"It's got to be more than that". Parents and speech-language pathologists discuss training content for families with a new speech generating device.

PURPOSE: Although parent-implemented interventions for children with a speech-generating device (SGD) have been well researched, little is known about parents' or speech-language pathologists' (SLPs) views around parent training content. In this project, we aimed to identify areas that parents and SLPs consider should be included in training for families with a new SGD.

METHODS: Seven parents of children with an SGD and three SLPs who were new to the SGD field, participated in individual semi-structured interviews. Ten SLPs experienced in SGD practice took part in two focus groups. Data were analysed using grounded theory methods.

RESULTS: Participants identified the following areas suitable for inclusion in a family SGD training package: (a) content aimed at improving acceptance and uptake of the SGD, including technical guidance, customisation and reassurance around SGD misconceptions; (b) content around aided language development and (c) home practice strategies, including responsivity, aided language stimulation and managing children's motivation.

CONCLUSIONS: Participants identified diverse training targets, many of which are unexplored in parent-training research to date. Their recounted experiences illustrate the diversity of family capacity, knowledge and training priorities, and highlight the need for collaborative planning between families and SLPs at all stages of SGD training. Implications for Rehabilitation Training needs for families with a new speech generating device (SGD) are diverse, ranging from technology-specific competencies to broader areas, such as advocacy, teamwork and goal-setting skills. Each family with a new SGD will have a unique profile of training needs, determined by individual learning capacity, priorities, prior knowledge and experience, as well as their child's current communication skills and future support needs. Parents and speech-language pathologists (SLPs) may hold different priorities concerning family SGD training, necessitating ongoing team discussion.

Acoustic and device feature fusion for load recognition

Appliance-specific Load Monitoring (LM) provides a possible solution to the problem of energy conservation which is becoming increasingly challenging, due to growing energy demands within offices and residential spaces. It is essential to perform automatic appliance recognition and monitoring for optimal resource utilization. In this paper, we study the use of non-intrusive LM methods that rely on steady-state appliance signatures for classifying most commonly used office appliances, while demonstrating their limitation in terms of accurately discerning the low-power devices due to overlapping load signatures. We propose a multi-layer decision architecture that makes use of audio features derived from device sounds and fuse it with load signatures acquired from energy meter. For the recognition of device sounds, we perform feature set selection by evaluating the combination of time-domain and FFT-based audio features on the state of the art machine learning algorithms. Further, we demonstrate that our proposed feature set which is a concatenation of device audio feature and load signature significantly improves the device recognition accuracy in comparison to the use of steady-state load signatures only.

A trial of e-simulation of sudden patient deterioration (FIRST2ACT WEB) on student learning

BACKGROUND: High-fidelity simulation pedagogy is of increasing importance in health professional education; however, face-to-face simulation programs are resource intensive and impractical to implement across large numbers of students. OBJECTIVES: To investigate undergraduate nursing students' theoretical and applied learning in response to the e-simulation program-FIRST2ACT WEBTM, and explore predictors of virtual clinical performance. DESIGN AND SETTING: Multi-center trial of FIRST2ACT WEBTM accessible to students in five Australian universities and colleges, across 8 campuses. PARTICIPANTS: A population of 489 final-year nursing students in programs of study leading to license to practice. METHODS: Participants proceeded through three phases: (i) pre-simulation-briefing and assessment of clinical knowledge and experience; (ii) e-simulation-three interactive e-simulation clinical scenarios which included video recordings of patients with deteriorating conditions, interactive clinical tasks, pop up responses to tasks, and timed performance; and (iii) post-simulation feedback and evaluation. Descriptive statistics were followed by bivariate analysis to detect any associations, which were further tested using standard regression analysis. RESULTS: Of 409 students who commenced the program (83% response rate), 367 undergraduate nursing students completed the web-based program in its entirety, yielding a completion rate of 89.7%; 38.1% of students achieved passing clinical performance across three scenarios, and the proportion achieving passing clinical knowledge increased from 78.15% pre-simulation to 91.6% post-simulation. Knowledge was the main independent predictor of clinical performance in responding to a virtual deteriorating patient R(2)=0.090, F(7, 352)=4.962, p<0.001. DISCUSSION: The use of web-based technology allows simulation activities to be accessible to a large number of participants and completion rates indicate that 'Net Generation' nursing students were highly engaged with this mode of learning. CONCLUSION: The web-based e-simulation program FIRST2ACTTM effectively enhanced knowledge, virtual clinical performance, and self-assessed knowledge, skills, confidence, and competence in final-year nursing students.

Learning multi-faceted activities from heterogeneous data with the product space hierarchical Dirichlet processes

Hierarchical Dirichlet processes (HDP) was originally designed and experimented for a single data channel. In this paper we enhanced its ability to model heterogeneous data using a richer structure for the base measure being a product-space. The enhanced model, called Product Space HDP (PS-HDP), can (1) simultaneously model heterogeneous data from multiple sources in a Bayesian nonparametric framework and (2) discover multilevel latent structures from data to result in different types of topics/latent structures that can be explained jointly. We experimented with the MDC dataset, a large and real-world data collected from mobile phones. Our goal was to discover identity–location– time (a.k.a who-where-when) patterns at different levels (globally for all groups and locally for each group). We provided analysis on the activities and patterns learned from our model, visualized, compared and contrasted with the ground-truth to demonstrate the merit of the proposed framework. We further quantitatively evaluated and reported its performance using standard metrics including F1-score, NMI, RI, and purity. We also compared the performance of the PS-HDP model with those of popular existing clustering methods (including K-Means, NNMF, GMM, DP-Means, and AP). Lastly, we demonstrate the ability of the model in learning activities with missing data, a common problem encountered in pervasive and ubiquitous computing applications.