Wednesday, January 8, 2020

Indexing DICOM files in a directory into a CSV file using PowerShell and MODALIZER-SDK

This post is continues my series on PowerShell DICOM scripting. It covers a very common task that repeats itself in all kind of forms where I have a directory full of DICOM files that piled up somewhere and needs to be indexed and processed. PowerShell scripting comes handy in these cases where there's a need to do something fast and probably change it on the fly. Of course, you will need our MODALIZER-SDK for this one.

#!PowerShell ... jeje ... this doesn't work on widows ;-) 

# Well, this is something very useful (I think) for anyone managing a PACS.
# Lets say you have a directory full with DICOM files you have no idea what's in them (you do, right?) and you want to scan through them.
# So, here's a little powershell script for this:

# Ok, so first we get the directory to scan as a command line parameter
param ($ScanPath)

# This is a function. It extract one element from a DICOM object and return an empty string if the element doesn't exist or doesn't have a value. We use it later on.
function Get-Value($obj, $tag
{
  $e = $obj.GetElement($tag)
  if ($e -ne $null)
  {
    if ($e.Value -ne $null)
    { 
      return $e.Value
    }
    else
    {
      return ""
    }
  }
  else
  {
    return "";
  }
}

# Modify this array to add more tags. The pair items are used for headers (and readabity).
$tags = @(
    0x00080016, "sop class uid",
    0x00080018, "sop instance uid",
    0x00080020, "study date",
    0x00080030, "study time",
    0x00080050, "accession number",
    0x00100010, "patient name",
    0x00100020, "patient id",
    0x00100030, "patient birth date",
    0x00100040, "patient sex",
    0x0020000d, "study Instance UID",
    0x0020000e, "$series Instance UID")

# This prints a nice CSV header line. 
# Note all the '","' to enclose values in double quotes and separate them with commas. 
function Print-Header
{
    $line = '"Filename","Status'
    for (($i=1); ($i -lt $tags.Count); ($i+=2))
    {
        $line += '","'
        $line += $tags[$i]
    }
    $line += '"'
    $line
}

# This is where the element values are extracted and one line is returned. 
# The filename is in the first column and a status is in the second. Then the values from the tags array follow.
# The status  can be "OK" or "ERROR". You will also see errors in stderr.
function Parse-DicomFile($filename)
{
    $line = '"'
    $line += $filename
    $line += '","OK'
    try {
        $obj = New-Object -ComObject rzdcx.DCXOBJ
        $obj.openFile($filename)
        for (($i=0); ($i -lt $tags.Count); ($i+=2))
        {
            $line += '","'
            $line += Get-Value $obj $tags[$i]
        }
        $line+='"'
    } catch { $line = '"' + $filename+ '","ERROR"' }
    return $line
}

# Here it starts. We first print the header and then scan through the folder and extract the data from every DICOM file. 
# Note ignoring directories
Print-Header
Get-ChildItem -Recurse -Path $ScanPath | Foreach-Object {    
    if ($_.Attributes -ne "Directory") {
        Parse-DicomFile $_.FullName
    }

# Enjoy!!!



No comments:

Post a Comment