Otherstudentsposting.docx

Christina

Unstructured data are usually free form narratives that can’t be easily categorized and organized with analytic software  (Managing Unstructured Healthcare Data, n.d.).  In contrast, structured data is information that can be categorized, organized, and displayed in a consistent format. 

After we are done gathering images on a patient for an echocardiogram, sonographers proceed to type a report of the findings into our reporting system which is on a PACS viewer.  Within the reporting structure, there are many fields with drop down menus to describe technical findings.  For example, under the aortic valve, there is a drop down menu that has options such as normal/trileaflet, calcified (with another drop down to specify: mild, moderate severe), bicuspid, etc.  It is a long list but it is not exhaustive, about 99% of the time it has options that we use.  For those few times that it the descriptions do not suffice, we use a free form manual text option.  We are discouraged from using manual texts because these manual texts then become not searchable.  For example, let’s say that I’m trying to find all the echos within a time frame in which there was a vegetation on the aortic valve.  If I had used the drop down menu, the echo would be pulled up.  But, if I wanted to further describe the vegetation in my own words, search options would not pull up my study.  In this way, I think that the drop down options in my report could be classified as structured data and the manual entries as unstructured data within the PACS system.   In this situation, I think that the way to overcome the obstacle is to add more entries to the drop down menus.  Although, I think that it can become a problem if these lists get too long because this alone would discourage a person to use the drop down menus.  If it’s going to take me 2 minutes to hunt for the description I need, I’d rather manually enter it.  Another thought is that maybe sonographers chose a disease type, and use the manual entry only for the description that they want to add.  Continuing with our example from above, I’d chose aortic valve vegetation from the drop down and under manual entry, I’d only put in my description that was not on the drop down.

Reference

Managing Unstructured Healthcare Data. (n.d.). HealthTerm. Retrieved February 3, 2023, from 

Meron

Information that cannot be effectively managed using specified data models is referred to as unstructured data. which makes it challenging to maintain and store in a common relational database. Since unstructured data lacks a predefined data model or structure, it does have an inherent structure. It could be textual or not. It may be created by a machine or a human. 

One example of unstructured data is both machines and human-generated unstructured data are produced in significant quantities in the healthcare industry. Machine-generated data consists of information gathered by medical imaging tools such as surgery robots, endoscopes, and emergency video cameras as well as biosignal information from patient monitors in operating rooms and intensive care units. There is also a ton of data produced by wearable health monitoring gadgets. The interactions between patients and medical staff that are written down or captured as audio recordings could be considered human-generated data. another one is when we look at the vast majority of documents used in business, including emails, reports, and presentations, which are unstructured and contain data in the form of text, photos, numbers, or video. In the enterprise, these documents serve as significant knowledge reservoirs, but they are currently mostly underutilized if they cannot be linked to structured information systems. Making content readable, searchable, and organized is the first step toward overcoming the problem of unstructured data. You must preprocess data if you want to increase the quality of your datasets. Unstructured data needs to be saved, arranged, and cleansed before analysis. using text analysis technologies with AI. 

Reference:

Camilla. (2019, July 10).  Managing unstructured healthcare data. HealthTerm. Retrieved February 6, 2023,

from 

Syrel

Unstructured data is data that cannot be stored in a relational database and is too complex to be interpreted. Unstructured data can include data such as videos, audio, or photos. An example of unstructured data in ancillary services can be texts that contain a lot of words such as evaluations or medical notes and using AI technology can help overcome interpreting this unstructured data which can extract the key ideas in a medical report making the information easier to understand.

Reference:

Victor. (2022, December 13).  Structured vs unstructured data in Healthcare (with examples). TrueNorth ITG. Retrieved February 6, 2023, from