Biomedical Engineering Reference
In-Depth Information
example will only lightly touch issues of Health Insurance Portability and
Accountability Act (HIPPA) security including the agreements, auditing,
and proper practices that need to be in place to allow end-users to operate
on the data for purposes of healthcare operations (e.g. large insurance
company), or public health (e.g. Center For Disease Control). Although
non-trivial, such considerations apply to any technological solution and
form a discussion out of scope for the main emphasis of this chapter.
20.2 Interoperability
The fi rst step in clinical analytics is to carefully gather as much meaningful
data as possible and process it into a form amenable to downstream analysis.
The diagnosis and treatment of patients generates a tremendous amount of
data from a large number of sources. A typical hospital will have any number
of different systems such as health information systems (HIS [2]), radiology
information systems (RIS), computerized physician order entry (CPOE), and
electronic health records (EHR). Many hospitals may even have multiple
systems from different vendors. How do they make sense of all of this
complex health information coming from various different sources?
This was the reason that Health Level Seven International (HL7 [3, 4])
was founded. It is a not-for-profi t organization developing standards 'for
the exchange, integration, sharing, and retrieval of electronic health
information that supports clinical practice and the management, delivery
and evaluation of health services' [3]. Its standards attempt to navigate
the thin line between enabling interoperability and allowing vendors
and hospitals fl exibility in a complex domain. This is an important
point because the data feeds from the individual systems may require
transformation in order to be consistent in the analytical data store. HL7
covers a broad range of topics, including application interoperability at
the user interface (UI) level (e.g. Clinical Context Object Workgroup or
CCOW) and Medical Logic Modules. The two aspects most relevant for
analytics are the messaging standard and Clinical Document Architecture
(CDA). HL7 V2.x is still the most widely used set of standards. The V2.x
message is a character string with delimiters differentiating parts of the
message. Messages are made up of segments as shown in Figure 20.2.
Figure 20.2
HL7 V2.x message sample
Search WWH ::
Custom Search