activAnalyzer 2.1.2

• Removed an undesired comma from a req() function in app_server.R that caused an error with more recent versions of package dependencies (likely Shiny).
• Updated the figures for the comparisons with norms and recommendations (legend is now correctly placed at the top).

activAnalyzer 2.1.1

• When collapsing data from a shorter epoch to a longer epoch, NAs are introduced in the 'wearing' column provided by the PhysicalActivity package during nonwear time analysis. This generated a bug when computing the accumulation metrics because the detection of the bouts partly depends on the labels provided in this column. This has been corrected by replacing NAs by "Nonwear".
• In version 2.1.0, MX5 had the same value than MX15 due to a coding error. The result for MX5 has now been corrected.

activAnalyzer 2.1.0

• Moved 'assertthat' from Suggests to Imports field in DESCRIPTION.
• Added a German translation for most figures.
• Corrected an ERROR: replaced the mean by the minimum of the counts/(epoch duration) related to the most X active minutes for computing MX metrics.
• Updated guides for the MX definitions: eg, for M30: the count value (in counts/epoch duration) at and above which the most active 30 minutes are accumulated over the day.
• Added an internal function (compute_peak_step_acc()) to compute peak step accumulation only when there is the minimum number of minutes required to perform computation. It returns NA otherwise.
• Added an internal function (compute_mx()) to compute MX metrics only when there is the minimum amount of time required to perform the targetted computation. It returns NA otherwise.

activAnalyzer 2.0.2

• Corrected figure in the vignette (a step-based metrics image was used instead of a figure related IG metrics).
• Replaced ifelse() functions by dplyr::if_else() to increase speed.
• Updated read_agd(): ticks_to_dttm function within dplyr::across() functions has be rewritten so that it fits the requirements of the dplyr::across() function as of dplyr 1.1.0.
• Updated recap_by_day() function. The separation of the timestamp column in two columns (date and time) generated NAs for rows with 00:00:00. This has been solved by formating the timestamp column before separating it.
• Launching the app in the default web brower is now the default setting of the run_app() function to prevent the problems that appear when closing the app after a use in the RStudio window.

activAnalyzer 2.0.1

• Increased speed of the compute_accumulation_metrics() function.
• Added progress bar for waiting for initialization of graphics related to final results.
• Corrected a bug that appeared when the width of the intensity bin set by the user to compute intensity gradient is too large, leading to time spent in only one bin during a day and thus leading to the impossibility to compute the slope of the log-log model used to get intensity gradient.

activAnalyzer 2.0.0

• Total kcal and PAL are now correctly computed when the user does not analyse the whole day. There was no problem when the entire day was considered. However, when the user wanted to analyse a given period of the day, BMR information used in the calculations of kcal and PAL was still related to the whole day, not the period of the day to analyse. Total kcal and PAL were thus wrong when the period of the day to be analysed was not the entire day. This has been corrected (#28).
• Updated arguments passed to dplyr::left_join() function following the release of dplyr package v1.1.0.
• The app now provides intensity distribution metrics.
• The app now provides activity accumulation metrics for sedentary behaviour and physical activity: alpha coefficient, median bout duration, usual bout duration, Gini index.
• The addition of new metrics led to a breaking change in the package: the recap_by_day() function now returns a list, not a dataframe.
• The buttons and tables related to PROactive instruments panels now correctly disappear when the user runs an new analysis that leads to no valid days for the accelerometer.
• Some variable names in the source code are no longer hard coded to provide more flexibility when analysing personal datasets outside the app. Default variable names remain the same.
• The app now allows to export all results to .html reports.
• The package does not export the following function anymore: tbl_agd.
• Now the Y labels of the graphics showing counts and steps data include the epoch duration for better data vizualisation context.
• The function create_fig_res_by_day() now allows to visualise all metrics by day (activity volume, step accumulation, intensity distribution).
• Added functions to get a radar plot showing MX metrics by day of measurement and for the mean or median of the considered days.
• The app now shows all results (previously several results were shown only in the report).
• Updated the vignette.

Note about the user's experience: Because the app now provides more tables and graphics, getting the final results will take more time than the previous version (eg, up to 2'30 for a 14-day file using 60-s epochs with the current version).

activAnalyzer 1.1.0

• Added the intersex and prefer not to say categories to provide a more inclusive classification of sex. As it seems there is no scientific study about what should be the calculation of resting and activity energy expenditures for intersex people, the values provided for Basal metabolic rate (BMR) and METs are the averages of two values: the value that would be computed for a male, and the value that would be computed for a female. For people reporting prefer not to say, computations for females are used by default.
• Updated in the guide the description of the computation of BMR: "If the patient considers their sex as undefined or chooses the prefer not to say option, then an equation for females is used. If the patient falls into the intersex category, then the average of the results for a male and for a female of the considered age is used (WARNING: At the time of writing this guide, there is no scientific data to justify any calculation for intersex people).".
• Updated in the guide the description of the computation of METs: "METs, by using the MET equation provided by the user (if the patient considers their sex as undefined or chooses the prefer not to say option, then equations including sex information, when selected, are used as if the patient were a female; when the intersex category is used, an average of the METs related respectively to a male and to a female is used with the equations using sex information; of note, at the time of writing this guide, there is no scientific data to justify any calculation for intersex people);".
• The compute_bmr() and compute_mets() functions now use the dplyr::case_when() function to determine the appropriate value of BMR and METs, respectively.
• Corrected typos ("MPVA" -->"MVPA") in the English version of the guide.
• Added a class "icon-widget" to the sidebar icons to keep control on the appropriate size of the icons.
• Now the prepare_dataset() function imports data using the read_agd() function (instead of the PhysicalActivity::readActigraph() function). This modification now allows to import data from the GT3X device (previously only data from GT3X+ and newer devices could be used). This was not possible before because the structure of the .agd file obtained with a GT3X device is not accepted by the PhysicalActivity::readActigraph() function.
• There is no more constraints about the necessity to have inclinometer information in the .agd file. This information is not used anymore in the app.
• Added a file info message (measurement information) when loading the data file.
• Changed the size arguments of the internal geom_line(), geom_segment() and geom_rect() functions by linewidth arguments in relation to the v3.4.0 {ggplot2} update.
• Added the possibility to zoom in on the figures of the app.
• Added the possibility to provide information about potential relevant physical activity periods that would have modified physical activity level but that could not be recorded due to accelerometer removal.
• Added an alert message for the figure with steps (cf. report) when the score is beyond the upper limit of the x axis.
• Corrected a bug: the app crashed when changing age or weight inputs after a data file was loaded. This is now fixed.

This version is now archived in a Zenodo repository (https://zenodo.org/record/7384191#.Y4oT-ZrMJPY) following the acceptance of the related paper in Journal of Open Source Software (https://joss.theoj.org/papers/10.21105/joss.04741).

activAnalyzer 1.0.5

• Updated the version numbers of the package dependencies.
• Replaced as.character() by format() in the mark_wear_time() function so that there is no more error when checking for R dev versions.
• Added the argument verify_fa = FALSE to icon() functions in the UI to remove an error message that appeared when running the app.
• Updated the README by indicating the minimum versions to install to use the app.
• Updated the mark_intensity() function: the intensity category numbers associated to the Nonwear, SED, LPA, and MVPA categories (that are present only in the exported marked whole dataset when using the app) were not as expected because they were obtained by converting a factor vector to a numeric vector. Now the conversion is done from a character vector to a numeric vector, which keeps the numerical order as expected. This error had no impact on the results, nor on the figures provided by the package/app. The exported marked dataset has now the corrected intensity category numbers, that is: 0 for Nonwear, 1 for SED, 2 for LPA, and 4 for MVPA.

activAnalyzer 1.0.4

• Added a block of code into the plot_data_with_intensity() function that was erroneously removed during a previous commit. This block of code allows to show grey bands on the figure with intensity data to indicate what periods of the day will not be considered for analysis.
• This version has been accepted on CRAN. From now on, future shinyapps.io and desktop versions will be updated after new CRAN releases to keep consistency across the different ways to use the app.

activAnalyzer 1.0.3

• Added an alerte message with box when validating intensity analysis while all inputs are not correctly defined (before the user could not quickly see that there were some problems if any).
• Improved UI reactivity depending on size screen.
• Added auto-fill button for patient information and intensity analysis.
• Added tests related to default values configuration when clicking on corresponding buttons.