Imputation
Imputation refers to statistical methods for creating data when it is missing from a data set. Missing data is often not random (and can therefore lead to different forms of bias). Imputation theoretically improves research outcomes as opposed to simply discarding incomplete data subsets. Several methods have traditionally been used to impute single missing values which included linear regression i.e. replacing the missing value with a value that would fit a linear regression line and substituting the missing value with the mean of the entire data set, but various AI models can be used as well. Multiple imputation methodologies are increasingly reported in academic medical research generally 1 and have been utilized in radiology research specifically 3. In AI models, imputed data could be considered a subset of synthetic data.
Related Radiopaedia articles
Artificial intelligence
- artificial intelligence (AI)
- imaging data sets
- computer-aided diagnosis (CAD)
- natural language processing
- machine learning (overview)
- visualizing and understanding neural networks
- common data preparation/preprocessing steps
- DICOM to bitmap conversion
- dimensionality reduction
- scaling
- centering
- normalization
- principal component analysis
- training, testing and validation datasets
- augmentation
- loss function
-
optimization algorithms
- ADAM
- momentum (Nesterov)
- stochastic gradient descent
- mini-batch gradient descent
-
regularisation
- linear and quadratic
- batch normalization
- ensembling
- rule-based expert systems
- glossary
- activation function
- anomaly detection
- automation bias
- backpropagation
- batch size
- cost function
- confusion matrix
- convolution
- curse of dimensionality
- epoch
- gradient descent
- image registration
- imputation
- iteration
- linear algebra
- noise reduction
- R (Programming language)
- Python (Programming language)
- segmentation
- synthetic and augmented data
- overfitting
- transfer learning
Unable to process the form. Check for errors and try again.