In the context of radiological images, synthetic and augmented data are data that are not completely generated by direct measurement from patients.
Machine learning by definition improves with increased data, however, there is a relative lack of open, free available radiology data sets. Issues of patient privacy and legal restrictions on data use make machine learning without synthetic and augmented data challenging. Additionally, some diseases are so rare that even large data sets do not contain sufficient samples to generate robust machine learning algorithms.
In addition to contributing to making algorithms for image identification and classification, synthetic data can also be used in algorithms for artifact correction.
- synthetic data: partly or completely artificial. Synthetic data are often in adversarial network artificial intelligence (AI) algorithms
- augmented data: derived from real images with some sort of transformation such as whole image enlargement, translation, flipping, rotation, or the addition of noise
Criticisms of the use of synthetic and augmented data in radiology AI include the potential amplification of statistical biases and a general lack of research on their consequences. Nonetheless, synthetic and augmented are used routinely in the generation of AI applications for radiology.
- 1. Kanghyun Ryu, Yoonho Nam, Sung‐Min Gho, Jinhee Jang, Ho‐Joon Lee, Jihoon Cha, Hye Jin Baek, Jiyong Park, Dong‐Hyun Kim. Data‐driven synthetic MRI FLAIR artifact correction via deep neural network. (2019) Journal of Magnetic Resonance Imaging. doi:10.1002/jmri.26712 - Pubmed
- 2. Goodfellow I, Pouget-Abadie J, Mirza M, et al (2014) Generative Adversarial Nets. In: 1124Ghahramani Z, Welling M, Cortes C, et al (eds) Advances in Neural Information Processing 1125Systems 27. Curran Associates, Inc., pp 2672–2680.
Related Radiopaedia articles
- artificial intelligence (AI)
- imaging data sets
- computer-aided diagnosis (CAD)
- natural language processing
machine learning (overview)
- machine learning processes
- machine learning models
- visualizing and understanding neural networks
- common data preparation/preprocessing steps
- DICOM to bitmap conversion
- dimensionality reduction
- principal component analysis
- training, testing and validation datasets
- loss function
- optimization algorithms
- linear and quadratic
- batch normalization
- rule-based expert systems