Wearable sensor derived decompensation index for continuous remote monitoring of COVID-19 diagnosed patients

The COVID-19 pandemic has accelerated the adoption of innovative healthcare methods, including remote patient monitoring. In the setting of limited healthcare resources, outpatient management of individuals newly diagnosed with COVID-19 was commonly implemented, some taking advantage of various personal health technologies, but only rarely using a multi-parameter chest-patch for continuous monitoring. Here we describe the development and validation of a COVID-19 decompensation index (CDI) model based on chest patch-derived continuous sensor data to predict COVID-19 hospitalizations in outpatient-managed COVID-19 positive individuals, achieving an overall AUC of the ROC Curve of 0.84 on 308 event negative participants, and 22 event positive participants, out of an overall study cohort of 400 participants. We retrospectively compare the performance of CDI to standard of care modalities, finding that the machine learning model outperforms the standard of care modalities in terms of both numbers of events identified and with a lower false alarm rate. While only a pilot phase study, the CDI represents a promising application of machine learning within a continuous remote patient monitoring system.

A flexible electronic strain sensor for the real-time monitoring of tumor progression

Healthcare professionals and scientists utilize tumor shrinkage as a key metric to establish the efficacy of cancer treatments. However, current measurement tools such as CT scanners and calipers only provide brief snapshots of the dynamic geometric changes occurring in vivo, and they are unable to detect the micrometer-scale volumetric transformations transpiring at minute timescales. Here we present a stretchable electronic strain sensor, with a 10-micron scale resolution, capable of continuously monitoring tumor volume progression in real-time. In mouse models with subcutaneously implanted lung cancer or B-cell lymphoma tumors our sensors discerned a significant change in the tumor volumes of treated mice within 5 hours after small molecule therapy or immunotherapy initiation. Histology, caliper measurements, and luminescence imaging over a one-week treatment period validated the data from the continuous sensor. We anticipate that real-time tumor progression datasets could help expedite and automate the process of screening cancer therapies in vivo.

Using Learned Health Indicators and Deep Sequence Models to Predict Industrial Machine Health

In this paper, we describe a machine learning approach for predicting machine health indicators with a large time horizon into the future. The approach uses state-of-the-art neural network architectures for sequence modelling and can incorporate numerical-sensor and categorical data using entity embeddings. Moreover, we describe an unsupervised labelling approach where classes are generated using continuous sensor values in the training data and a clustering algorithm. To validate our approach, we performed an ablation study to verify the effectiveness of each of our model’s components. In this context, we show that entity embeddings can be used to generate effective features from categorical inputs, that state-of-the-art models, while originally developed for a different set of problems, can nonetheless be transferred to perform industrial asset health classification and provide a performance boost over simpler networks that have been traditionally used, such as relatively shallow recurrent or convolutional networks. Taken together, we present a machine health monitoring system that can accurately generate asset health predictions. This system can incorporate both numerical and categorical information, the current state-of-the-art for sequence modelling, and generate labels in an unsupervised fashion when explicit labels are unavailable.

A Sensor Data Stream Collection Scheme Considering Phase Differences for Load Balancing

In the internet of things (IoT), various devices (things) including sensors generate data and publish them via the internet. The authors define continuous sensor data with difference cycles as a sensor data stream and have proposed methods to collect distributed sensor data streams. In this paper, the authors describe a skip graph-based collection scheme for sensor data streams considering phase differences. In the proposed scheme considering phase differences, the collection time is balanced within each collection cycle by the phase differences, and the probability of load concentration to the specific time or node is decreased. The simulation results show that the proposed scheme can equalize the loads of nodes even if the distribution of collection cycles is not uniform.

Real-Time Detection of Fouling-Layer with a Non-Intrusive Continuous Sensor (NICS) during Thermal Processing in Food Manufacturing

The fouling of indirect shell and coil heat exchanger by heavy whipping cream (HWC) and non-fat dry milk (NFDM) was studied at aseptic Ultra-High Temperature (UHT) processing conditions (140 °C) using a novel non-intrusive sensor. The sensor emitted a heat pulse intermittently throughout the duration of the process causing an incremental increase in temperature at the tube external surface. The temperature response of the sensor varied due to the radial growth of the fouling layer formed by certain components of the products. Each heating pulse and the temperature response was studied to estimate the thermal conductivity of the fouling layer using inverse problems and parameter estimation. The changes in thermal conductivity were used as an indication of the fouling layer development during food processing at UHT temperatures. The estimated parameters from experimental results showed a decreasing trend in the thermal conductivity of HWC and NFDM from 0.35 to 0.10 and 0.63 to 0.37, respectively. An image analysis tool was developed and used to measure the fouling layer thickness at the end of each trial. The measured thickness was found to be 0.58 ± 0.15 for HWC and 0.56 ± 0.07 mm for NFDM. The fouling layer resistance for HWC and NFDM was 5.95 × 10−3 ± 1.53 × 10−3 and 1.53 × 10−3 ± 2.0 × 10−4 (m2K)/W, respectively.

Sensors for Continuous Monitoring of Surgeon’s Cognitive Workload in the Cardiac Operating Room

Monitoring healthcare providers’ cognitive workload during surgical procedures can provide insight into the dynamic changes of mental states that may affect patient clinical outcomes. The role of cognitive factors influencing both technical and non-technical skill are increasingly being recognized, especially as the opportunities to unobtrusively collect accurate and sensitive data are improving. Applying sensors to capture these data in a complex real-world setting such as the cardiac surgery operating room, however, is accompanied by myriad social, physical, and procedural constraints. The goal of this study was to investigate the feasibility of overcoming logistical barriers in order to effectively collect multi-modal psychophysiological inputs via heart rate (HR) and near-infrared spectroscopy (NIRS) acquisition in the real-world setting of the operating room. The surgeon was outfitted with HR and NIRS sensors during aortic valve surgery, and validation analysis was performed to detect the influence of intra-operative events on cardiovascular and prefrontal cortex changes. Signals collected were significantly correlated and noted intra-operative events and subjective self-reports coincided with observable correlations among cardiovascular and cerebral activity across surgical phases. The primary novelty and contribution of this work is in demonstrating the feasibility of collecting continuous sensor data from a surgical team member in a real-world setting.

Ethics and Privacy Considerations Before Deploying Sensor Technologies for Exposure Assessment in the Workplace: Results of a Structured Discussion Amongst Dutch Stakeholders

Will sensor-based exposure assessment be the future in workplace settings? Static instruments with embedded sensors are already applied to monitor levels of dangerous substances—in the context of acute health effects—at critical locations. However, with wearable, lightweight, miniaturized (low-cost) sensors developing quickly, much more is possible with sensors in relation to exposure assessment. Sensors can be applied in the work environment, on machines, or on employees and may include sensors that measure chemical exposures, but also sensors or other technologies that collect contextual information to support the exposure measurements. Like every technology it also has downsides. Sensors collect data on individuals that, depending on the purpose, need to be shared with others (e.g. health, safety and environment manager). One can imagine that people are afraid of misuse. To explore possible ethical and privacy issues that may come along with the introduction of sensors in the workplace, we organized a workshop with stakeholders (n = 32) to discuss three possible sensor-based scenarios in a structured way around five themes: purpose, efficacy, intrusiveness, proportionality, and fairness. The main conclusion of the discussions was that stakeholders currently see benefits in using sensors for applied targeted studies (short periods, clear reasons). In order to find acceptance for the implementation of sensors, all individuals affected by the sensors or its data need to be involved in the decisions on the purpose and application of sensors. Possible negative side effects need to be discussed and addressed. Continuous sensor-based monitoring of workers currently appears to be a bridge too far for the participants of this workshop.

Developing Sensor Proxies for “Chemical Cocktails” of Trace Metals in Urban Streams

Understanding transport mechanisms and temporal patterns in the context of metal concentrations in urban streams is important for developing best management practices and restoration strategies to improve water quality. In some cases, in-situ sensors can be used to estimate unknown concentrations of trace metals or to interpolate between sampling events. Continuous sensor data from the United States Geological Survey were analyzed to determine statistically significant relationships between lead, copper, zinc, cadmium, and mercury with turbidity, specific conductance, dissolved oxygen, and discharge for the Hickey Run, Watts Branch, and Rock Creek watersheds in the Washington, D.C. region. We observed a significant negative linear relationship between concentrations of Cu and dissolved oxygen at Rock Creek (p < 0.05). Sometimes, turbidity had significant positive linear relationships with Pb and Hg concentrations. There were negative or positive linear relationships between Pb, Cd, Zn, and Hg and specific conductance. There also appeared to be relationships between watershed areal fluxes of Pb, Cu, Zn, and Cd in streams with turbidity. Watershed monitoring approaches using continuous sensor data have the potential to characterize the frequency, magnitude, and composition of pulses in concentrations and loads of trace metals, which could improve the management and restoration of urban streams.