Big Data and machine learning in radiation oncology: State of the art and future prospects

- Computing power and data storage costs are continuously decreasing.
- Electronic Health Records can now be used to create comprehensive phenotypic profiles.
- Genomics can be correlated to these phenotypic profiles to better understand treatment response and toxicity.
- Combining EHR and Genomics through Machine Learning could generate high-quality evidence for precision medicine.
- These methods could be used to create a “learning health system” to predict the outcome of any treatment.

Precision medicine relies on an increasing amount of heterogeneous data. Advances in radiation oncology, through the use of CT Scan, dosimetry and imaging performed before each fraction, have generated a considerable flow of data that needs to be integrated. In the same time, Electronic Health Records now provide phenotypic profiles of large cohorts of patients that could be correlated to this information. In this review, we describe methods that could be used to create integrative predictive models in radiation oncology. Potential uses of machine learning methods such as support vector machine, artificial neural networks, and deep learning are also discussed.