simrisc (14.05.00)

  * Added option --err (-e) to select the previously used but inorrect
    algorithm for computing the Beir7 risk vector, instead of using the
    (now default) correct algorithm for computing the risk vector.

 -- Frank B. Brokken <f.b.brokken@rug.nl>  Sat, 22 May 2021 22:09:08 +0200

simrisc (14.04.00)

  * Modalities support spread-variation. See the simriscparams(7) man-page for
    a description of how the configuration file was extended wrt the
    Modality-specifications.  When 'spead: true' is specified then
    Loop::iterate() now calls d_modalities.vary(), calling ModBase vary for
    the various Modalities.  Previously defined 'double' Modality values now
    use VSDs

  * Age range specifications do not end in ':'s anymore. E.g.,
            ageGroup:  1 - 50

  * Beir7's parameter specification first specifies eta, then beta.

  * The Case-specific data matrix defines an extra (18th) colum, showing
    the results of the screening rounds for each simulated case.

  * When 'spread: true' is specified the actually used and original parameter
    values are listed in a file, by default 'spread-$.txt', where $ is
    replaced by the loop iteration index. Use the option '-s' to specify a
    non-default filename (cf. simrisc(1));

  * The man-pages were update

 -- Frank B. Brokken <f.b.brokken@rug.nl>  Sun, 11 Apr 2021 11:00:58 +0200

simrisc (14.03.00)

  * To specify no screening round ages 'round: none' should be specified. When
    'round: none' is specified then the rounds file is not written.

 -- Frank B. Brokken <f.b.brokken@rug.nl>  Sun, 21 Feb 2021 13:30:38 +0100

simrisc (14.02.00)

  * Option --cases was renamed to --last-case (-l) to avoid confusion with
    the cases: specification in the configuration file

  * Default option values are reset at each Analysis: specification.

  * Analysis files now use (capitalized) Analysis: specifications.

  * Fixed some specification errors in the simriscparams(7) man-page

  * Updated the simrisc(1) man-page

  * Updated the usage info

 -- Frank B. Brokken <f.b.brokken@rug.nl>  Wed, 27 Jan 2021 16:39:02 +0100

simrisc (14.01.00)

  * 'm:' parameter specifications for the four bi-rad categories were added to
    the Mammo specifications in the default configuration file

  * The 'm:' parameter corresponding to the cases bi-rad category given the
    case's age is used when computing the sensitivity associated with the
    Mammo modality

  * The simriscparams(7) man-page was updated accordingly

  * The simrisc --help info lists the --cases (-c) option instead of --nCases
    (-n). 

 -- Frank B. Brokken <f.b.brokken@rug.nl>  Fri, 22 Jan 2021 11:14:22 +0100

simrsic (14.00.00)

  * Case-specific data are reproducible when specifying 'generator: fixed'
    or 'generator: increasing'

  * Parameter variation when specifying 'spread: true' is available

  * The organization of the configuration file was drastically modified

  * The validity of parameter value specifications is checked

 -- Frank B. Brokken <f.b.brokken@rug.nl>  Wed, 20 Jan 2021 20:43:53 +0100

simrsic (13.03.00)

  * Redefined variables in the Costs class: the discount percentage in fact is
    a proportion, and its type was adapted accordingly

  * MRI simulation was incorrectly implemented in the original program. After
    fixing this error handling False Negatives is now using a standard
    procedure for all Modalities

  * The configuration file and the simriscparams man-page were adapted
    accordingly. 

  * The classes Options, Random, and Error are now singletons.

 -- Frank B. Brokken <f.b.brokken@rug.nl>  Tue, 27 Oct 2020 19:56:12 +0100

simrsic (13.02.01)

  * Added a Copyright-notice file

 -- Frank B. Brokken <f.b.brokken@rug.nl>  Mon, 12 Oct 2020 09:34:04 +0200

simrsic (13.02.00)

  * Specifications using multiple analyses no longer overwrite existing files

  * Added specification 'generator' to the scenario specifications to specify
    the way the random number generator is initialized

  * The classes Pool, IDpool, and RandomPool are superfluous

  * MRI handling is activated.

  * The loop Status enum now uses 'PRESENT' instead of 'ALIVE' and Status enum
    values beginning with 'LEFT_' instead of 'NATURAL_'

 -- Frank B. Brokken <f.b.brokken@rug.nl>  Tue, 29 Sep 2020 16:04:26 +0200

simrsic (13.01.00)

  * Added missing registration of d_roundDetected to loop/maybedetect.cc

  * loop/selfdetected.cc and loop/treatmentdeath.cc are identical except for
    used Status values: now inline, calling new function in
    loop/characteristics.cc 

  * Added a new section OUTPUT to the simrisc(3) man-page, containing
    descriptions of the content of generated files

  * By default the id-base option uses value 1



simrsic (13.00.00-pre)

  * Reorganized config/simrisc, costs associated with Modalities are now
    defined by the individual modality-specifications

  * Repaired the computation of d_prePeriod in tumor/reset.cc 

  * Implemented Danielle's code and modified the handling of multiple
    modalities 

  * Standardized the handling of Modalities by defining one standard interface

  * Updated the man-pages

 -- Frank B. Brokken <f.b.brokken@rug.nl>  Thu, 16 Jul 2020 13:45:32 +0200

simrsic (12.03.01)

  * Dropped 'oneAnalysis' and 'showAll' from Options, added specificAges(),
    fixing a bug in handling specific ages / N-cases specifications
    
 -- Frank B. Brokken <f.b.brokken@rug.nl>  Fri, 29 May 2020 15:52:30 +0200

simrsic (12.03.00)

  * Fixed handling of filename specifications in analysis: sections

  * The --one-analysis option no longer requires command-line arguments

  * Analysis: specifications may be empty

  * Minor update of the man-page

 -- Frank B. Brokken <f.b.brokken@rug.nl>  Fri, 29 May 2020 13:39:29 +0200

simrsic (12.02.02)

  * Changed the content of /usr/share/doc/simrisc: it now contains the
    README.hierarchy file and the default configuration file (simrisc.gz)

 -- Frank B. Brokken <f.b.brokken@rug.nl>  Tue, 26 May 2020 13:38:10 +0200

simrsic (12.02.01)

  * Updated the short option letters of the program's usage info 

 -- Frank B. Brokken <f.b.brokken@rug.nl>  Tue, 26 May 2020 11:14:14 +0200

simrsic (12.02.00)

  * Added option --id-base to use id-based random numbers instead of
    random-pool based random numbers.

  * All options referring to filesystem entries use capitalized single letter
    options; all other options are lowercase

  * Updated the man-pages.

  * The default parameter specification file is available as the file
    'simrisc' in the distribution's .config/ directory

 -- Frank B. Brokken <f.b.brokken@rug.nl>  Mon, 25 May 2020 15:48:49 +0200

simrsic (12.01.01)

  * New sub-minor release due to a man-page bug.

 -- Frank B. Brokken <f.b.brokken@rug.nl>  Sat, 23 May 2020 21:20:04 +0200

simrsic (12.01.00)

  * Reorganized the class hierarchy, preparing for multi-threading: see
    README.hierarcy. 

  * Base directories are created if not available

  * Output files are provided with a timestamp and optionally analysis labels

  * Output files are cleaned up: improved table formats, and CSVs for the data
    files 

  * The file echoing the used parameters is by default not produced. The
    --parameters option is used to specify the name of that file if requested.

  * When specifying configuration parameters in the context of an analysis
    either all, or none of the parameters of a category must be provided. This
    procedure is used to allow reduction of, e.g., screening rounds.

 -- Frank B. Brokken <f.b.brokken@rug.nl>  Sat, 23 May 2020 19:26:48 +0200

simrsic (12.00.00)

  * Following a request by Marcel Greuter (m.j.w.greuter@umcg.nl) SimRiSc was
    completely rewrittin in this release. Earlier specific changelog
    version information is not available. In the distribution earlier
    changelog info is available under 'legacy'

  * This is the initial conversion, offering the basic facilities. 
    The current program produces the same results as the original program,
    for a single scenario, but specifying 100'000 women is OK.

  * Lots of work is still required to update the man-pages, and to completely
    generalize the program.

  * Debian package construction is (locally) operational.

 -- Frank B. Brokken <f.b.brokken@rug.nl>  Sun, 17 May 2020 17:47:06 +0200


