LifeFlow allows the users to load multiple data files into the system at the same time. Each loaded data file is called dataset. Having multiple datasets loaded, the users can switch between datasets easily. Users can also create a new dataset from a subset of the loaded dataset (Section 3.3). For example, users loaded the dataset and remove patients who survived from the dataset then save it as new dataset called "dead patients". All datasets are listed in the Data Manager Tab.
To start the analysis, let's load the dataset (Section 3.2).
Locating an interesting subset of data is an iterative process, and may involve a number of selection/de-selection actions, in a trial-and-error manner. Therefore, LifeFlow supports setting up checkpoint versions of the data, and enables the users to revert to the dataset to the previous checkpoint, or to the original (from file) version. After the users successfully locate an interesting subset of records, they can also rename or save the data as a new dataset. All manipulation actions can be reached from the menu "Dataset" on the top of the application.
Figure: Dataset manipulation menu
This operation changes the name of the dataset. For example, after selecting all the records end with event "Die" and removing the others. The data can be renamed into "ends with Die" as a note to the users.
This operation creates a new dataset with the current data.
Revert the visualization to the original stage, where all records are visible and none is selected. Note that renaming the dataset does NOT change the original stage. In case the users want to change the original stage, the dataset must be saved into another file.
After you have finished exploring the dataset in LifeFlow, you may want to save your work. LifeFlow supports saving the dataset after it has been modified (e.g. selection of a subset of the original dataset), so the users can load the modified dataset and continue the analysis later. LifeFlow provides two ways to save the dataset: save to the original file, or save to a new file.
Figure: Save dataset to file from the Data Manager Tab.
LifeFlow allow users to save configurations, such as colors of the events and other display options, which makes it more convenient when the users want to perform the analysis later, or copy the same configuration to use with other datasets.
By default, when the users load the data file,
LifeFlow will try to load configuration (config) file with the name <filename>_config.xml in the same directory with the data file.
If the file does not exist, it will try to find the default config file
(default_config.xml) in the same directory with the data file.
If the default config file also does not exist, no config file will be loaded automatically.
However, users still can load other config files manually (Section 3.5.1).
If the config file is loaded, the Data Manager Tab will show the name of the current config file. (See Figure)
The config file is XML, so users can also modify it manually outside of LifeFlow.
Figure: Managing config file from the Data Manager Tab.
This feature is useful when the users modify the XML config file manually outside of LifeFlow.
<filename>_config.xml
or default_config.xml.