RESUMEN
The aim of this study is to evaluate the item-level psychometrics of the Ascertain Dementia Eight-Item Informant Questionnaire (AD-8) by examining its dimensionality, rating scale integrity, item fit statistics, item difficulty hierarchy, item-person match, and precision. We used confirmatory factor analysis and the Rasch rating scale model for analyzing the data extracted from the proxy versions of the 2019 and 2020 National Health and Aging Trends Study, USA. A total of 403 participants were included in the analysis. The confirmatory factor analysis with a 1-factor model using the robust weighted least squares (WLSMV) estimator indicated a unidimensional measurement structure (χ2 = 41.015, df = 20, p = 0.004; root mean square error of approximation = 0.051; comparative fit index = 0.995; Tucker-Lewis Index = 0.993;). The findings indicated that the AD-8 has no misfitting items and no differential item functioning across sex and gender. The items were evenly distributed in the item difficulty rating (range: -2.30 to 0.98 logits). While there were floor effects, the AD-8 revealed good reliability (Rasch person reliability = 0.67, Cronbach's alpha = 0.89). The Rasch analysis reveals that the AD-8 has excellent psychometric properties that can be used as a screening assessment tool in clinical settings allowing clinicians to measure dementia both quickly and efficiently. To summarize, the AD-8 could be a useful primary screening tool to be used with additional diagnostic testing, if the patient is accompanied by a reliable informant.
Asunto(s)
Demencia , Demencia/diagnóstico , Análisis Factorial , Femenino , Humanos , Masculino , Psicometría , Reproducibilidad de los Resultados , Encuestas y CuestionariosRESUMEN
Railway point devices act as actuators that provide different routes to trains by driving switchblades from the current position to the opposite one. Point failure can significantly affect railway operations, with potentially disastrous consequences. Therefore, early detection of anomalies is critical for monitoring and managing the condition of rail infrastructure. We present a data mining solution that utilizes audio data to efficiently detect and diagnose faults in railway condition monitoring systems. The system enables extracting mel-frequency cepstrum coefficients (MFCCs) from audio data with reduced feature dimensions using attribute subset selection, and employs support vector machines (SVMs) for early detection and classification of anomalies. Experimental results show that the system enables cost-effective detection and diagnosis of faults using a cheap microphone, with accuracy exceeding 94.1% whether used alone or in combination with other known methods.