loader
Pezoulas, Vasileios C., et al. “A computational pipeline for data augmentation towards the improvement of disease classification and risk stratification models: A case study in two clinical domains.” Computers in Biology and Medicine 134 (2021): 104520.
 

Virtual population/Synthetic data generation

The PRECIOUS platform provides an advanced synthetic data generation and data augmentation service which aims: (i) to increase the population size for pharma research, where there is a reported lack for clinical trials, (ii) to increase the predictive power of the existing AI models for disease prediction and prevention. To this end, we focus on the use of machine learning algorithms towards the generation of high-quality synthetic data which can “mimic” the real data. The synthetic data are used: (i) for imputing real patient data, where the real missing data of a patient are replaced with the values of its optimal virtual patient profile (i.e., the virtual patient with the highest similarity), and (ii) to augment the real data and use them for ML training and validation.