Dataset Information

Avatar-Based Patient Monitoring With Peripheral Vision: A Multicenter Comparative Eye-Tracking Study.

ABSTRACT:

Background

Continuous patient monitoring has been described by the World Health Organization as extremely important and is widely used in anesthesia, intensive care medicine, and emergency medicine. However, current state-of-the-art number- and waveform-based monitoring does not ideally support human users in acquiring quick, confident interpretations with low cognitive effort, and there are additional problematic aspects such as alarm fatigue. We developed a visualization technology (Visual Patient), specifically designed to help caregivers gain situation awareness quickly, which presents vital sign information in the form of an animated avatar of the monitored patient. We suspected that because of the way it displays the information as large, colorful, moving graphic objects, caregivers might be able to perform patient monitoring using their peripheral vision, which may facilitate quicker detection of anomalies, independently of acoustic alarms.

Objective

In this study, we tested the hypothesis that avatar-based monitoring, when observed with peripheral vision only, increases the number of perceptible changes in patient status as well as caregivers' perceived diagnostic confidence compared with a high-fidelity simulation of conventional monitoring, when observed with peripheral vision only.

Methods

We conducted a multicenter comparative study with a within-participant design in which anesthesiologists with their peripheral field of vision looked at 2 patient-monitoring scenarios and tried to identify changes in patient status. To ensure the best possible experimental conditions, we used an eye tracker, which recorded the eye movements of the participants and confirmed that they only looked at the monitoring scenarios with their peripheral vision.

Results

Overall, 30 participants evaluated 18 different patient status changes with each technology (avatar and conventional patient monitoring). With conventional patient monitoring, participants could only detect those 3 changes in patient status that are associated with a change in the auditory pulse tone display, that is, tachycardia (faster beeping), bradycardia (slower beeping), and desaturation (lower pitch of beeping). With the avatar, the median number of detected vital sign changes quadrupled from 3 to 12 (P<.001) in scenario 1, and more than doubled from 3 to 8 (P<.001) in scenario 2. Median perceived diagnostic confidence was confident for both scenarios with the avatar and unconfident in scenario 1 (P<.001), and very unconfident in scenario 2 (P=.024) with conventional monitoring.

Conclusions

This study introduces the concept of peripheral vision monitoring. The test performed showed clearly that an avatar-based display is superior to a standard numeric display for peripheral vision. Avatar-based monitoring could potentially make much more of the patient monitoring information available to caregivers for longer time periods per case. Our results indicate that the optimal information transmission would consist of a combination of auditory and avatar-based monitoring.

SUBMITTER: Pfarr J

PROVIDER: S-EPMC6668297 | biostudies-literature | 2019 Jul

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Avatar-Based Patient Monitoring With Peripheral Vision: A Multicenter Comparative Eye-Tracking Study.

Pfarr Juliane J Ganter Michael T MT Spahn Donat R DR Noethiger Christoph B CB Tscholl David W DW

Journal of medical Internet research 20190717 7

<h4>Background</h4>Continuous patient monitoring has been described by the World Health Organization as extremely important and is widely used in anesthesia, intensive care medicine, and emergency medicine. However, current state-of-the-art number- and waveform-based monitoring does not ideally support human users in acquiring quick, confident interpretations with low cognitive effort, and there are additional problematic aspects such as alarm fatigue. We developed a visualization technology (Vi ...[more]

PMID: 31317870

Similar Datasets

Project description:BACKGROUND:Patient monitoring is central to perioperative and intensive care patient safety. Current state-of-the-art monitors display vital signs as numbers and waveforms. Visual Patient technology creates an easy-to-interpret virtual patient avatar model that displays vital sign information as it would look in a real-life patient (eg, avatar changes skin color from healthy to cyanotic depending on oxygen saturation). In previous studies, anesthesia providers using Visual Patient perceived more vital signs during short glances than with conventional monitoring. OBJECTIVE:We aimed to study the deeper mechanisms underlying information perception in conventional and avatar-based monitoring. METHODS:In this prospective, multicenter study with a within-subject design, we showed 32 anesthesia providers four 3- and 10-second monitoring scenarios alternatingly as either routine conventional or avatar-based in random sequence. All participants observed the same scenarios with both technologies and reported the vital sign status after each scenario. Using eye-tracking, we evaluated which vital signs the participants had visually fixated (ie, could have potentially read and perceived) during a scenario. We compared the frequencies and durations of participants' visual fixations of vital signs between the two technologies. RESULTS:Participants visually fixated more vital signs per scenario in avatar-based monitoring (median 10, IQR 9-11 versus median 6, IQR 4-8, P<.001; median of differences=3, 95% CI 3-4). In multivariable linear regression, monitoring technology (conventional versus avatar-based monitoring, difference=-3.3, P<.001) was an independent predictor of the number of visually fixated vital signs. The difference was less prominent in the longer (10-second) scenarios (difference=-1.5, P=.04). Study center, profession, gender, and scenario order did not influence the differences between methods. In all four scenarios, the participants visually fixated 9 of 11 vital signs statistically significantly longer using the avatar (all P<.001). Four critical vital signs (pulse rate, blood pressure, oxygen saturation, and respiratory rate) were visible almost the entire time of a scenario with the avatar; these were only visible for fractions of the observations with conventional monitoring. Visual fixation of a certain vital sign was associated with the correct perception of that vital sign in both technologies (avatar: phi coefficient=0.358; conventional monitoring: phi coefficient=0.515, both P<.001). CONCLUSIONS:This eye-tracking study uncovered that the way the avatar-based technology integrates the vital sign information into a virtual patient model enabled parallel perception of multiple vital signs and was responsible for the improved information transfer. For example, a single look at the avatar's body can provide information about: pulse rate (pulsation frequency), blood pressure (pulsation intensity), oxygen saturation (skin color), neuromuscular relaxation (extremities limp or stiff), and body temperature (heatwaves or ice crystals). This study adds a new and higher level of empirical evidence about why avatar-based monitoring improves vital sign perception compared with conventional monitoring.

Project description:BackgroundA new patient monitoring technology called Visual Patient, which transforms numerical and waveform data into a virtual model (an avatar) of the monitored patient, has been shown to improve the perception of vital signs compared to conventional patient monitoring. In order to gain a deeper understanding of the opinions of potential future users regarding the new technology, we have analyzed the answers of two large groups of anesthetists using two different study methods.MethodsFirst, we carried out a qualitative analysis guided by the "consolidated criteria for reporting qualitative research" checklist. For this analysis, we interviewed 128 anesthesiologists, asking: "Where do you see advantages in Visual Patient monitoring?" and afterward identified major and minor themes in their answers. In a second study, an online survey with 38 anesthesiologists at two different institutions, we added a quantitative part in which anesthesiologists rated statements based on the themes identified in the prior analysis on an ordinal rating scale.ResultsWe identified four high-level themes: "quick situation recognition," "intuitiveness," "unique design characteristics," and "potential future uses," and eight subthemes. The quantitative questions raised for each major theme were: 1. "The Visual Patient technology enabled me to get a quick overview of the situation." (63% of the participants agreed or very much agreed to this statement). 2. "I found the Visual Patient technology to be intuitive and easy to learn." (82% agreed or very much agreed to this statement). 3. "The visual design features of the Visual Patient technology (e.g., the avatar representation) are not helpful for patient monitoring." (11% agreed to this statement). 4. "I think the Visual Patient technology might be helpful for non-monitor experts (e.g., surgeons) in the healthcare system." (53% of the participants agreed or strongly agreed).ConclusionThis mixed method study provides evidence that the included anesthesiologists considered the new avatar-based technology to be intuitive and easy to learn and that the technology enabled them to get an overview of the situation quickly. Only a few users considered the avatar presentation to be unhelpful for patient monitoring and about half think it might be useful for non-experts.

Project description:BACKGROUND:Visual Patient is an avatar-based alternative to standard patient monitor displays that significantly improves the perception of vital signs. Implementation of this technology in larger organizations would require it to be teachable by brief class instruction to large groups of professionals. Therefore, our study aimed to investigate the efficacy of such a large-scale introduction to Visual Patient. OBJECTIVE:In this study, we aimed to compare 2 different educational methods, one-on-one instruction and class instruction, for training anesthesia providers in avatar-based patient monitoring. METHODS:We presented 42 anesthesia providers with 30 minutes of class instruction on Visual Patient (class instruction group). We further selected a historical sample of 16 participants from a previous study who each received individual instruction (individual instruction group). After the instruction, the participants were shown monitors with either conventional displays or Visual Patient displays and were asked to interpret vital signs. In the class instruction group, the participants were shown scenarios for either 3 or 10 seconds, and the numbers of correct perceptions with each technology were compared. Then, the teaching efficacy of the class instruction was compared with that of the individual instruction in the historical sample by 2-way mixed analysis of variance and mixed regression. RESULTS:In the class instruction group, when participants were presented with the 3-second scenario, there was a statistically significant median increase in the number of perceived vital signs when the participants were shown the Visual Patient compared to when they were shown the conventional display (3 vital signs, P<.001; effect size -0.55). No significant difference was found for the 10-second scenarios. There was a statistically significant interaction between the teaching intervention and display technology in the number of perceived vital signs (P=.04; partial ?2=.076). The mixed logistic regression model for correct vital sign perception yielded an odds ratio (OR) of 1.88 (95% CI 1.41-2.52; P<.001) for individual instruction compared to class instruction as well as an OR of 3.03 (95% CI 2.50-3.70; P<.001) for the Visual Patient compared to conventional monitoring. CONCLUSIONS:Although individual instruction on Visual Patient is slightly more effective, class instruction is a viable teaching method; thus, large-scale introduction of health care providers to this novel technology is feasible.

Project description:BackgroundMaintaining adequate situation awareness is crucial for patient safety. Previous studies found that the use of avatar-based monitoring (Visual Patient Technology) improved the perception of vital signs compared to conventional monitoring showing numerical and waveform data; and was further associated with a reduction of perceived workload. In this study, we aimed to evaluate the effectiveness of Visual Patient Technology on perceptive performance and perceived workload when monitoring multiple patients at the same time, such as in central station monitors in intensive care units or operating rooms.MethodsA prospective, within-subject, computer-based laboratory study was performed in two tertiary care hospitals in Switzerland in 2018. Thirty-eight physician and nurse anesthetists volunteered for the study. The participants were shown four different central monitor scenarios in sequence, where each scenario displayed two critical and four healthy patients simultaneously for 10 or 30 s. After each scenario, participants had to recall the vital signs of the critical patients. Perceived workload was assessed with the National Aeronautics and Space Administration Task-Load-Index (NASA TLX) questionnaire.ResultsIn the 10-s scenarios, the median number of remembered vital signs significantly improved from 7 to 11 using avatar-based versus conventional monitoring with a mean of differences of 4 vital signs, 95% confidence interval (CI) 2 to 6, p < 0.001. At the same time, the median NASA TLX scores were significantly lower for avatar-based monitoring (67 vs. 77) with a mean of differences of 6 points, 95% CI 0.5 to 11, p = 0.034. In the 30-s scenarios, vital sign perception and workload did not differ significantly.ConclusionsIn central monitor multiple patient monitoring, we found a significant improvement of vital sign perception and reduction of perceived workload using Visual Patient Technology, compared to conventional monitoring. The technology enabled improved assessment of patient status and may, thereby, help to increase situation awareness and enhance patient safety.

Project description:BackgroundCognitive ergonomics design of patient monitoring may reduce human factor errors in high-stress environments. Eye-tracking is a suitable tool to gain insight into the distribution of visual attention of healthcare professionals with patient monitors, which may facilitate their further development.MethodsThis prospective, exploratory, high-fidelity simulation study compared anesthesia personnel's visual attention (fixation count and dwell-time) to 15 areas of interest on the patient monitor during non-critical and critical anesthesia situations. Furthermore, we examined the extent to which participants' experience influenced visual attention and which vital signs displayed on the patient monitor received the most visual attention. We used mixed zero-inflated Poisson regression and mixed linear models to analyze the data.ResultsAnalyzing 23 ten-minute scenarios, we found significantly more fixations to the areas of interest on the patient monitor during critical than non-critical situations (rate ratio of 1.45; 95% CI 1.33 to 1.59; p < 0.001). However, the dwell-time on the areas of interest did not significantly differ between the non-critical and critical situations (coefficient of - 1.667; 95% CI - 4.549 to 1.229; p = 0.27). The professional experience did not significantly influence the visual attention (fixation: rate ratio of 0.88; 95% CI 0.54 to 1.43; p = 0.61 and dwell-time: coefficient of 0.889; 95% CI - 1.465 to 3.229; p = 0.27). Over all situations, anesthesia personnel paid the most attention to the vital signs blood pressure (fixation: mean [SD] of 108 [74.83]; dwell-time: mean [SD] of 27 [15.90] seconds), end-expiratory carbon dioxide (fixation: mean [SD] of 59 [47.39]; dwell-time: mean [SD] of 30 [21.51] seconds), and the electrocardiogram (fixation: mean [SD] of 58 [64.70]; dwell-time: mean [SD] of 15 [14.95] seconds).ConclusionsCritical anesthesia situations increased anesthesia personnel's visual interaction with the patient monitor. Furthermore, we found that their visual attention focused mainly on a few vital signs. To assist clinicians in critical situations, manufacturers should optimize monitors to convey necessary information as easily and quickly as possible and optimize the visibility of less frequently observed but equally critical vital signs, especially when they are in an abnormal range.

Project description:Monitoring animals in their natural habitat is essential for advancement of animal behavioural studies, especially in pollination studies. Non-invasive techniques are preferred for these purposes as they reduce opportunities for research apparatus to interfere with behaviour. One potentially valuable approach is image-based tracking. However, the complexity of tracking unmarked wild animals using video is challenging in uncontrolled outdoor environments. Out-of-the-box algorithms currently present several problems in this context that can compromise accuracy, especially in cases of occlusion in a 3D environment. To address the issue, we present a novel hybrid detection and tracking algorithm to monitor unmarked insects outdoors. Our software can detect an insect, identify when a tracked insect becomes occluded from view and when it re-emerges, determine when an insect exits the camera field of view, and our software assembles a series of insect locations into a coherent trajectory. The insect detecting component of the software uses background subtraction and deep learning-based detection together to accurately and efficiently locate the insect among a cluster of wildflowers. We applied our method to track honeybees foraging outdoors using a new dataset that includes complex background detail, wind-blown foliage, and insects moving into and out of occlusion beneath leaves and among three-dimensional plant structures. We evaluated our software against human observations and previous techniques. It tracked honeybees at a rate of 86.6% on our dataset, 43% higher than the computationally more expensive, standalone deep learning model YOLOv2. We illustrate the value of our approach to quantify fine-scale foraging of honeybees. The ability to track unmarked insect pollinators in this way will help researchers better understand pollination ecology. The increased efficiency of our hybrid approach paves the way for the application of deep learning-based techniques to animal tracking in real-time using low-powered devices suitable for continuous monitoring.

Dataset Information

Avatar-Based Patient Monitoring With Peripheral Vision: A Multicenter Comparative Eye-Tracking Study.

Background

Objective

Methods

Results

Conclusions

Publications

Avatar-Based Patient Monitoring With Peripheral Vision: A Multicenter Comparative Eye-Tracking Study.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets