Data Analytics

Healthcare Analytics

teh systematic use of data and statically methods to analyze and interpret health-related information

Descriptive analytics

provides insight into what has happened, such as tracking patient outcomes or identifying trends in healthcare utilization

Predictive analytics

Uses historical data to predict future outcome, such as forecasting disease outbreaks or patient readmissions

Prescriptive analytics

recommends options to optimize healthcare practices, such as suggesting personalized treatment plans based on patient data

Unit of Analysis

refers to the primary entity or object of study or measurement

Simple random sampling

every indivusal or unit in the population has an equal chance of being selected

stratified sampling

the population is divided into distinct subgroups or strata (e.g by age, gender, outcome), and random samples are taken from each group

Systematic Sampling

every “kth” element in the population is selected, starting from a randomly selected point

Cluster Sampling

the population is divided into clusters (eg. geographical areas, schools), and entire lusters are randomly selected for sampling

Multistage sampling

like a cluster sample, but rather than keeping all observations in each cluster, we collect a random sample within each selected cluster

electronic health records (EHRs)

can contain all Medical history (medical image, prescription)

biomedical image analysis

magnetic resonance image (MRI), computed tomography (CT), positron emission tomography (PET), and ultrasound

Sensor Data Analysis

Electrocardiogram (ECG or EKG)

quantitative

quantity, something that you can measure, usually describes data sets or study’s

data overload

(Overwhelming amount of data) when collecting data sometimes more data doesnt matter

discrete numbers

whole numbers

continuous numbers

Decimal (any value between two numbers )

Noise/ signal-to-noise ratio

Extra data info to identify, less noise = better analysis

causality

Relationship between cause and effect, relationships between variable

mining

Mining the data to find a signal data point

pervasive healthcare

Care about a large group of people for a long time, some in livement of digital minoring

Destructive analytics

Describe a data set, describing factors, describing the situation

predictive analytics

gather historical data to predict what will happen in the future

Prescribitive analytics

Prescribing a solution based on understanding the cause, reaction and outcome of the data

Nominal

Can be categorized but not ranked or ordered

ordinal

can be categorized and ranked and ordered

R (correlation coefficient)

Measures the strength and direction of linear relationships between two variables. It’s unitless and always between -1 and 1

Slope

Tells you how much Y changes for a one - unit change in X. Depends on the units of the variables involved

p-value

Tell how likely it is to see results just by chance of the two variables were actually unrelated. A low number (less than 0.05) means there’s a good change the variables are related in the population

chi square

a statistical test used to determine where there’s a significant association between two categorical variables

regression

Used to predict the value of one variable based on another and can imply a directional influence, making it useful for forecasting and modeling relationships

Dependent variable (y)

This is the outcome or the variable e want to predict or explain

independent variable (x)

this is the predictor variable. Or the variable we use to make prediction about the dependent variable

Direct and linear

As one variable increases, the other increases at a constant rate, forming a straight upward-sloping line.

direct and nonlinear

As one variable increases, the other also increases, but not at a constant rate, forming a curved upward trend.

indeirect and linear

As one variable increases, the other decreases at a constant rate, forming a straight downward-sloping line.

Indirect and nonlinear

As one variable increases, the other decreases, but not at a constant rate, forming a curved downward trend.

no relationship

Changes in one variable show no consistent pattern with changes in the other.

Correlation

measures the strength and direction of the relationship between two variables. Tells us is two varibles are related and how strongly but doesnt imply that one causes the other

Symmetric Distribution

If the histogram is symmetrical (e.g., bell-shaped normal distribution), the teeter-totter is balanced because the weight (data values) is evenly distributed on both sides of the mean.

right - skewed distribution (positive )

If the histogram has a long right tail (e.g., income data where a few people earn much more), the teeter-totter tilts right because the extreme high values pull the mean in that direction.

Left-Skewed Distribution (Negatively Skewed) → Tilted Left

If the histogram has a long left tail (e.g., test scores where most students score high but a few score very low), the teeter-totter tilts left because the extreme low values pull the mean downward.

Bimodal Distribution → Two Peaks, Unstable Teeter-Totter

If the histogram has two peaks (e.g., heights of adults where men and women form separate peaks), the teeter-totter might wobble because the data is clustered in two separate areas, making it hard to find a single point of balance.

Boxplots

Summarize data using a five-number summary (minimum, first quartile, median, third quartile, and maximum) and highlight potential outliers.

Histogram

Show the distribution of numerical data by grouping values into bins and displaying the frequency of occurrences.