EuroCAT
EuroCAT – Euregional Computer Assisted Theragnostics: Distributed Learning for Individualized Medicine
We aim to greatly improve Cancer treatment and research by using a new internationally advanced computer network for clinical research and decision supporting software.
Starting September 2010 nine organisations are collaborating on the ambitious Euregional Computer Assisted Theragnostics project (EuroCAT) in the Meuse-Rhine region (The Netherlands, Belgium and Germany). As a result physicians will be able to dramatically improve the quality of treatment each individual cancer patient receives as well as the speed and quality of clinical research. This is possible by using software that skilfully mines the data and treatment results in a new international patient database and predicts treatment outcomes for each new patient. This international advanced computer network will rapidly identify patients for clinical trials and automate many of the standard processes.
The goal of the project is two fold: a) to develop a shared database of medical characteristics in cancer patients, tumours and treatments. Copying data from existing databases and linking them together on a larger scale will greatly improve the ability to learn and predict the outcome of individual treatments within the next three years. And b) find patients for trials and decrease and speed up the administration and analysis around clinical trials. The project is made possible through extensive, cross-border cooperation among the parties involved.
The euroCAT consortium and project ran from 2010 to 2015. It is in many aspects the seed project of the Personal Health Train as it was the first project (2010) to embark on distributed learning FAIR-avant-la-lettre data.
The project was made possible through extensive, cross-border cooperation among the parties involved in addition to a sizeable European Interreg grant from Interreg IV-a.Interreg.
The Project Lead
Partners
- CHU, Liege, Belgium
- MAASTRO, Maastricht, The Netherlands
- Klinikum, Aachen, Germany
- Catharina Ziekenhuis, Eindhoven, The Netherlands
- LOC, Hasselt, Belgium
- Varian Medical Systems, Palo Alto, USA
Publications
Infrastructure and distributed learning methodology for privacy-preserving multi-centric rapid learning health care: euroCAT
We developed and implemented an IT infrastructure in five radiation clinics across three countries (Belgium, Germany, and The Netherlands). We present here a proof-of-principle for future ‘big data’ infrastructures and distributed learning studies. Lung cancer patient data was collected in all five locations and stored in local databases. Exemplary support vector machine (SVM) models were learned using the Alternating Direction Method of Multipliers (ADMM) from the distributed databases to predict post-radiotherapy dyspnea grade
. The discriminative performance was assessed by the area under the curve (AUC) in a five-fold cross-validation (learning on four sites and validating on the fifth). The performance of the distributed learning algorithm was compared to centralized learning where datasets of all institutes are jointly analyzed.
The euroCAT infrastructure has been successfully implemented in five radiation clinics across three countries. SVM models can be learned on data distributed over all five clinics. Furthermore, the infrastructure provides a general framework to execute learning algorithms on distributed data. The ongoing expansion of the euroCAT network will facilitate machine learning in radiation oncology. The resulting access to larger datasets with sufficient variation will pave the way for generalizable prediction models and personalized medicine.