Machine Learning for Motor Imagery Wrist Dorsiflexion Prediction in Brain-Computer Interface Assisted Stroke Rehabilitation.

Stroke is a life-changing event that can affect the survivors' physical, cognitive and emotional state. Stroke care focuses on helping the survivors to regain their strength; recover as much functionality as possible and return to independent living through rehabilitation therapies. Automated training protocols have been reported to improve the efficiency of the rehabilitation process. These protocols also decrease the dependency of the process on a professional trainer. Brain-Computer Interface (BCI) based systems are examples of such systems where they make use of the motor imagery (MI) based electroencephalogram (EEG) signals to drive the rehabilitation protocols. In this paper, we have proposed the use of well-known machine learning (ML) algorithms, such as, the decision tree (DT), Naive Bayesian (NB), linear discriminant analysis (LDA), support vector machine (SVM), ensemble learning classifier (ELC), and artificial neural network (ANN) for MI wrist dorsiflexion prediction in a BCI assisted stroke rehabilitation study conducted on eleven stroke survivors with either the left or right paresis. The doubling sub-band selection filter bank common spatial pattern (DSBS-FBCSP) has been proposed as feature extractor and it is observed that the ANN based classifier produces the best results.

Population control charts for population data.

Healthcare managers are beginning to collect full population data, rather than sample data, on some patient and performance measures. For example, hospitals and healthcare systems already gather and store comprehensive data on admissions, ambulatory encounters, and other procedures. And as the electronic medical record is more widely used, complete population data will be collected on an even wider range of clinical measures, such as blood pressure and Laboratory values, in both inpatient and outpatient settings. To correctly monitor process quality when working with full population data, rather than sample data, healthcare managers will need appropriate statistical tools. Traditional control charts, which are used for tracking processes over time, are not suitable for such population data because they are based on the assumption that sample data are being collected. The author proposes a new type of control chart specifically for use with such population data: population control charts. These control charts can be used for monitoring processes that have output measures with continuous, binomial, or nonbinomial rate variables.

CAN'T MISS: conquer any number task by making important statistics simple. Part 7. Statistical process control: x-s control charts.

Statistical process control (SPC) can be thought of as the frequent monitoring of processes using inferential statistics. The feature that distinguishes SPC from the typical use of inferential statistics for analyzing populations is that in the former frequent samples are taken over time, whereas in inferential statistics a single sample is generaLLy taken before and after some intervention or treatment. An x-s control chart is used to monitor a continuous variable that reflects the output of a process. The x-s control chart is a graph that includes serial sample means (x) as the variables of interest, a centerline that represents the grand mean of the samples (x), and upper control limit (UCL) and lower control limit (LCL) that represent three standard errors (SEx) above and below the centerline. An x-s control chart is used to estimate with 99.7% confidence that the population mean of a continuous output variable was within the interval defined by the UCL and LCL during a period of baseline monitoring. It is further assumed that if the process remains stable, future population means wiLL remain between the control Limits for additional process outputs. Control charts allow the evaluation of both common- and special-cause variation. AnaLysis of the common-cause variation aLLows an assessment of the current process performance. Special-cause variation is identified when there is a sample mean that is beyond the UCL or LCL.

CAN'T MISS: conquer any number task by making important statistics simple. Part 8. Statistical process control: n, np, c, u control charts.

A p control chart is used to monitor a proportion from a binomial variable that reflects the output of a process. The p control chart is a graph that includes serial sample proportions (p) as the variables of interest, a centerline that represents the overall proportion of the samples (p), and upper control limits (UCLs) and Lower control Limits (LCLs) that represent three standard errors (SEp) above and below the centerline. We use a p control chart to estimate with 99.7% confidence that the population proportion of an output variable was within the interval defined by the UCLs and LCLs during a period of baseline monitoring. We further assume that if the process remains stable, the values of future population proportions will remain between the established control limits. An np control chart simply plots the numerators of the sample proportions as the variables of interest. A u control chart is used to monitor ratios. In u control charts, ratios are quantities in which the numerator can have one or more occurrences in reference to the denominator. A c control chart is analogous to an np control chart in that it graphs the numerators from the ratios.

Can't miss: conquer any number task by making important statistics simple. Part 4. Confidence intervals with t distributions, standard error, and confidence intervals for proportions.

Healthcare quality professionals need to understand and use inferential statistics to interpret sample data from their organizations. Since in quality improvement and healthcare research studies, all the data from a population often are not available, investigators take samples and make inferences about that population using inferential statistics. This series of six articles will give readers an understanding of the concepts of inferential statistics, as well as the specific tools for calculating confidence intervals and tests of statistical significance for samples of data. The statistical principles are equally applicable to quality improvement and healthcare research studies. This article, Part 4, starts with a review of the information contained in Parts 1, 2, and 3, which appeared in the July/August 2003 issue of the Journal for Healthcare Quality. This article describes t distributions and how these are used to calculate confidence intervals for estimating a population mean based on a sample mean of a continuous variable. Part 4 concludes with a discussion of standard error, margin of error, and confidence intervals for estimating a population proportion based on a sample proportion from a binomial variable.

Can't miss: conquer any number task by making important statistics simple. Part 5. Comparing two confidence intervals, standard error of the difference between two means and between two proportions.

Healthcare quality professionals need to understand and use inferential statistics to interpret sample data from their organizations. Since in quality improvement and healthcare research studies all the data from a population often are not available, investigators take samples and make inferences about that population using inferential statistics. This series of six articles will give readers an understanding of the concepts of inferential statistics as well as the specific tools for calculating confidence intervals and tests of statistical significance for samples of data. This article, Part 5, demonstrates the comparison of two confidence intervals as a method for estimating the difference between two population means. The concept of the standard error of the difference between two sample means is presented along with the confidence interval for estimating the difference between two population means. The article concludes with the standard error and confidence interval for estimating the difference between two population proportions from a binomial variable.

Can't miss: conquer any number task by making important statistics simple. Part 6. Tests of statistical significance (z test statistic, rejecting the null hypothesis, p value), t test, z test for proportions, statistical significance versus meaningful difference.

Healthcare quality professionals need to understand and use inferential statistics to interpret sample data from their organizations. Since in quality improvement and healthcare research studies all the data from a population often are not available, investigators take samples and make inferences about that population using inferential statistics. This series of six articles will give readers an understanding of the concepts of inferential statistics as well as the specific tools for calculating confidence intervals and tests of statistical significance for samples of data. This article, Part 6, merges the four concepts of the (1) standard error of the difference between sample means, (2) the z test statistic, (3) rejecting the null hypothesis, and (4) the p value to provide a comprehensive view of tests of statistical significance. This is followed by a description of t tests, statistical tests for comparing two sample proportions, and Type I and Type II errors. The series of articles concludes with a description of statistical significance versus meaningful difference.

CAN'T MISS--conquer any number task by making important statistics simple. Part 1. Types of variables, mean, median, variance, and standard deviation.

Healthcare quality improvement professionals need to understand and use inferential statistics to interpret sample data from their organizations. In quality improvement and healthcare research studies all the data from a population often are not available, so investigators take samples and make inferences about the population by using inferential statistics. This three-part series will give readers an understanding of the concepts of inferential statistics as well as the specific tools for calculating confidence intervals for samples of data. This article, Part 1, presents basic information about data including a classification system that describes the four major types of variables: continuous quantitative variable, discrete quantitative variable, ordinal categorical variable (including the binomial variable), and nominal categorical variable. A histogram is a graph that displays the frequency distribution for a continuous variable. The article also demonstrates how to calculate the mean, median, standard deviation, and variance for a continuous variable.

CAN'T MISS--conquer any number task by making important statistics simple. Part 2. Probability, populations, samples, and normal distributions.

Healthcare quality improvement professionals need to understand and use inferential statistics to interpret sample data from their organizations. In quality improvement and healthcare research studies all the data from a population often are not available, so investigators take samples and make inferences about the population by using inferential statistics. This three-part series will give readers an understanding of the concepts of inferential statistics as well as the specific tools for calculating confidence intervals for samples of data. This article, Part 2, describes probability, populations, and samples. The uses of descriptive and inferential statistics are outlined. The article also discusses the properties and probability of normal distributions, including the standard normal distribution.

CAN'T MISS--conquer any number task by making important statistics simple. Part 3. Standard error, estimation, and confidence intervals.

Healthcare quality improvement professionals need to understand and use inferential statistics to interpret sample data from their organizations. In quality improvement and healthcare research studies all the data from a population often are not available, so investigators take samples and make inferences about the population by using inferential statistics. This three-part series will give readers an understanding of the concepts of inferential statistics as well as the specific tools for calculating confidence intervals for samples of data. This article, Part 3, describes standard error and margin of error for a continuous variable and how they are calculated from the sample size and standard deviation of a sample. The article then demonstrates how the standard error and margin of error are used to calculate the confidence interval for estimating a population mean based on a sample mean.