ECC use case summary

The University of St Andrews are developing a dashboard to help oncologists observe, monitor, and analyse the condition of patients over time. It can also be used to analyse the effect of different chemotherapy treatments when given to patients with similar characteristics, and consequently influence future decisions to improve the well-being and survival rate of patients. The ultimate aim is to build a toxicity predictor to predict the toxicity of chemotherapy treatments based on history and feedback from patients.  Data was extracted for training the machine learning models from three main databases (Chemocare, Trak, and OncologyDB) within the Edinburgh Cancer Centre (ECC). The data contained information on treatment cycles, recorded side effects (in this case, toxicity level), comorbidities (other conditions affecting the patient), and various observations concerning breast cancer patients for three years (from 2014 to 2016).

Derived rules from the data are added to a platform created to generate synthetic data, thus improving the predictive ability of the system with regards to potential patient outcome. IBMs Data Fabrication Platform (DFP) is a web based central platform, providing a consistent and organisational-wide methodology for generating high-quality data for testing, development and training. Fabrication of synthetic data consists of two stages – data modelling and data generation. The platform can generate data from scratch, inflate existing databases, move existing data, and transform data from previously existing resources, such as old test databases or production data. The platform provides a comprehensive and hybrid solution that can create a mixture of synthetic and real data according to user requirements.