Biomedical Engineering Reference
In-Depth Information
possible formats and data types, means that manually reading data from AFM files is
rather complicated. Additional information that might be included in the header include
information about the instrument and software used for acquisition and descriptions of the
sample, and even the probe used (these would be entered by the user during acquisition).
There can be hundreds of lines of text stored in the header block of an AFM file. Following
the header, the data section stores the individual data points as a simple block of
consecutive 16 bit numbers (double bytes). The scan size and distances between these
points are obtained from the information stored in the header.
AFM analysis software
As may be surmised from the brief description of AFM file formats above, writing
software to open AFM files correctly is not a simple task. The best way for the user to
analyse their data is always by opening the file in the analysis software provided by the
manufacturer of the instrument on which the file was acquired. This will be the only
analysis package that is sure to open and read the data correctly. There are many (more
than 50) different AFM file formats, which are all mutually incompatible. Furthermore,
many formats change over time, as the capabilities of the instruments are improved, so
a program that can open old files from one format many not be compatible with the
newer files. All this means that creating tools to read, manipulate and write these
formats is far from trivial. However, some third party software does exist, and a list of
third party software capable of opening and manipulating AFM data is included in
Appendix C.
In this chapter, the operations carried out on AFM data are separated into processing,
display, and analysis steps. The most important thing to remember when manipulating
AFM data is to maintain the integrity of the original file as saved by the data acquisition
software. The user should never save the AFM file over the original after changing the data
in any way. The results of any operations which need to be saved or exported should
be saved as new files-either new AFM data files, or as simple bit-map image files (.bmp,
.jpeg, etc). All AFM analysis software has facilities to export to such bit-map files. One
reason to avoid ever altering the original file is that AFM data files always contain more
information than the user can see at one time. This can be in the form of alternate channels,
more data resolution than is visible (there are more than 65,000 possible levels in the
z
-scale, whereas fewer than a hundred can be distinguished by humans), or in the form of
the 'metadata' stored in the header block of the file. Furthermore, many of the processing
steps that are referred to below cannot be undone. Thus, by processing, we can lose the
ability to see the original data. Therefore, the user should always keep a backup copy of
their original AFM data, and any processed files should be saved with a new name.
Processing steps change the AFM data; they include functions like filtering and back-
ground subtraction. Occasionally an image may not need any processing at all, but this is
not common with most samples. All processing is done with the aim of clarifying the data
already within the file. In other words, the purpose is to make it easier to measure and
observe the features that have been measured.
