RESUMEN
This paper proposes new improved binary versions of the Sine Cosine Algorithm (SCA) for the Feature Selection (FS) problem. FS is an essential machine learning and data mining task of choosing a subset of highly discriminating features from noisy, irrelevant, high-dimensional, and redundant features to best represent a dataset. SCA is a recent metaheuristic algorithm established to emulate a model based on sine and cosine trigonometric functions. It was initially proposed to tackle problems in the continuous domain. The SCA has been modified to Binary SCA (BSCA) to deal with the binary domain of the FS problem. To improve the performance of BSCA, three accumulative improved variations are proposed (i.e., IBSCA1, IBSCA2, and IBSCA3) where the last version has the best performance. IBSCA1 employs Opposition Based Learning (OBL) to help ensure a diverse population of candidate solutions. IBSCA2 improves IBSCA1 by adding Variable Neighborhood Search (VNS) and Laplace distribution to support several mutation methods. IBSCA3 improves IBSCA2 by optimizing the best candidate solution using Refraction Learning (RL), a novel OBL approach based on light refraction. For performance evaluation, 19 real-wold datasets, including a COVID-19 dataset, were selected with different numbers of features, classes, and instances. Three performance measurements have been used to test the IBSCA versions: classification accuracy, number of features, and fitness values. Furthermore, the performance of the last variation of IBSCA3 is compared against 28 existing popular algorithms. Interestingly, IBCSA3 outperformed almost all comparative methods in terms of classification accuracy and fitness values. At the same time, it was ranked 15 out of 19 in terms of number of features. The overall simulation and statistical results indicate that IBSCA3 performs better than the other algorithms.
RESUMEN
BACKGROUND: Published training recommendations exist for youth athletes aimed at reducing injury risk. No studies have assessed the impact of counselling interventions using training recommendations on risk of injury in young athletes. OBJECTIVES: To determine if online training counselling regarding safe sport participation can reduce injury risk in youth athletes and to assess recommendation compliance, including barriers to compliance. METHODS: A multicentre randomised intervention trial was performed at two Midwestern academic institutions with expertise in treating young athletes. Enrolled subjects ages 8-17 completed a baseline risk assessment survey and were randomised to a control or intervention group. Both groups completed exposure surveys every 3 months for 1 year. The intervention group also received online training counselling on safe sport participation every 3 months. Training characteristics including training volume, degree of specialisation and adherence to recommendations were captured. Differences in self-reported injury between groups, compliance to recommendations and barriers to compliance were evaluated. RESULTS: At baseline, n=357 subjects were enrolled (n=172 control and n=185 intervention). Controls were nearly twice as likely to be injured during the intervention period after controlling for age, sex, baseline injury and level of specialisation. No improvement in recommendation compliance was detected among intervention subjects. Primary barriers to compliance were no prior knowledge of recommendations, personal choice and following coaches' recommendations. CONCLUSIONS: In this convenience sample of youth athletes, electronic training counselling surrounding safe sports participation was not determined to affect injury risk. Lack of knowledge and adherence to appropriate training recommendations is evident and barriers to compliance exist.
Asunto(s)
Traumatismos en Atletas/prevención & control , Consejo , Medición de Riesgo , Adolescente , Atletas , Niño , Femenino , Humanos , Internet , Masculino , Cooperación del Paciente/estadística & datos numéricos , Estudios Prospectivos , DeportesRESUMEN
BACKGROUND: Multiple sclerosis (MS) is the most common demyelinating disease and characterized by immunological changes. Oligoclonal bands of IgG in CSF not seen in corresponding serum have been used for many years as part of the diagnostic criteria. However, considerably less is known about the role of IgM, despite several studies showing marked changes to IgM metabolism in MS. Bands of oligoclonal IgM (o-IgM) are more difficult to determine than oligoclonal IgG, thus limiting their study, and there is no agreement as to whether o-IgM in CSF should be part of the clinical work-up of MS. Nevertheless, there is a possibility that such bands might provide a prognostic marker if a cut-off could be established. MATERIALS AND METHODS: In this pilot study, paired samples of CSF and serum from 37 patients with relapsing-remitting MS (RRMS) and 57 controls with no subsequent signs of neurological disease were analysed for total IgM, and bands of o-IgM were visualised by isoelectric focusing and western blot. Patient records were used to compare mean changes in Expanded Disability Status Scale (EDSS) over a maximum of 17 years. RESULTS: None of the controls displayed extra o-IgM in CSF compared to corresponding serum, whereas additional o-IgM band(s) were seen in CSF in most patient samples (70%). After five years of disease, there was a significant difference in the EDSS between patients with no extra o-IgM compared to patients with at least one extra o-IgM band. This difference increased over time. If a cut-off of two or more extra bands of o-IgM in CSF was applied, this difference was not found. CONCLUSION: These exploratory data suggest that o-IgM support the prognostic potential for RRMS, and though tentative, the occurrence of any bands of o-IgM restricted to CSF seems to result in poorer prognosis. Despite the small size of the groups, the data infer that the absence of CSF-restricted o-IgM is good news for the patient. The results need to be reproduced in a more comprehensive study.