Automatic Evaluation of Trainee Nurses' Patient Transfer Skills Using Multiple Kinect Sensors

Abstract

SUMMARYTo help student nurses learn to transfer patients from a bed to a wheelchair, this paper proposes a system for automatic skill evaluation in nurses' training for this task. Multiple Kinect sensors were employed, in conjunction with colored markers attached to the trainee's and patient's clothing and to the wheelchair, in order to measure both participants' postures as they interacted closely during the transfer and to assess the correctness of the trainee's movements and use of equipment. The measurement method involved identifying body joints, and features of the wheelchair, via the colors of the attached markers and calculating their 3D positions by combining color and depth data from two sensors. We first developed an automatic segmentation method to convert a continuous recording of the patient transfer process into discrete steps, by extracting from the raw sensor data the defining features of the movements of both participants during each stage of the transfer. Next, a checklist of 20 evaluation items was defined in order to evaluate the trainee nurses' skills in performing the patient transfer. The items were divided into two types, and two corresponding methods were proposed for classifying trainee performance as correct or incorrect. One method was based on whether the participants' relevant body parts were positioned in a predefined spatial range that was considered 'correct' in terms of safety and efficacy (e.g., feet placed appropriately for balance). The second method was based on quantitative indexes and thresholds for parameters describing the participants' postures and movements, as determined by a Bayesian minimum-error method. A prototype system was constructed and experiments were performed to assess the proposed approach. The evaluation of nurses' patient transfer skills was performed successfully and automatically. The automatic evaluation results were compared with evaluation by human teachers and achieved an accuracy exceeding 80%.

Self-Help Training System for Nursing Students to Learn Patient Transfer Skills

“…To indicate the accuracy of performance of the trainee against each evaluation item, we used our previously developed automatic skill evaluation method [41], [42]. As described in Section 2.1, the evaluation factors and corresponding motions for each item were different.…”

“…In this step, one image sequence was extracted for each evaluation item. The accuracy of segmentation was examined in our previous work [41] and was found to be 96.4 percent.…”

“…The calculation of posture data was based on the spatial positions of the body joints, indicated by X, Y, and Z. Additionally, data on the wheelchair's position and posture (items 1 and 2) were measured using the preset color threshold to determine the minimum bounding rectangle of its cushion [41]. The states of brakes (item 3) were detected using the attached red markers, which became visible only when the brakes were applied [41]. The resolutions for distance and angle measurements achieved using the above methods were 1 cm and 1 degree, respectively.…”

“…2 shows the hardware of the self-help skill training system prototype for patient transfer, which comprises a dual-Kinect sensor system, liquid crystal display, wireless mouse, and wireless keyboard. The dual-Kinect sensor system was designed to capture information on the posture of trainees and patients, and state of the wheelchair [41], [42]. The sensor system contains two Microsoft Kinect sensors, two personal computers (PCs), and a router.…”

“…The self-help skill training system consists of three parts: (1) a sensor system comprising Kinect sensors and color markers (which has been proposed in our previous studies) to measure and evaluate the performance of trainees [41], [42]; (2) a feedback method to help trainees check and correct their errors; and (3) an operation interface to enable trainees to operate the system by themselves.…”