Brainscape's Knowledge GenomeTM


Top Flashcards (Certified)
All Flashcards

HI-410 > HI-410 Exam 2 > Flashcards

HI-410 Exam 2 Flashcards

(98 cards)

Start Studying
1
Q

Comma-separated values (CSV)

A

-A flat file database format where fields are delimited by a comma.
-Might include the variable names in the first row.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

Flat File

A

-Text file, usually delimited by a comma or tab, with one record found on each row.

  • It has only one table of data.
    -An example: if a practice manager wished to track the number of records coded by each employee, then this is sufficient.
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

Relational Databases

A

-Data with a common purpose, concept, or source are arranged into tables
- Structured in a way that helps ensure data integrity.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

Relational Databases are

A

-Displayed in an entity relationship diagram (ERD). ERD visually illustrates the relationship between different entities such as patients and items.
- ERD may be used in industries such as engineering, business, and healthcare.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

How are relational databases structured?

A

1.) Primary Key,
2.) Foreign Key 3.) Cardinality

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

Primary Key

A

Uniquely identifies the row in the database.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

Foreign Key

A

Is a variable in one table that is a primary key in another table

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

Entity Relationship Diagram (ERD)

A

Visually illustrates the relationship between different entities such as patients and items.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

Cardinality

A

Represents the relationship between the two tables

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

Object-oriented

A

-Designed to handle data types beyond text and numbers,
-May be used to store images or videos.
-Stores two types of information about the object.
- The first element is the data itself (audio clip, image, video file, and such). -The second element stored describes how to use the data and is called the method.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

Hierarchical databases

A

Common in EHRs, has patient-child relations and great for maternity wards.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

Data dictionary

A

A tool that provides metadata, or data about data, to support and adopt more consistent use of data elements and terminology. to improve the use of data in reporting

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

What does the data dictionary provide?

A

Provides standardization to promote clearer understanding and promotes consistent and efficient use of information

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

SQL (Structured Query Language)

A

Programming language that is used to manipulate data in a relational database.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

How can Structured Query Language(SQL) be used?

A

1.) Is a client/server language. Personal computer programs use this to communicate over a network with database servers that store shared data.

2.) A database programming language. Programmers embed these commands into their application programs to access the data in a database.

3.) An Internet data access language. Internet web servers that interact with corporate data and Internet application servers all use this as a standard language for accessing corporate databases.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

Entity Relationship Diagram(EDR)

A

-“A specific type of data modeling used in conceptual data modeling and the logical-level modeling of relational databases

-Visually illustrates the relationship between different entities such as patients and items

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
17
Q

DBMS (database management system)

A

Provides a method for adding, updating, or deleting data and also supports methods for extracting data for various purposes to support organizational decision-making

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
18
Q

Examples of DBMS

A

-Microsoft Access
-Microsoft SQL Server
-Oracle
-MySQL
-PostgreSQL

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
19
Q

Data Flow Diagram

A

The conceptual data model may be mapped using a context-level data flow diagram, which maps out the database’s boundary and scope.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
20
Q
  • The conceptual data model may be mapped using
A

a context-level data flow diagram

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
21
Q

Diagram 0

A

Expands on the context diagram and adds details regarding the tables and their relationships.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
22
Q

Data warehouse

A

-Stores large amounts of data (important aspects) for decision support databases.

-To have one you have to have a common verbage. Everyone has to use the same verbiage.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
23
Q

Normalized tables

A

Reduce redundancy and improve data integrity by organizing data into related tables.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
24
Q

One-to-One Relationship

A

A record in Table A corresponds to one record in Table B.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
25
One-to-Many Relationship
-A record in Table A can relate to multiple records in Table B. -Each row in one table may relate to many rows in a second table; each row in the second table relates to only one row in the table.
26
Many-to-Many Relationship
-Records in Table A can relate to multiple records in Table B and vice versa, often requiring a junction table. -Each row in one table may relate may relate to many rows in a second table, each row in the second table may relate to many rows in first table.
27
Extract
-The process of copying the data into the ETL system for manipulation. -Once the data are extracted, they are essentially part of the data warehouse.
28
Transform
Ensures the data are compatible. Any data mapping, correction, or translation occurs here.
29
Load (ETL)
-The final step in moving data from the source to the data warehouse. -This is the step when the data are physically loaded into tables that make up the data warehouse.
30
DSS (Decision Support Systems)
Computer-based system that gathers data from various sources and assists in providing structure to the data.
31
What are the two types of DSS's?
1.) Administrative Support Systems(ADS) 2.)Clinical Decision Support Systems(CDSS)
32
Administrative Decision Systems (ADS)
-Supports organizational decisions using data sources that could include reimbursement, utilization of services, and aggregate patient sociological data. -Focuses on primarily improving efficiency of care being provided -Improves administration
33
Clinical Decision Support Systems (CDSS)
-Includes alerts in the electronic health record (EHR) such as an allergy to medications, reminders for preventive healthcare services, and links for providers to find references or order sets to assist clinicians in making patient care decisions. -Provide correct information, to the right person, in correct format, and correct channel at the same time.
34
NLP (Natural Language Processing)
-Technology that converts human language (structured or unstructured) into data that can be manipulated by computer systems. -Examples would be: voice→text and dragon (these are efficiency tools)
35
Computer-assisted coding (CAC)
The process of extracting and translating dictated and then transcribed free-text data into ICD and CPT with menu-driven prompts evaluation and management codes for billing and coding purposes" -Looks at unstructured data and pick up keywords
36
Clinical Data
Mining discrete patient healthcare data to make clinical decisions or to aid in translating data for research and to further healthcare treatment
37
Data Analytics
-This use of statistical analysis of data to make business decision.
38
Computerized Provider Order Entry (CPOE)
An electronic system that allows healthcare providers to enter and manage patient treatment orders.
39
ITSA Framework
Framework that outlines the integration of information technology and clinical services to improve healthcare delivery.
40
Unintended Consequences of DSS
-Over-reliance on technology, reduced critical thinking, potential data privacy issues, and possible inaccuracies if the underlying data is flawed.
41
Data Cleaning
-The processing correcting or removing inaccurate records from datasets. -Which is a procedure used to manipulate and cleaning raw data so that they are ready for further analysis
42
Vulnerable Subject
-Individuals at higher risk in research studies due to their circumstances.
43
What types of people are considered Vulnerable subjects?
-Are children, pregnant women, human fetuses, neonates, mentally disabled individuals, educationally or economically disadvantaged, prisoners or persons with incurable or fatal diseases
44
Metadata
-Is information about the data in the database, typically maintained at the system level and the user level. -Data that provides information about other data, such as its source, format, and structure.
45
What does metadata include?
The names of all tables, objects, fields, indexes, primary and foreign keys, usernames, and privileges for the database.
46
Quantitative Research
research that collects and reports data primarily in numerical form and they are measurable and very definitive.
47
Qualitative Research
-Exploring non-numerical observations and perceptions. -Example: Survey based on opinon/ perception of patient study
48
Basic Research
-Can also be called pure or theoretical. -Aims to expand knowledge without immediate application
49
Applied Research
-Aimed at solving practical problems -Also called pratical or clinical research "focuses on the use of scientific theories to improve actual practice"
50
Correlational Research
Looking to see if there is any relationship between two different variables. Example: Increasing the pay for overtime→ you see the correlation with people taking longer to give reports at the end of the shifts so they get more time to get paid)
51
Quasi-Experimental Research
-When you know there is about to be a change -Appropriate when variables CANNOT & SHOULD NOT change. -Example: measuring the well-being of employees before and after merger
52
Clinical Outcome Research
-This is where your quality research is coming from and minor modifications are coming from here. -This research looks to improve the delivery of patient care by studying the end result of healthcare services. -Example: looking to see if a new process that's being implemented for putting a stent in is going to decrease the number of infections.
53
Independent Variable
Variable believed to cause change in another variable.
54
Dependent Variable
Variable being measured in a study.
55
Reliability
-Consistency of a measure across repeated tests. -Refers to the consistency of a measure
56
Validity
-Accuracy of a measure in representing its intent. -Is the extent to which an instrument measures what it is supposed to measure
57
Nominal Data
-Expressed in categories -Categorical data with a defined order (rankings.)
58
Ordinal Data
Categorical data with a defined order.
59
Interval Data
Numeric data with equal intervals, no true zero. -Example: Temperature -Continuous data
60
Ratio Data
Numeric data with a true zero point. Example: Weight
61
Data Cleaning
- A procedure used to manipulate and cleaning raw data so that they are ready for further analysis. -Includes removing unnecessary data or outliers, removing duplicates.
62
Crosstabs
-Displays relationships between two categorical variables. -Serve to highlight any errors where there is an expected or explicit relationship between two data elements.
63
Descriptive Statistics
-Summarizes data features like mean and median. -Used to analyze study variables like age and number of ED visits because they can describe the features of the data, but they are not used for hypothesis testing.
64
Structured Data
-Organized data in a defined format. -Is appropriate and highly recommended for data entry when the options are limited or are required to conform to a specific standard. -Highly desired for uses such as statistical analysis, clinical decision support, and quality measure reporting.
65
Unstructured Data
Data lacking a specific format or organization. Example: Text -Might be organized into data fields or text boxes such as the history of illness and a progress note for the physician's hospital visit.
66
Categorical Data
Represents distinct groups like gender and race and comorbidity
67
Continuous Data
Data that can take any value within a range.
68
Primary Data
Firsthand data collected for clinical care and facilitated billing
69
Secondary Data
-Analyzed data from existing sources. -The "non-direct care use of personal health information. -Example: One example is performing research studies on the data to determine the risk factors for 30-day readmission after hospital discharge.
70
Why is unstructured data preferred more than Structured Data?
-Often preferred because they enable providers to document details and nuances that are usually not available with structured data.
71
How to handle Missing Data?
Identify and address missing values in datasets. -Deletion, mean substitution, and regression imputation are commonly used to handle missing values.
72
Deletion
One can simply delete the data points that have the missing values. However, this could result in the loss of data, which is critical when the data set is small.
73
Mean Subsitution:
-the mean values of a variable are used in place of the missing values in that variable. This method assumes that the mean is a realistic estimate for a randomly selected observation from a normal distribution. -May result in an inconsistent bias when missing values are not strictly random, especially when there is a large disparity in the number of missing values for the different variables.
74
Regression:
Imputation is a comprehensive technique that replaces the missing values with a probable value, estimated by statistical regression techniques using existing values
75
Other ways to handle missing data?
-Could be coded with an identifier such as -9999 to be more easily located and dealt with during the data analysis. -Be provided using a process called imputation in which missing data values are replaced with some calculated values. Imputation methods must be described fully and imputed values clearly labeled.
76
Regression Imputation
Estimates missing values using statistical regression.
77
Imputation
Replacing missing data with calculated values.
78
PCA
-Is another commonly used feature extraction method. -It combines highly correlated variables together to form a smaller data set. -Projects the high-dimensional input data into lower-dimensional data by combining features. The data set thus gets smaller, with fewer important features.
79
Supervised Learning
-Takes the data as pairs (input, output) and learns a mapping function that could calculate the output, sometimes also called the response or target, from the input. -The model is trained on labeled data.
80
Unsupervised Learning
-Refers to the use of algorithms to identify patterns from unlabeled data. The learning model groups input data into clusters based on their common characteristics.
81
Reinforcement Learning
Learning through rewards and penalties.
82
What are the four types of analytics?
-Descriptive analytics -Diagnostic analytics -Predictive analytics -Prescriptive analytics
83
Descriptive Analytics
Analyzes historical data for reporting. -Usually helps answer questions, such as "What is happening?" or "What has happened?" It is useful in understanding more in-depth insights into specific queries with the data collected from the past.
84
Predictive Analytics
Estimates likelihood of future events. -It can be used any time you need to know something about the future; for example, "What will happen?" -Uses statistical models and forecasting techniques to identify trends, correlation, and causation.
85
Diagnostic Analytics
-Finds reasons for past events. -Can answer the question, "Why did this happen?" -Will be able to find the cause of an issue based on the collected descriptive data; for instance, what is the cause of death of many patients in an intensive care unit (ICU)?
86
Prescriptive Analytics
-Helps to identify the course of action to achieve something by leveraging optimization and simulation algorithms to advise on possible outcomes. -Answers the question, "What should we do?" For example, what is a more cost-effective way to manage patients at high risk for readmission: at the hospital or at the patient's home? - Recommends one or more possible courses of action.
87
Machine Learning:
A technology whereby computers are trained to make decisions or predictions based on given data. Computers learn patterns in the data to make informed decisions in the future
88
What are the three sections of machine learning?
-Supervised learning -Unsupervised learning -Reinforcement learning
89
What are the steps in the data analytical process?
1.) Data Collection 2.) Data Preprocessing 3.) Exploratory Data Analysis 4.) Feature Engineering 5.) Statistical Modeling Machine Learning 6.) Optimization 7.) Performance Evaluation 8.) Data Visualization Communication
90
Data Collection
Gathering relevant health-related data from various sources suchas EHR's and Clinical Trials
91
Data Preprocessing
Cleaning and transforming data for quality and usability.
92
Exploratory Data Analysis
Investigating dataset to uncover patterns, trends, and relationships
93
Feature Engineering
Creating new variables from existing data to enhance model performance.
94
Statistical Modeling
Applying statistical methods to derive insights and make predicitons
95
Optimization
Applied to improve model performance through tuning.
96
Performance Evaluation
Assessing model effectiveness using metrics such as: AUC-ROC
97
Data Visualization Communication
-Visualizing the results is crucial for making the findings understandable and actionable
98
AUC-ROC
Evaluates trade-off between sensitivity and specificity.

HI-410 (3 decks)

Brainscape helps you reach your goals faster, through stronger study habits.
© 2025 Bold Learning Solutions. Terms and Conditions