Page 53 - Big Data Analytics for Intelligent Healthcare Management

3.3 HEALTHCARE DATA 45

3.3 HEALTHCARE DATA
Data generated and obtained in healthcare industries are the healthcare data. This type of data can be
mainly classified as structured, semistructured, and unstructured data. These different types of data can
be defined as below [6].

3.3.1 STRUCTURED DATA
Structured data are those data that can be captured and stored as discrete coded values. One example of
such data is Logical Observation Identifiers Names and Codes (LOINC), which is a database and uni-
versal standard for identifying medical laboratory observations [6]. When healthcare data is stored in a
structured way, it becomes very easy for computer systems to pass data back and forth. This makes the
analysis phase easier. Nowadays, structured data is given more preference such as the data in diagnosis,
laboratories, procedure orders, and medications are kept as structured data [7].

3.3.2 UNSTRUCTURED DATA
Unstructured data is generally captured and stored as free text. Although humans can easily read this
data, they are not readable for computers. The common examples of unstructured data are progress
notes, pathology results, and radiology reports etc. [7].

3.3.3 SEMISTRUCTURED DATA
Semistructured data can be defined as a mixture of structured and unstructured data. One example of
semistructured data can be a data entry interface that may have a grouping of structured data capture
and free text.
If we categorize the healthcare big data per application, then the following classification is used [3].

3.3.4 GENOMIC DATA
Genomic data generally means the genomic and DNA data of an organism. In bioinformatics, scientists
have been gathering, accumulating, and processing the genomes of different living things. This type of
data is generally very big in size and requires big storage and specially built bioinformatics software to
analyze them.

3.3.5 PATIENT BEHAVIOR AND SENTIMENT DATA
These are the data generally collected or gathered through different social media avenues. Patients can
provide their valuable feedback and information about their own experiences with doctors, nurses, and
other staff members of a particular hospital or clinic [8] [9]. By sharing this information, they can seek
help in different social media channels, patient surveys, and discussion forms. This information is col-
lected in great amounts and valuable facts can be extracted from them if they are analyzed properly with
an efficient big data analytics tool.

3.3.6 CLINICAL DATA AND CLINICAL NOTES
Clinical data has always proved to be the main resource for health and medical research problems [10].
Clinical data are generally collected during the course of ongoing patient care. These data can also be
formulated as a part of a formal clinical trial program. These data are mostly unstructured (about 80%)
for example, documents, images, and clinical or transcribed notes.