Items tagged “statistics”
43 results found
Article
Case report
Case reports are a type of radiology research literature. They belong to the class of descriptive studies.
Purpose
The purpose of a radiology case report is to describe the patient history, clinical course, and imaging for a notable or unusual case. The case may be intended to aid other practi...
Article
Case series
A case series is essentially a collection of case reports around a common theme. It belongs to the class of descriptive studies.
Structure
A case series typically contains:
a short introduction
the series of cases
patient history, presentation, imaging, and clinical course are described in ...
Article
Likelihood ratios
Likelihood ratios (LR) are an alternative to positive and negative predictive values for estimating the likelihood of disease after diagnostic testing. The general formula for a likelihood ratio is the probability (P) that someone with a disease will have a particular test result divided by the ...
Article
Multiple regression analysis
Multiple regression analysis is a commonly used statistical technique in radiology research. It allows the examination of the relationship between multiple variables in a quantifiable manner.
This technique is often used where there are multiple explanations (independent variables / co-variates...
Article
Pearson's chi-squared test
The Pearson's chi-squared test is one of the most common statistical tests found in radiology research. It is a type of non-parametric test, used with two categorical variables (not continuous variables).
Concept
The heart of the chi-squared test is a 2 x 2 contingency table.
We usually have ...
Article
Paired t-test
The paired t-test is the appropriate method when the researcher takes an experimental group, measures the baseline, subjects the members to an intervention, and then measures the results.
Testing in a before-and-after manner like this ("matched data" or "repeated measures") requires a different...
Article
Kappa
Kappa is a nonparametric test that can be used to measure interobserver agreement on imaging studies. Cohen's kappa compares two observers, or in the case of machine learning can be used to compare a specific algorithm's output versus labels. Fleiss' kappa assesses interobserver agreement betwee...
Article
Selection bias
Selection bias is a type of bias created when the data sampled is not representative of the data of the population or group that a study or model aims to make a prediction about. Selection bias is the result of systematic errors in data selection and collection. Practically-speaking selection bi...
Article
Confusion matrix
Confusion matrices, a key tool to evaluate machine learning algorithm performance in classification, are a statistical tool.
Contingency tables, a type of confusion matrix, are used in the evaluation of many diagnostic exams for sensitivity, specificity, positive and negative predictive values....
Article
Single linear regression
Single linear regression, also known as simple linear regression, in statistics, is a technique that maps a relationship between one independent and one dependent variable into a first-degree polynomial. Linear regression is the simplest example of curve fitting, a type of mathematical problem i...
Article
Retrospective study
Retrospective studies are cohort or case-control studies that analyze existing data from before the time point at which the study began. These studies are more susceptible to bias and confounding than prospective studies.
Article
Prospective study
Prospective studies are usually longitudinal or cohort studies, although they can be case control studies, that examine whether an outcome occurs during the time period of the study.
Article
Bayes' theorem
Bayes' theorem, also known as Bayes' rule or Bayes' law, is a theorem in statistics that describes the probability of one event or condition as it relates to another known event or condition. Mathematically, the theory can be expressed as follows: P(A|B) = (P(B|A) x P(A) )/P(B), where given that...
Article
Bayes' factor
A Bayes' factor is a number that quantifies the relative likelihood of two models or hypotheses to each other if made into a ratio e.g. if two models are equally likely based on the prior evidence ( or there is no prior evidence) then the Bayes factor would be one.
Such factors have several use...
Article
Incidence
The incidence is an epidemiological term for the number of new cases of a condition (e.g. brain tumors or diabetes), in a given population, during a specified time interval. The formulas for incidence can express it as a rate or a proportion, e.g. x new cases / y population count / z time period...
Article
Prevalence
Prevalence is an epidemiological term referring to the proportion that reflects the total disease/condition burden in a population at a specific time and should not be confused with incidence.
There are different kinds of prevalence:
point prevalence is the amount of disease at one point in t...
Article
Impact factor
Impact factor is a bibliometric index. It expresses the "impact" of a publication on the reference scientific community. Specifically, it measures the average number of citations of a scientific article by other researchers.
Development of the impact factor
The impact factor was invented by an...
Article
Probability vs odds
The likelihood of an occurrence (called an “event”) can be expressed as a probability or as odds.
Probability (P) tells us how often a particular event occurs on average over the course of many trials. For example, the probability of rolling a 4 with a fair 6-sided die is 1/6. Think of probabil...
Article
Diagnostic test accuracy
The accuracy of a diagnostic test is defined as how often the test correctly classifies someone as having or not having the disease. The formula for accuracy is:
(true positive + true negative) / (true positive + true negative + false positive + false negative)
or correct results / all results...
Article
Alpha vs beta error
Clinical trials may have incorrect results due to random error or bias. The 2 types of random error are called alpha (α) and beta (β), also known as type I and type II errors respectively. Alpha and beta errors are both conditional probabilities. For Radiologists, it may be helpful to think of α...