Enterprise Imaging: The Enterprise and More

Introduction
Just as PACS is not just about radiology, Enterprise Imaging is not just about the Enterprise.   Incorporating images from outside the Enterprise must be managed as well.   Images outside the firewall include not only images sent in for consultation but those associated with various image sharing needs and telehealth programs operated by the Enterprise. The workflow needs to be defined and the future state optimized for best results for managing images from inside and outside the Enterprise.

Outside the Firewall — Image sharing and Telehealth

Images from outside the enterprise will come in from a variety of routes: CD; VPN; and the cloud. The workflow varies depending on the source of the images and the reason for sending the images.  DICOM CD’s were designed to provide a patient’s prior images for comparison with a current exam or to eliminate a redundant exam.

Other images may be read to determine whether a patient needs to be transferred. Pediatric echo cardiograms are often sent to a children’s hospital for reading, and the diagnosis may or may not include a recommendation for transferring the patient.

Images may also be read as part of an outreach service extending the services of a specialist to outlying areas. Telepathology uses high definition grossing cameras to acquire images of specimen slides that are sent to a pathologist for examination and report.

Images also need to be sent out from the Enterprise for a variety of reasons. Patients admitted to other facilities need imaging exams sent. Participation in clinical trials may require images sent to core labs. Patient access to their imaging exams via patient portals is a recent requirement.

The Enterprise’s current and planned telehealth initiatives and referral patterns can help in prioritizing the order of implementation of the other “-ologies” in an Enterprise Imaging rollout.

Inside the firewall
Looking beyond radiology and cardiology to the other “-ologies” to create an Enterprise Image Archive, it is necessary to understand the status of the technology and workflow in each of the other clinical areas. This information and the telehealth initiatives can be used to create a roadmap for Enterprise Imaging.

The variety of imaging devices in the other “-ologies” can be daunting. One organization[1] found over 250 different endoscopic devices in use. These endoscopes had unique light sources, unique user interfaces, and unique data requirements.

Other “-ologies” include:

  • Bronchoscopy, endoscopic imaging of the lung uses multiple light sources and fluorescence
  • Otolaryngology, endoscopic imaging of the ear, nose, and throat employs numerous types and styles of endoscopes with video frames up to 200 fps and synchronized audio
  • Ophthalmology uses many different types of image devices, ultrasound, angiograpy, corneal tomographic mapping, and other devices generating film
  • Pathology is one of the highest volume image generators (can be greater than 300,000 slides per year) of high resolution images (about 20 GB per image)
  • Dermatology and wound care generate images with a variety of mobile devices (hand held cameras, smartphones, tablets..) that create image quality and uniformity issues
  • Sleep labs store video synchronized with EEG waveforms acquired over a period of hours.

Approach
In defining the Enterprise Image Archive requirements, one needs to identify all of the potential image sources and for each one, determine the image volume estimation, unique user interface requirements, optimal workflow, training and support issues, IS equipment issues, and the EMR integration.

One challenge in defining the Enterprise Image Archive is that some of the “-ologies”  may have a low degree of automation and technology. The current standard for image storage can often be video tape, CD’s, thermal printers, film or nothing. Scheduling may not be done and reporting and billing can be departmental. Standards addressing the other modalities are limited although DICOM supports other image objects including visible light objects and waveforms.

Organizations that have successfully addressed Enterprise Image Archives[1][2] have found advantages in using the PACS model including a RIS or RIS equivalent and DICOM. Some big advantages of DICOM are that it automatically associates image data with demographics and other contextual information. It also allows leveraging of the existing PACS archive. PACS supports the concepts of “like priors” which can retrieve and compare an ENT visible light image with the correlating CT and PET images.

The RIS or RIS Equivalent reliably associates image data with the patient demographics. Anyone who has worked with a system that does not have an ADT feed knows how critical this is. Thus, organizations that have been successful in Enterprise Imaging have extended the PACS model beyond radiology. They employ DICOM, either converting the images or putting a DICOM “wrapper” on the images as appropriate. The RIS is employed to order exams and associate the image data with the patient demographics.

In general, the following steps need to be addressed.

  • Determine the image acquisition model which may be modality based like radiology and cardiology with DICOM always available. It may be non-modality based where the patient goes to a device that does not have DICOM and may need to have images converted and contextual information added but where exams are scheduled. Or it could be mobile based using a smart phone or tablet where orders may have to be created on the fly and images converted to DICOM with contextual information added.
  • Identification of the “RIS equivalent”  that will tie the images to the demographics not just for this acquisition but for all subsequent demographic changes.
  • Storage retention model which defines which data need to be kept and for how long. In dealing with much larger image requirements, on the order of thousands of petabytes for pathology, it is not practical to store and maintain forever.
  • Usage model defining who needs access to the data and how they use the data.
  • Associated reports and meta data. Define what these are and how they are entered into the system.
  • Workflow optimization. Each “-ology” will be different.
  • Support model and budget for the Enterprise Imaging System.

Summary
Enterprise Imaging Systems need to factor in not only the Enterprise requirements but also the needs of users and applications outside the Enterprise. Just as PACS is more than radiology, Enterprise Imaging is more than the Enterprise. Telehealth and image sharing needs can help define the requirements from outside the firewall. Business Continuity Plans and Disaster Recovery Systems will need to be expanded beyond radiology and cardiology imaging to include enterprise imaging. EMR links to the Enterprise System may need to developed and tested.

There are many advantages of using the PACS/DICOM/RIS model for the Enterprise Imaging System. The model is standard based, proven, and mature. It satisfies many of the common needs of the “-ologies” to associate patient demographics with the images, tie metadata to the images, and hanging protocols to automatically retrieve relevant priors.

The variety and complexity of the other “-ologies”, the lack of maturity of the technology, and the manual-based workflow can make an Enterprise Imaging System seem daunting. However, the experience gained in implementing radiology and cardiology PACS, its standards, and the maturity of this technology can be used to advantage in solving the Enterprise Imaging needs. Combining the PACS experience with a structured approach to defining the Enterprise Imaging requirements while concentrating on workflow optimization has yielded positive results.

[1] Wendt, GJ, Medical Imaging In & Outside the Organization, University of Wisconsin-Madision, SIIM 2013, Learning Track 5, June 7, 2013

[2] Chang, PJ, Enterpirse Image Management; “-ology” Image Considerations, University of Chicago, SIIM 2013, Learning Track 5, June 7, 2013

This entry was posted in Enterprise Imaging, General. Bookmark the permalink.

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>