Development and Usability of a Novel Interactive Tablet App (PediAppRREST) to Support the Management of Pediatric Cardiac Arrest: Pilot High-Fidelity Simulation-Based Study.

BACKGROUND: Pediatric cardiac arrest (PCA), although rare, is associated with high mortality. Deviations from international management guidelines are frequent and associated with poorer outcomes. Different strategies/devices have been developed to improve the management of cardiac arrest, including cognitive aids. However, there is very limited experience on the usefulness of interactive cognitive aids in the format of an app in PCA. No app has so far been tested for its usability and effectiveness in guiding the management of PCA. OBJECTIVE: To develop a new audiovisual interactive app for tablets, named PediAppRREST, to support the management of PCA and to test its usability in a high-fidelity simulation-based setting. METHODS: A research team at the University of Padova (Italy) and human-machine interface designers, as well as app developers, from an Italian company (RE:Lab S.r.l.) developed the app between March and October 2019, by applying an iterative design approach (ie, design-prototyping-evaluation iterative loops). In October-November 2019, a single-center nonrandomized controlled simulation-based pilot study was conducted including 48 pediatric residents divided into teams of 3. The same nonshockable PCA scenario was managed by 11 teams with and 5 without the app. The app user's experience and interaction patterns were documented through video recording of scenarios, debriefing sessions, and questionnaires. App usability was evaluated with the User Experience Questionnaire (UEQ) (scores range from -3 to +3 for each scale) and open-ended questions, whereas participants' workload was measured using the NASA Raw-Task Load Index (NASA RTLX). RESULTS: Users' difficulties in interacting with the app during the simulations were identified using a structured framework. The app usability, in terms of mean UEQ scores, was as follows: attractiveness 1.71 (SD 1.43), perspicuity 1.75 (SD 0.88), efficiency 1.93 (SD 0.93), dependability 1.57 (SD 1.10), stimulation 1.60 (SD 1.33), and novelty 2.21 (SD 0.74). Team leaders' perceived workload was comparable (P=.57) between the 2 groups; median NASA RTLX score was 67.5 (interquartile range [IQR] 65.0-81.7) for the control group and 66.7 (IQR 54.2-76.7) for the intervention group. A preliminary evaluation of the effectiveness of the app in reducing deviations from guidelines showed that median time to epinephrine administration was significantly longer in the group that used the app compared with the control group (254 seconds versus 165 seconds; P=.015). CONCLUSIONS: The PediAppRREST app received a good usability evaluation and did not appear to increase team leaders' workload. Based on the feedback collected from the participants and the preliminary results of the evaluation of its effects on the management of the simulated scenario, the app has been further refined. The effectiveness of the new version of the app in reducing deviations from guidelines recommendations in the management of PCA and its impact on time to critical actions will be evaluated in an upcoming multicenter simulation-based randomized controlled trial.

Measuring the effects of visual demand on lateral deviation: a comparison among driver's performance indicators.

In this study we compare the efficacy of three driver's performance indicators based on lateral deviation in detecting significant on-road performance degradations while interacting with a secondary task: the High Frequency Component of steering wheel (HFC), and two indicators described in ISO/DIS 26022 (2007): the Normative and the Adapted Lane Change Test (LCT). Sixteen participants were asked to perform a simulated lane-change task while interacting, when required, with a visual search task with two levels of difficulty. According to predictions, results showed that the Adapted LCT indicator, taking into consideration individual practices in performing the LCT, succeeded in discriminating between single and dual task conditions. Furthermore, this indicator was also able to detect whether the driver was interacting with an easy or a difficult secondary task. Despite predictions, results did not confirm Normative LCT and HFC to be reliable indicators of performance degradation within the simulated LCT.

Field tests and machine learning approaches for refining algorithms and correlations of driver's model parameters.

This paper describes the field tests on a driving simulator carried out to validate the algorithms and the correlations of dynamic parameters, specifically driving task demand and drivers' distraction, able to predict drivers' intentions. These parameters belong to the driver's model developed by AIDE (Adaptive Integrated Driver-vehicle InterfacE) European Integrated Project. Drivers' behavioural data have been collected from the simulator tests to model and validate these parameters using machine learning techniques, specifically the adaptive neuro fuzzy inference systems (ANFIS) and the artificial neural network (ANN). Two models of task demand and distraction have been developed, one for each adopted technique. The paper provides an overview of the driver's model, the description of the task demand and distraction modelling and the tests conducted for the validation of these parameters. A test comparing predicted and expected outcomes of the modelled parameters for each machine learning technique has been carried out: for distraction, in particular, promising results (low prediction errors) have been obtained by adopting an artificial neural network.