. In this work, we computed an established respiratory quality index (RQI) on four respiratory signals derived from three common respiratory sensing modalities (respiratory effort (RSP), electrocardiogram-derived respiration (EDR), and Impedance Pneumography (IP)) in a complex stress protocol wherein 15 subjects underwent three stressors while receiving 1 of 3 active PNS modalities or sham. Our results indicate that the RQI of all signals changed substantially throughout the protocol with a maximal reduction from baseline of 21.28% (p< 0.001) during stressors involving speech. Further, RSP and IP resulted in the highest average quality and both were significantly higher in quality than the two EDR signals (p < 0.001). These results provide unique information that may inform the design of new technologies and studies leveraging respiratory markers for continuous stress detection and mitigation.

Chronic respiratory diseases affect millions and are leading causes of death in the US and worldwide. Pulmonary auscultation provides clinicians with critical respiratory health information through the study of Lung Sounds (LS) and the context of the breathing-phase and chest location in which they are measured. Existing auscultation technologies, however, do not enable the simultaneous measurement of this context, thereby potentially limiting computerized LS analysis. In this work, LS and Impedance Pneumography (IP) measurements were obtained from 10 healthy volunteers while performing normal and forced-expiratory (FE) breathing maneuvers using our wearable IP and respiratory sounds (WIRS) system. Simultaneous auscultation was performed with the Eko CORE stethoscope (EKO). The breathing-phase context was extracted from the IP signals and used to compute phase-by-phase (Inspiratory (I), expiratory (E), and their ratio (I:E)) and breath-by-breath acoustic features. Their individual and added value was then elucidated through machine learning analysis. We found that the phase-contextualized features effectively captured the underlying acoustic differences between deep and FE breaths, yielding a maximum F1 Score of 84.1 ±11.4% with the phase-by-phase features as the strongest contributors to this performance. Further, the individual phase-contextualized models outperformed the traditional breath-by-breath models in all cases. The validity of the results was demonstrated for the LS obtained with WIRS, EKO, and their combination. These results suggest that incorporating breathing-phase context may enhance computerized LS analysis. Hence, multimodal sensing systems that enable this, such as WIRS, have the potential to advance LS clinical utility beyond traditional manual auscultation and improve patient care.

Impedance pneumography (IP), a measure of the changes in the lung and thoracic bioimpedance, holds promise for non-invasive monitoring of pulmonary health. A key limitation of IP is the need for complex and frequent calibrations that require the subject to perform various maneuvers. In this work, we explore different calibration approaches to reduce the effects of inter-subject variability and postural changes on IP by utilizing a dual-frequency calibration approach. Dual-frequency IP was deployed for the first time in this work and its performance in estimating tidal volume (TV) was evaluated and compared to the conventional single frequency approaches. TV values obtained from a spirometer estimated with the subject- and posture-specific IP calibration approach are shown to correlate highly with the ground truth TV$(r > 0.9)$in all postures, including supine, left/right lateral, and seated postures for both 5 kHz and 100 kHz IP signals. Eliminating posture specificity results in a correlation of$r > 0.8$• With the globalized calibration approach that does not require any subject or posture-specific calibration, a correlation of$r=0.75$was achieved with the dual-frequency approach, and this was higher than the corresponding correlation of around$r=0.68$using any single frequency. This result has implications for the feasibility of dual-frequency IP for mitigating inter-subject variability and posture-specific calibrations.