Sensor-based Measurement Research Articles

Machine Learning-Aided Remote Monitoring of NOx Emissions from Heavy-Duty Diesel Vehicles Based on OBD Data Streams

Most of the current, popular approaches to monitoring real driving NOx emissions are based on direct measurement. However, due to the uncertainty of sensor-based measurements, such methods cannot always be used to accurately screen out the malfunctions of an emission control system. In this paper, a random forest (RF) model which extracts information from on-board diagnostics (OBD) data streams transmitted by a remote emission management vehicle terminal (REMVT) is proposed to provide a specific emission method for the online screening of high NOx emissions. First, two particular forms of modeling, random forest and logistic regression (LR), are laid out as representatives of nonparametric models and specified linear models. These two models were trained, validated and compared using OBD data collected from three China-VI heavy-duty diesel vehicles (HDDVs). The results show that as a data-driven, highly adaptive and robust learning method, the RF model can more accurately identify an abnormal emission state. Finally, a further validation was conducted, in which another China-VI HDDV was tested in two typical states, including a fault state and a normal state. The results indicated that the RF model could clearly distinguish the out-of-control emission condition from the normal operation state. The outcome of this research verifies the feasibility of using a machine learning model to process remote OBD data on HD vehicles and to identify high emissions in the case of an in-use fleet. On this basis, more sophisticated combined models and multi-stage models could be developed.

Action Observation Training to Improve Upper Limb Function in Infants with Unilateral Brain Lesion – a Feasibility Study

ABSTRACT In this study, the feasibility of Action Observation Training (AOT) in combination with sensor-based measurements in infants at high risk of Unilateral Spastic Cerebral Palsy (UCP) were evaluated. Over a four-week period, eight infants at high risk of UCP performed AOT at home while wearing sensors with assistance of caregivers. Sensor data were compared to clinical assessments of upper limb function, the Hand Assessment for Infants (HAI) and the Mini-Assisting Hand Assessment (Mini-AHA). AOT training time and acceptance by the caregivers were considered as indicators for feasibility. The excellent training adherence and positive feedback of the caregivers showed that the AOT was feasible in this patient group and setting. Sensor measurements were accepted and displayed significant correlations with hand function. These preliminary results indicate the potential of wearable sensors to record upper limb function over the course of AOT for infants at high risk of UCP. Thus, AOT in combination with sensor measurements are proposed as a feasible training tool to complement usual care.

Wrist-worn sensor-based measurements for drug effect detection with small samples in people with Lewy Body Dementia

IntroductionFew late-stage clinical trials in Parkinson's disease (PD) have produced evidence on the clinical validity of sensor-based digital measurements of daily life activities to detect responses to treatment. The objective of this study was to assess whether digital measures from patients with mild-to-moderate Lewy Body Dementia demonstrate treatment effects during a randomized Phase 2 trial. MethodsSubstudy within a 12-week trial of mevidalen (placebo vs 10, 30, or 75 mg), where 70/344 patients (comparable to the overall population) wore a wrist-worn multi-sensor device. ResultsTreatment effects were statistically significant by conventional clinical assessments (Movement Disorder Society-Unified Parkinson's Disease Rating Scale [MDS-UPDRS] sum of Parts I-III and Alzheimer's Disease Cooperative Study-Clinical Global Impression of Change [ADCS-CGIC] scores) in the full study cohort at Week 12, but not in the substudy. However, digital measurements detected significant effects in the substudy cohort at week 6, persisting to week 12. ConclusionsDigital measurements detected treatment effects in a smaller cohort over a shorter period than conventional clinical assessments. Trial registrationclinicaltrials.gov, NCT03305809

Empowering drug development: Leveraging insights from imaging technologies to enable the advancement of digitalhealth technologies.

The US Food and Drug Administration (FDA) has publicly recognized the importance of improving drug development efficiency, deeming translational biomarkers a top priority. The use of imaging biomarkers has been associated with increased rates of drug approvals. An appropriate level of validation provides a pragmatic way to choose and implement these biomarkers. Standardizing imaging modality selection, data acquisition protocols, and image analysis (in ways that are agnostic to equipment and algorithms) have been key to imaging biomarker deployment. The best known examples come from studies done via precompetitive collaboration efforts, which enable input from multiple stakeholders and data sharing. Digital health technologies (DHTs) provide an opportunity to measure meaningful aspects of patient health, including patient function, for extended periods of time outside of the hospital walls, with objective, sensor-based measures. We identified the areas where learnings from the imaging biomarker field can accelerate the adoption and widespread use of DHTs to develop novel treatments. As with imaging, technical validation parameters and performance acceptance thresholds need to be established. Approaches amenable to multiple hardware options and data processing algorithms can be enabled by sharing DHT data and by cross-validating algorithms. Data standardization and creation of shared databases will be vital. Pre-competitive consortia (public-private partnerships and professional societies that bring together all stakeholders, including patient organizations, industry, academic experts, and regulators) will advance the regulatory maturity of DHTs in clinical trials.

Development of a LoRaWAN IoT Node with Ion-Selective Electrode Soil Nitrate Sensors for Precision Agriculture

Crop productivity is highly dependent on the availability of soluble nitrogen (N), e.g. nitrate, in soil. When N levels are low, fertilisers are applied to replenish the soil's reserves. Typically the timing of these applications is based on paper-based guidance and sensor-based measurements of canopy greenness, which provides an indirect measure of soil N status. However this approach often means that N fertiliser is applied inappropriately or too late, resulting in excess N being lost to the environment, or too little N to meet crop demand. To promote greater N use efficiency and improve agricultural sustainability, we developed an Internet of Things (IoT) approach for the real-time measurement of soil nitrate levels using ion-selective membrane sensors in combination with digital soil moisture probes. The node incorporates state-of-the-art IoT connectivity using a LoRaWAN transceiver. The sensing platform can transfer real-time data via a cloud-connected gateway for processing and storage. In summary, we present a validated soil sensor system for real-time monitoring of soil nitrate concentrations, which can support fertiliser management decisions, improve N use efficiency and reduce N losses to the environment.

Nursing Assessment of Pressure Injury Risk with the Braden Scale Validated against Sensor-Based Measurement of Movement.

Nursing staff assessment to accurately identify pressure injury (PrI) risk is a hallmark in PrI prevention care. Risk scores from the Braden Scale for Predicting Pressure Sore Risk© (hereafter Braden), a commonly used tool for assessing PrI risk, signal the need for preventative care. Braden Mobility, Activity, and Sensory Perception subscale subgroups associated with repositioning movement features help identify preventative strategies that minimize pressure intensity and duration. Evidence confirming subscale rating accuracy is needed. This study compared assessment score accuracy with movement data collected via accelerometer sensor. Sample included 913 nursing home residents from the Turn Everyone and Move for Pressure Ulcer Prevention (TEAM-UP) cluster randomized trial. Movements and Braden Mobility and Activity subscale scores were evaluated for significant differences and associations. Mobility subgroups explained a small-medium amount of variance in mean lying and upright movement features (0.002 ≤ R2 ≤ 0.195). Activity subgroups explained a small-medium amount of variance in mean lying, upright, and ambulating movements (0.016 ≤ R2 ≤ 0.248). Significant associations occurred among subscale subgroups and most movements. Nursing assessment ratings using Braden scale's Mobility and Activity subscale scores are accurate indicators of actual repositioning movements and can be relied upon for PrI prevention care planning for older adults.

A Novel Automated RGB-D Sensor-Based Measurement of Voluntary Items of the Fugl-Meyer Assessment for Upper Extremity: A Feasibility Study.

Motor function assessment is essential for post-stroke rehabilitation, while the requirement for professional therapists’ participation in current clinical assessment limits its availability to most patients. By means of sensors that collect the motion data and algorithms that conduct assessment based on such data, an automated system can be built to optimize the assessment process, benefiting both patients and therapists. To this end, this paper proposed an automated Fugl-Meyer Assessment (FMA) upper extremity system covering all 30 voluntary items of the scale. RGBD sensors, together with force sensing resistor sensors were used to collect the patients’ motion information. Meanwhile, both machine learning and rule-based logic classification were jointly employed for assessment scoring. Clinical validation on 20 hemiparetic stroke patients suggests that this system is able to generate reliable FMA scores. There is an extremely high correlation coefficient (r = 0.981, p < 0.01) with that yielded by an experienced therapist. This study offers guidance and feasible solutions to a complete and independent automated assessment system.

A structured-light 3D sensor-based shape measurement method for casting allowance evaluation

The evaluation of casting allowance currently relies on manual operation, which is time-consuming and unstable. The structured-light three-dimensional (3D) sensor-based 3D shape measurement technology is characterized by non-contact, high accuracy, and fast measuring speed, which provides the complete 3D shape of casting for accurate allowance evaluation. However, measuring a complex shape casting still requires numerous manual operations, including attaching marks, planning the measurement process, and data processing. To solve these problems, a robot-driven structured-light 3D sensor-based measurement method is proposed, with a six-degree of freedom (6-DOF) industrial robot and a turntable to control the structured-light 3D measurement sensor, to transform the 3D results from different measuring viewpoints into a coherent coordinate system, and to form a complete 3D shape of the casting. Then, the mechanical processing allowance will be calculated by comparing the 3D measurement result and the standard CAD model automatically. The experiment results show that the proposed method is accurate and efficient, and the casting allowance evaluation time is about ten times faster than that of manual operation. Moreover, the proposed method would provide valuable experience for other industrial applications.

Reliability of Lower-Cost Sensors in the Analysis of Indoor Air Quality on Board Ships

Air quality in and around ships is governed by a variety of pollution sources that are unique for the shipping context. This makes the living and working conditions on ships substantially different from situations in cities or inside buildings. To gain insight into these differences, information about trends and absolute pollutant amounts on board ships is needed. However, the installation of reference instruments to monitor NO2, NO, O3, particulate matter and other environmental parameters is often not possible because of their size, weight or because of safety reasons. For that reason, more compact devices incorporating a variety of sensors are a good alternative. However, the use of such sensors is only possible when their behaviour and performance in a shipping context are well understood. To study this context, we were allowed to compare sensor-based measurements performed on a 36-year old ship dedicated to near shore operations with measurements of reference-grade instruments. Additional behavioural information of sensors is obtained by measuring campaigns organized on several inland ships. This contribution demonstrates that trends registered by gas and particulate matter sensors are reliable but that insufficient detection limits, higher noise, imperfect calibration and sensor errors result in some reliability constraints.

Machine Vision-Based Fatigue Crack Propagation System.

This paper introduces a machine vision-based system promising low-cost solution for detecting a fatigue crack propagation caused by alternating mechanical stresses. The fatigue crack in technical components usually starts on surfaces at stress concentration points. The presented system was designed to substitute a strain gauge sensor-based measurement using an industrial camera in cooperation with branding software. This paper presents implementation of a machine vision system and algorithm outputs taking on fatigue crack propagation samples.

Improvement of Fiber Bragg Grating Wavelength Demodulation System by Cascading Generative Adversarial Network and Dense Neural Network

A high-performance, low-cost demodulation system is essential for fiber-optic sensor-based measurement applications. This paper presents a demodulation system for FBG sensors based on a long-period fiber grating (LPG) driven by artificial intelligence techniques. The LPG is applied as an edge filter to convert the spectrum drift of the FBG sensor into transmitted intensity variation, which is subsequently fed to the proposed sensor demodulation network to provide high-precision wavelength interrogation. The sensor demodulation network consists of a generative adversarial network (GAN) for data augmentation and a dense neural network (DNN) for wavelength interrogation, the former addresses the drawback that traditional machine learning models rely on a large-scale dataset for satisfactory performance, while the latter is used to model the relationship between transmitted intensity and wavelength for demodulation. Experiments demonstrate that the proposed system has excellent performance and can achieve wavelength interrogation precision of ±3 pm. In addition, the effectiveness of the GAN is demonstrated. With a wide demodulation range, high performance, and low cost, the system can provide a new platform for fiber-optic sensor-based measurement applications.

Soft Landing Parameter Measurements for Candidate Navigation Trajectories Using Deep Learning and AI-Enabled Planetary Descent

Smart instruments, sensors, and AI technologies are playing an important role in many fields such as medical science, Earth science, astronomy physics, and space study. This article attempts to study the role of sensors, instruments, and AI (artificial intelligence) based smart technologies in lunar missions during navigation of trajectories. Lunar landing missions usually divide the power descent phase into three to four sub-phases. Each sub-phase has its own set of initial and final constraints for the desired system state. The landing systems depend on human competencies for making the most crucial landing decisions. Trajectory planning and designing are very significant in lunar missions, and it requires inputs with precision. The manual systems may be prone to errors. In contrast, AI and smart sensor-based measurements give an accurate idea about the trajectory paths and make appropriate decisions where manual systems may turn into disasters. The manual systems are either pre-fed or have manual controls to guide the trajectory. For autonomous landing problems, trajectory design is a very crucial task. The automated trajectories play a vital role in the measurement and prediction of landing state parameters of the space rocket. Nowadays, sensors, intelligent instruments, and the latest technologies go hand in hand to devise measurement methods for accurate calculations and make appropriate decisions during landing space rockets at the designated destination. Space missions are very expensive and require huge efforts to design smart systems for navigation trajectories. This paper attempts to design all possible candidates of reference navigation trajectories for autonomous lunar descent by employing 3D non-linear system dynamics with randomly chosen initial state conditions. The generated candidates do not rely on multiple hops and thus exhibit an ability to serve autonomous missions. This research work makes use of smart sensors and AI federated techniques for smartly training the system to serve the ultimate purpose. The trajectories are simulated in an automated simulating environment to perform exhaustive analyses. The results accurately approximate the trajectories analogous to their numerical counterparts and converge to their measured final state estimates. The generation rate of feasible trajectories measures the accuracy of the algorithm. The algorithm’s accuracy is near 0.87 for 100 sec flight time, which is reasonable.

Integrating Wearable Sensors and Video to Determine Microlocation-Specific Physiologic and Motion Biometrics-Method Development for Competitive Climbing.

Competitive indoor climbing has increased in popularity at the youth, collegiate, and Olympic levels. A critical aspect for improving performance is characterizing the physiologic response to different climbing strategies (e.g., work/rest patterns, pacing) and techniques (e.g., body position and movement) relative to location on climbing wall with spatially varying characteristics (e.g., wall inclinations, position of foot/hand holds). However, this response is not well understood due to the limited capabilities of climbing-specific measurement and assessment tools. In this study, we developed a novel method to examine time-resolved sensor-based measurements of multiple personal biometrics at different microlocations (finely spaced positions; MLs) along a climbing route. For the ML-specific biometric system (MLBS), we integrated continuous data from wearable biometric sensors and smartphone-based video during climbing, with a customized visualization and analysis system to determine three physiologic parameters (heart rate, breathing rate, ventilation rate) and one body movement parameter (hip acceleration), which are automatically time-matched to the corresponding video frame to determine ML-specific biometrics. Key features include: (1) biometric sensors that are seamlessly embedded in the fabric of an athletic compression shirt, and do not interfere with climbing performance, (2) climbing video, and (3) an interactive graphical user interface to rapidly visualize and analyze the time-matched biometrics and climbing video, determine timing sequence between the biometrics at key events, and calculate summary statistics. To demonstrate the capabilities of MLBS, we examined the relationship between changes in ML-specific climbing characteristics and changes in the physiologic parameters. Our study demonstrates the ability of MLBS to determine multiple time-resolved biometrics at different MLs, in support of developing and assessing different climbing strategies and training methods to help improve performance.

Relating subjective ride quality ratings to objective measures

Agencies have long used subjective roughness ratings from panels of users to inform policy development on road maintenance strategies. The commoditization of electronics motivated the development of more objective, automated, and cost-effective measurement technologies. Consequently, there has been an explosion of ensemble measurements using smartphones or connected vehicles. Nevertheless, agencies have no means of relating those sensor-based measurements to their customary linguistic scale of human perceived roughness levels. This research relates subjective ratings of roughness from regular passengers of public bus transit to simultaneous smartphone-based objective measures of roughness. The findings are that regular bus riders consistently distinguished between the extreme values of measured roughness but not the intermediate values. Ratings are also less distinguishable for smoother rides than for rougher rides. The experiments also reveal a phenomenon of roughness acclimation that leads to biased ratings from regular users of a road segment.

An examination of daily sleep characteristics and subsequent eating disorder behavior among individuals with binge-spectrum eating disorders.

Specific characteristics of sleep (e.g., duration, quality, and fatigue) are positively associated with (ED) behaviors, specifically binge eating (BE) potentially through decreased self-regulation and increased appetite. However, prior work has been largely cross-sectional and has not examined temporal relationships between sleep characteristics and next-day ED behaviors. Thus, the present study examined daily relationships between sleep and ED behaviors among individuals with binge-spectrum EDs. Participants (N = 96) completed 7 daily ecological momentary assessment (EMA) surveys over 7-14days; morning surveys assessed sleep characteristics and 6 randomly timed surveys each day captured ED behaviors. Analyses examined within-subject and between-subject effects of sleep quality, duration, and fatigue on BE, compensatory purging behaviors, and maladaptive exercise. Within-subject sleep quality was significantly negatively associated with engagement in maladaptive exercise later that day. Additionally, between-subject sleep duration was significantly negatively associated with engagement in compensatory purging behaviors. Within- and between-subjects associations between sleep quality and duration and compensatory behavior engagement indicate that sleep plays an important role in ED behaviors. Future research should incorporate sensor-based measurement of sleep and examine how specific facets of sleep impact BE and treatment response. Level II: Evidence obtained from controlled trial without randomization.

A Remote Digital Monitoring Platform to Assess Cognitive and Motor Symptoms in Huntington Disease: Cross-sectional Validation Study.

BackgroundRemote monitoring of Huntington disease (HD) signs and symptoms using digital technologies may enhance early clinical diagnosis and tracking of disease progression, guide treatment decisions, and monitor response to disease-modifying agents. Several recent studies in neurodegenerative diseases have demonstrated the feasibility of digital symptom monitoring.ObjectiveThe aim of this study was to evaluate a novel smartwatch- and smartphone-based digital monitoring platform to remotely monitor signs and symptoms of HD.MethodsThis analysis aimed to determine the feasibility and reliability of the Roche HD Digital Monitoring Platform over a 4-week period and cross-sectional validity over a 2-week interval. Key criteria assessed were feasibility, evaluated by adherence and quality control failure rates; test-retest reliability; known-groups validity; and convergent validity of sensor-based measures with existing clinical measures. Data from 3 studies were used: the predrug screening phase of an open-label extension study evaluating tominersen (NCT03342053) and 2 untreated cohorts—the HD Natural History Study (NCT03664804) and the Digital-HD study. Across these studies, controls (n=20) and individuals with premanifest (n=20) or manifest (n=179) HD completed 6 motor and 2 cognitive tests at home and in the clinic.ResultsParticipants in the open-label extension study, the HD Natural History Study, and the Digital-HD study completed 89.95% (1164/1294), 72.01% (2025/2812), and 68.98% (1454/2108) of the active tests, respectively. All sensor-based features showed good to excellent test-retest reliability (intraclass correlation coefficient 0.89-0.98) and generally low quality control failure rates. Good overall convergent validity of sensor-derived features to Unified HD Rating Scale outcomes and good overall known-groups validity among controls, premanifest, and manifest participants were observed. Among participants with manifest HD, the digital cognitive tests demonstrated the strongest correlations with analogous in-clinic tests (Pearson correlation coefficient 0.79-0.90).ConclusionsThese results show the potential of the HD Digital Monitoring Platform to provide reliable, valid, continuous remote monitoring of HD symptoms, facilitating the evaluation of novel treatments and enhanced clinical monitoring and care for individuals with HD.

Cardio-metabolic risk modeling and assessment through sensor-based measurements

ObjectiveCardio-metabolic risk assessment in the general population is of paramount importance to reduce diseases burdened by high morbility and mortality. The present paper defines a strategy for out-of-hospital cardio-metabolic risk assessment, based on data acquired from contact-less sensors. MethodsWe employ Structural Equation Modeling to identify latent clinical variables of cardio-metabolic risk, related to anthropometric, glycolipidic and vascular function factors. Then, we define a set of sensor-based measurements that correlate with the clinical latent variables. ResultsOur measurements identify subjects with one or more risk factors in a population of 68 healthy volunteers from the EU-funded SEMEOTICONS project with accuracy 82.4%, sensitivity 82.5%, and specificity 82.1%. ConclusionsOur preliminary results strengthen the role of self-monitoring systems for cardio-metabolic risk prevention.

The Need for Objective Physical Activity Measurements in Routine Bariatric Care

PurposeThis study aims to (1) quantify physical behavior through self-reports and sensor-based measures, (2) examine the correlation between self-reported and sensor-based physical activity (PA) and (3) assess whether bariatric patients adhere to PA guidelines.MethodsA Fitbit accelerometer was used to collect minute-to-minute step count and heart rate data for 14 consecutive days. Total physical activity levels (PAL), moderate-to-vigorous intensity physical activity (MVPA) and sedentary behavior (SB) were used to quantify physical behavior. Self-reported PA was assessed with the International Physical Activity Questionnaire (IPAQ). To analyze the association between sensor-based and self-reported PA, Spearman’s correlation was used. A minimum of 150 MVPA minutes per week was considered as compliance with the PA guidelines.ResultsFitbit data of 37 pre- and 18 post-surgery patients was analyzed. Participants averaged 7403 ± 3243 steps/day and spent most of their time sedentary (832 min, IQR: 749 – 879), especially in prolonged periods of ≥ 30 min (525, IQR: 419 – 641). Median MVPA time was 5.6 min/day (IQR: 1.7 – 10.6). Correlations between self-reported and sensor-based MVPA and SB were respectively 0.072 and 0.455. Only 17.1% was objectively adherent to MVPA guidelines ≥ 150 min/week, while 94.3% met the guidelines in case of self-reports.ConclusionPA quantification confirmed that bariatric patients are highly sedentary and rarely engage in MVPA, despite a relatively high daily step count. Moreover, bariatric patients are not able to assess MVPA and moderately their SB by self-reports. Our results indicate the need for sensor-based PA monitoring in routine bariatric care.Graphical abstract

Conventional and sensor-based streamflow data acquisition system for sustainable water resources management and agricultural applications: an extensive review of literature

This paper presents an extensive review of literature of conventional and sensor-based methods for streamflow data collection intended for sustainable water resources management and agricultural applications. Conventional methods reviewed are direct measurements, e.g., floating method, volumetric method, current meter, trajectory, dilution method, Acoustic Doppler current profiler and electromagnetic. Hydraulic structure methods are also included, and remote sensing and particle image velocimetry as non-contact approaches. This paper also presents a review of published literature on local studies in the Philippines that employed sensor-based measurement of streamflow for water resources management. A comparative analysis of the various methods was then made based on their applicability with respect to the characteristics of the streams, ease of operation, effectiveness, accuracy, and the cost. Results showed that for most cases, when resources are available, current meter method is still, the best method to be used due to its high accuracy, time efficiency and versatility. For hilly terrain which covers larger areas, Particle Image Velocimetry (PIV) is suitable with accurate results. For long term measurements, hydraulic structures maybe considered due to its high accuracy when installed properly and if the conditions are met. Additionally, when other methods are impractical and impossible to use, float method is best due to its advantage in operation and cost especially in flat and unobstructed terrain. For channels with small quantity of streamflow that converges into a single descend like a waterfall, volumetric method is suitable. Sensor-based methods are the most effective for real-time monitoring of streamflow and they are applicable to almost any terrain and river size. In selecting the most suitable method for streamflow monitoring, the characteristics of the river or stream, ease of access, accuracy, and availability of financial and physical resources must be considered. Finally, recommendations are offered to maximize the potential applicability of the most effective methods for sustainable water resources management and agricultural applications particularly in developing countries.

Objective Nontechnical Skills Measurement Using Sensor-based Behavior Metrics in Surgical Teams.

The purpose of this study was to identify objective measures that predict surgeon nontechnical skills (NTS) during surgery. NTS are cognitive and social skills that impact operative performance and patient outcomes. Current methods for NTS assessment in surgery rely on observation-based tools to rate intraoperative behavior. These tools are resource intensive (e.g., time for observation or manual labeling) to perform; therefore, more efficient approaches are needed. Thirty-four robotic-assisted surgeries were observed. Proximity sensors were placed on the surgical team and voice recorders were placed on the surgeon. Surgeon NTS was assessed by trained observers using the NonTechnical Skills for Surgeons (NOTSS) tool. NTS behavior metrics from the sensors included communication, speech, and proximity features. The metrics were used to develop mixed effect models to predict NOTSS score and in machine learning classifiers to distinguish between exemplar NTS scores (highest NOTSS score) and non-exemplar scores. NTS metrics were collected from 16 nurses, 12 assistants, 11 anesthesiologists, and four surgeons. Nineteen behavior features and overall NOTSS score were significantly correlated (12 communication features, two speech features, five proximity features). The random forest classifier achieved the highest accuracy of 70% (80% F1 score) to predict exemplar NTS score. Sensor-based measures of communication, speech, and proximity can potentially predict NOTSS scores of surgeons during robotic-assisted surgery. These sensing-based approaches can be utilized for further reducing resource costs of NTS and team performance assessment in surgical environments. Sensor-based assessment of operative teams' behaviors can lead to objective, real-time NTS measurement.