Interobserver Agreement and Satisfaction With the use of Telemedicine for Evaluating low Back Pain: A Primary, Observational, Cross-Sectional, Analytical Study.

STUDY DESIGN: A primary, observational, cross-sectional, analytical study. OBJECTIVE: The development of a framework for systematic telemedicine (TM) for orthopedic physicians in frequent clinical care may increase agreement in diagnosis and satisfaction among users of TM. Therefore, this study aimed to estimate the agreement in the diagnosis of low back pain (LBP) between TM, systematized by a self-completed digital questionnaire, and face-to-face (FF) care in patients with LBP. METHODS: This study included adults up to 75 years of age with LBP for more than 6 weeks. They were evaluated at 2 independent time points (TM and FF) by different orthopedists with 3 different levels of expertise. Professionals evaluated the sample without prior knowledge of the diagnosis, and each orthopedist provided a diagnosis. Diagnostic agreement was the primary outcome. Secondary outcomes were the duration of the visit and satisfaction among healthcare professionals. RESULTS: A total of 168 participants were eligible, of whom 126 sought care through TM and 122 sought FF care (mean age, 47 years [range, 18-75 years]; 66.4% women). The agreement among professionals regarding the diagnosis was moderate (kappa = .585, P = .001). TM was faster than FF (11.9 minutes (standard deviation = 4.1) vs 18.6 (SD = 6.9), P < .001). Professional satisfaction was higher among spine specialists than among orthopedic residents and orthopedists who were not specialists in spine surgery. CONCLUSION: Agreement in diagnosis was moderate for TM, with a 30% shorter visit duration than FF. Satisfaction varied by professional expertise and was higher among spine specialists than among professionals with other expertise.

Variable step-size evolving participatory learning with kernel recursive least squares applied to gas prices forecasting in Brazil.

A prediction model is an indispensable tool in business, helping to make decisions, whether in the short, medium, or long term. In this context, the implementation of machine learning techniques in time series forecasting models has a notorious relevance, as information processing and efficient and dynamic knowledge uncovering are increasingly demanded. This paper develops a model called Variable step-size evolving Participatory Learning with Kernel Recursive Least Squares, VS-ePL-KRLS, applied to the forecast of weekly prices for S500 and S10 diesel oil, at the Brazilian level, for biweekly and monthly horizons. The presented model demonstrates a better accuracy compared with analogous models in the literature, without loss of computational performance for all time series analyzed.