Performance Issues with Orthanc and OHIF DICOM Viewer

General
1

Hello everyone,

We are using Orthanc internally to store DICOM files, which are then used by our OHIF DICOM viewer. When we had fewer than 500 studies, the image rendering time was manageable, averaging around 600ms per frame. However, now that the number of studies has increased to 2,000, the rendering time has slowed significantly, taking around 6–10 seconds per frame.

What could be causing this issue with Orthanc or OHIF? (There are no issues related to resources, internet speed, etc.). My guess is that the problem lies with the default SQLite database, which may also be preventing me from creating replicas.

We are using the following Orthanc Docker image: orthancteam/orthanc:23.12.1

Below is the Orthanc JSON configuration in use:

{
  "Name": "MASTER PACS SERVER",
  "RemoteAccessAllowed": true,
  "AuthenticationEnabled": false,
  "DicomServerEnabled": true,
  "StorageDirectory": "/var/lib/orthanc/db",
  "IndexDirectory": "/var/lib/orthanc/db",
  "HttpCompressionEnabled": true,
  "StorageCompression": true,
  "DicomAet": "MASTER_PACS",
  "DicomPort": 4242,
  "DicomModalities": {
    "MASTER_PACS": [
      "MASTER_PACS",
      "127.0.0.1",
      4242
    ]
  },
  "Plugins": [
    "/usr/share/orthanc/plugins",
    "/usr/local/share/orthanc/plugins"
  ],
  "DicomWeb": {
    "Enable": true,
    "Root": "/dicom-web/",
    "EnableWado": true,
    "WadoRoot": "/wado",
    "Ssl": false,
    "QidoCaseSensitive": true,
    "Host": "",
    "StudiesMetadata": "Full",
    "SeriesMetadata": "Full",
    "PublicRoot": "/dicom-web/"
  }
}

Any suggestions or guidance would be greatly appreciated. Thank you for your attention!

2

Hi,

The first thing to do is to analyse your logs in verbose mode and check the time-stamps to detect what operation is taking so much time.

SQLite should be able to handle this amount of study without any troubles.

Alain

3

Thank you, Alain, for your response. Yes, there is a delay of a few seconds between the REST call, reading the attachment, and sending the response.

I have configured the PostgreSQL database and migrated the data. The speed has improved significantly, now ranging from 100-600 ms, down from 6-10 seconds, for 2,000 larger studies with 400-500 DICOMs per series.

1 Like