laboration of the structure of MSOT data in particular. Include diagrams.
The data acquired from the MSOT is structured in a split structure partially determined by file structure and partially determined by folder structure.
At the highest level, a study folder is entirely self-contained and composed of instrumentation settings, analysis settings, metadata, raw binary data, and processed data. This facilitates the ability to copy MSOT data from one analysis environment to another. This document will review the structure of MSOT data, as well as the loading interface for MATLAB.
SS1.1 Gross folder structure
Studies may be organized at a gross level by simply saving study folders in different locations. The default name for a study is assigned at scan time, and is of the form "Study_<#>", but may be named anything which is a valid folder name.
Within a given study, the scans are organized by giving each scan an index corresponding to the next available name, which is automatically assigned, unchangeable, and of the form "Scan_<#>". There is also a *.nod file, of the same name as the overall study, which holds metadata for the entire study. This file is where the user-assigned scan names are stored.
Within each scan folder, there are five files and possibly some folders.
The files inherit the "Scan_<#>" name of their parent folder:
-Scan_<#>.bin: This file contains the raw ADC data acquired over the scan duration. The values are stored as 16-bit unsigned integers (uint16), each corresponding to the value acquired by a given transducer at a particular timepoint. The structure of this file is determined by the measurement settings contained in the corresponding .msot file.
-Scan_<#>.irf: This file contains the impulse response function which characterizes the electrical and physical response of the imaging system. This is used in image reconstruction, and is stored as double-precision floating point.
-Scan_<#>.msot: This file contains nearly all of the metadata for a particular scan, and is used extensively in the data analysis process. Data is stored as a modified XML schema using XMLBeans, compiled using a schema defined by iThera Gmbh. The structure of this file is elaborated below, in
The folders in the Scan_<#> folder only exist if processing has been performed using ViewMSOT software (or if they were defined for some reason): -RECONs: Contains .bin and .tif files corresponding to different reconstructions performed on the MSOT data. Each is assigned a GUID at write time, which functions as a unique identifier to allow metadata lookup for a given scan. The .bin file contains reconstructed images, stored as doubles, and with structure defined by the reconstruction settings. The .tif file contains thumbnail images which can be used for rapid dataset peeking. -MSPs: Contains .bin, .tif, and .csv files corresponding to instances of multispectral processing in ViewMSOT. A given MSP is defined by a GUID (separate from that of the corresponding recon), which is shared among all three files. The .bin and .tif files are similar to those in the recon folder, but with the wavelength dimension replaced by an endmember index. The function of the .csv file is to provide the stored spectra used in the unmixing process, and uses commas as a decimal separator while semicolons separate datapoints and newlines separate different variables (i.e., wavelength vs. absorption).The structure of the metadata is defined in the .msot file, but for the purposes of reconstruction on the cluster there are only a few particularities to use.