A novel approach: Simulating multiple simultaneous encounters to assess multitasking ability in emergency medicine.

STUDY OBJECTIVE: The purpose of this feasibility study is to develop and validate a new assessment tool and scoring system for multitasking competency for physicians in-training in a timed simulated setting. The multitasking competency includes ability to appropriately prioritize and implement tasks for different patients who present simultaneously. METHODS: We designed three single task stations with different levels of difficulty and priority. These skill stations were then combined to create a multitasking simulation scenario. Skill checklists and the global rating scale were utilized to assess the participants' performance. A multitasking score, multitasking index, and priority score were developed to measure the multitasking ability of participants. RESULTS: Thirty-three first-year postgraduate physicians were recruited for this prospective study. The total performance scores were significantly higher for the single-tasking stations than for the multitasking scenario. In terms of the time needed to complete the tasks, the participants spent more time on the multitasking scenario than on the single-tasking scenario. There were significant correlations between the global rating scale and the multitasking score (rho = 0.693, p < 0.001) and between the global rating scale and the multitasking index (rho = 0.515, p < 0.001). The multitasking score, multitasking index, and priority score did not have any significant correlations with the total single-tasking score. CONCLUSION: We demonstrated that the use of a simulated multitasking scenario could be an effective method of assessing multitasking ability and allow assessors to offer better quality feedback.

General Purpose Low-Level Reinforcement Learning Control for Multi-Axis Rotor Aerial Vehicles.

This paper proposes a multipurpose reinforcement learning based low-level multirotor unmanned aerial vehicles control structure constructed using neural networks with model-free training. Other low-level reinforcement learning controllers developed in studies have only been applicable to a model-specific and physical-parameter-specific multirotor, and time-consuming training is required when switching to a different vehicle. We use a 6-degree-of-freedom dynamic model combining acceleration-based control from the policy neural network to overcome these problems. The UAV automatically learns the maneuver by an end-to-end neural network from fusion states to acceleration command. The state estimation is performed using the data from on-board sensors and motion capture. The motion capture system provides spatial position information and a multisensory fusion framework fuses the measurement from the onboard inertia measurement units for compensating the time delay and low update frequency of the capture system. Without requiring expert demonstration, the trained control policy implemented using an improved algorithm can be applied to various multirotors with the output directly mapped to actuators. The algorithm's ability to control multirotors in the hovering and the tracking task is evaluated. Through simulation and actual experiments, we demonstrate the flight control with a quadrotor and hexrotor by using the trained policy. With the same policy, we verify that we can stabilize the quadrotor and hexrotor in the air under random initial states.

Impact of virtual reality anatomy training on ultrasound competency development: A randomized controlled trial.

PURPOSE: The use of Virtual Reality (VR) in health professions education has increased dramatically in recent years, yet there is limited evidence of its impact on educational outcomes. The purpose of the study was to assess the impact of VR anatomy instruction on the ultrasound competency of novice learners participating in a ultrasonography workshop. METHOD: We designed a VR-enhanced ultrasonography training program and utilized a plane transection tool to interact with a three-dimensional (3D) VR model of the human body which facilitated the 3D conceptualization of the spatial relationship of anatomical structures, leading to faster and better development of ultrasonographic competency. This was a randomized control study which enrolled third-year medical students (n = 101) without previous exposure to formal or informal ultrasonography training. The participants were randomly divided into an intervention and control group. We assessed participants' competency through ultrasound performance stations on live subjects, we also measured anatomical and ultrasound image identification ability using multiple choice tests. RESULT: Participants in the intervention group (median = 16; interquartile 13 to 19) had significantly higher scores in ultrasonography task performance tests than the control group (median = 10; interquartile 7 to 14; Mann-Whitney U = 595; P < 0.01). In sub-group analysis, the intervention group performed significantly better in the six out of ten ultrasound tasks. Participants in the intervention group also had greater improvement in ultrasonographic image identification MCQ tests than the control group (Mann-Whitney U = 914; P < 0.05). CONCLUSION: This study suggests that VR-enhanced anatomical training could be of significant benefit in ultrasonography training by promoting a better understanding of the spatial relationships of anatomical structures and the development of early psychomotor skills transferable to the handling of ultrasonographic probes.