PACS & IMACS SCIENTIFIC PRINCIPLES

In medical imaging informatics

The picture archiving and communication Image Acquisition:
system (PACS) is becoming more commonplace In modern radiology departments, most images
in today’s hospitals because hospital are acquired in a digital format, meaning that the
administrators have come to see the necessity of images are inherently digital and can be
having such a system to serve physicians and transferred via a computer network.
patients.
Ultrasound, computed tomography (CT),
PACS consists of digital acquisition, display magnetic resonance imaging (MRI), and nuclear
workstations, and storage devices interconnected medicine have been digital for many years and
through an intricate network. have been taking advantage of PACS far longer
than general radiography.

As stated earlier, the first PACS served a single

modality, namely, ultrasound. Ultrasound
mini-PACS networks were the norm in many
hospitals.
The PACS is an electronic version of the
radiologist reading room and the file room. The Radiologists routinely made diagnoses by looking
first PACSs were used in the early 1980s and at images on the modality’s computer screen. It
generally served one single modality. was a natural step from there to convert
ultrasound to softcopy reporting, i.e., reading
Digital imaging and communications in images on the computer without hardcopy films.
medicine (DICOM) is a universally accepted
standard for exchanging medical images among As the CT and MRI image sets became larger
the modality, viewing stations, and the archive. because of the increased number of
cross-sectional images per patient, radiologists
Now each modality and PACS communicates via routinely went to the modality to view the images.
DICOM, and it continues to be refined every year.
Every vendor and modality boasts DICOM This slowed down the scanning process for the
compatibility (Figure 8-2), but each DICOM technologists, and vendors began getting
statement must be read carefully to determine the requests for extra console stations for radiologist
extent of the compatibility. viewing.

To understand what a PACS is and how it is used, These workstations were directly connected to the
the following sections break down a PACS into its modalities. Radiologists could view the large
three fundamental parts (Figure 8-3): image stacks of images and perform simple image
acquisition, display workstations, and archive manipulation.
servers.
Display Workstations:
A display workstation is any computer that a
healthcare worker uses to view a digital image. It is
the most interactive part of a PACS, and these
workstations are used inside and outside of This section describes a generic film-based
radiology. workflow and then compares it with a generic
PACS workflow. The workflow in each radiology
The display station receives images from the department is different because there are many
archive or from the various radiology modalities variables.
and presents them for viewing.
Film-Based Workflow:
The display workstation has PACS application Most departments were designed years ago for
software that allows the user to perform minor film and chemical processing. Pass boxes were
image-manipulation techniques to optimize the built into walls that fed into darkrooms and into
image being viewed. Some display stations have large open reading rooms that had gigantic
advanced software to perform more complex multiviewer lightboxes lining the walls (see Figure
image-manipulation techniques. 8-11).

Archive Servers: The following list outlines a typical workflow in a

An archive server is the file room of the PACS. It is radiology department, from entering the order to
composed of a database server or image transcribing the report:
manager, short-term and long-term storage, and
a computer that controls the PACS workflow,
known as a workflow manager (Figure 8-5).

The archive is the central part of the PACS and

houses all of the historic data along with the
current data being generated. In many institutions
the archive serves as the central hub that receives
all images before being released to the
radiologists for interpretation.

Workflow:
Workflow is a term that can be used in any
industry or in any organization. It simply means (RIS), and a requisition is generated. A requisition
how a process is done, step by step. In radiology, generally contains the following information:
we have always used the term workflow to
describe how we complete an examination from ● Patient’s name
order entry to transcribed report. ● Patient’s hospital identification (ID)
number
 ● Date of birth historic images and reports and compares the
● Ordering physician’s name previous images with the current images.
● Examination ordered
● Reason for examination 6. The radiologist dictates a report and has it
● Chief complaint transcribed, or voice recognition software may be
used. If the radiologist uses voice recognition
Generic PACS Workflow: software, he or she can review the report right
The PACS workflow is in many ways different from after dictation, make corrections, and sign the
the film-based workflow (Figure 8-7). report, making it final.

The technologist may get the order via an With PACS it is possible that the time it takes from
electronic worklist or a paper requisition, but after performing the examination to completing the
that, things begin to change. final radiologist’s report is only a couple hours,
compared with a couple days for the film-based
1. Changes in the order entry are on the horizon, workflow.
but for now, the order-entry process is the same
as in film-based departments. The technologist System Architecture:
needs a requisition to verify the patient ID and to System architecture can be defined as the
take a patient history. hardware and software infrastructure of a
computer system. In a PACS, the system
2. The order is input into the RIS, and the RIS sends architecture normally consists of acquisition
a message to the PACS to find all historic images devices, storage, display workstations, and an
and put them on the short-term archive. This image management system.
eliminates waiting for the file room to retrieve a
film jacket from the off-site storage location. The following discussion outlines three common
PACS architectures and takes a look at the flow of
3. The technologist prepares the room, retrieves images after acquisition.
the patient, and performs the patient history. The
history is recorded on the paper requisition or Client/Server-Based Systems:
input electronically into the patient’s computerized In a client/server-based system, images are sent
medical record. directly to the archive server after acquisition and
are centrally located.
4. The technologist performs the examination, and
depending on the type of image acquisition The display workstation functions as a client of the
device, the images are processed and repeated archive server and accesses images based on a
as necessary and sent to the appropriate PACS centralized worklist that is generated at the
device. The patient images have been tagged with archive server.
information from the RIS so that historic image
reports are available at the PACS when the new The health care worker at the display workstation
images are sent. chooses a name from the central list, and the
archive server sends the image data to the display
5. The requisition is either taken to the radiologist, station.
or the radiologist may pull the images from an
electronic worklist. The radiologist also pulls up
After the “client” is finished, the image data is In some systems, the images are sent from the
flushed from its memory. Most systems allow basic modality to the archive server, and the archive
image manipulation at the display workstation or “ server distributes the images to the designated
client,” and the changes are saved on the archive workstation. The reading station designations may
server. be designed based on radiologist reading
preferences.
ADVANTAGES
1. Any examination sent to the PACS is available For example, MRI may be sent to one station and
anywhere without other interventions. CT to another, or all cross-sectional neurological
images may be sent to one station but all body
2. Only one person can open the study with the imaging is sent to another.
intent to read it. Others that open the study will
receive a message that the study is open and The designation is decided after extensive
being read. workflow observation. Moreover, in a distributed
model, the workstations can query and retrieve
3. There is no need to pull or send historic images images from the archive.
to a particular workstation because the old studies
are available with the new on the archive. All images are then stored locally and then are
sent to the archive server after they have been
DISADVANTAGES read. These images remain on the local hard drive
1. The archive server is seen as a single point of of the workstation until they are deleted either by a
failure. If the archive goes down, the entire system user or by system rules.
is down, and no image movement can take place.
All newly acquired images must remain at the ADVANTAGES
modality until the archive is up and can again 1. If the archive server goes down, local reading at
receive the images. the workstations is not interrupted, other than not
being able to get historic images. After the archive
2. The system is very network dependent. The comes back up, the images that have been
images are flying back and forth between the changed and signed off by the radiologist will be
archive and the workstations, and the network can forwarded automatically to the archive to be
become bogged down because of the large saved.
volume of data being moved.
2. Because the images can be distributed to many
3. The archive server is handling many requests at locations at once, copies of an examination exist
once and can become bottlenecked because of at various locations. Therefore it is less likely that
the high volume of requests. PACS data will be lost.

Distributed Systems: 3. The system is less dependent on the network for

In a distributed or stand-alone system, the its speed. The user can be working on one
acquisition modalities send the images to a examination while the workstation is pulling and
designated reading station and possibly to review getting the next examination ready to be read. The
stations, depending on where the order originated workstation can fetch historic images according
(i.e., ICU or ER) (Figure 8-9). to rules the user sets up.
DISADVANTAGES 2. The same application can be used on site and
1. There is heavy reliance on the assumption that at home in teleradiology situations.
the distribution of images is being done correctly.
If the distribution is wrong, the prefetching of Teleradiology is a term used to describe the
historic examinations will not be correct either. reading of images from outside of the hospital's
walls. It can be down the road at the radiologist’s
2. Each workstation has a different worklist, and home or on the other side of the world during
therefore only one person can be working on that nighttime hours.
list at a time.
DISADVANTAGES
3. It can be inconvenient to read additional 1. The system’s functionality may be limited
studies; the radiologist would have to move to because the software is not installed locally. The
another workstation to read the images bandwidth of the network connection limits the
designated for that workstation. amount of data that can be transmitted for
download, and some programs are too large to be
4. The users must depend on the transmitted over the network that is installed.
query-and-retrieve function when nonscheduled
examinations arrive at the workstation to be read. 2. As with client/server systems, the network is the
biggest obstacle to performance.
5. It is also possible for two radiologists to be
reading the same examination and not know that
the other has it until they try to start dictation. The
paper requisition is very important with this type of
PACS.

Web-Based Systems:
A web-based system is very similar to a
client/server system in how data flows. The
significant difference is that both the images and
the application software for the client display are
held centrally.

In a client/server system, the client still has

application software locally loaded to the client,
and only the images are held at the archive.

ADVANTAGES
1. The hardware at the client can be anything that
will support an appropriate web browser. This
allows for greater flexibility with hardware but can
also be a disadvantage because image displays
(monitors) may not be able to support diagnostic
quality.