TRAIN

TRAIN: Personal health Train for RAdiation oncology in India and the Netherlands

This project unites academic, industrial and clinical leaders to improve care for cancer patients via Big Data science.
TRAIN will focus on head & neck cancer, which is a societal challenge both in India and The Netherlands due to its disappointing outcomes, practice variations, high incidence and the level resources needed for diagnosis and treatment. TRAIN will address this challenge by the introduction of Decision Support Systems (DSSs) that can predict which treatment leads to an optimal outcome given the individual patient characteristics and local diagnostic and treatment capabilities.


These DSSs will be based on outcome prediction models, which will be learned from Big Data by machine learning algorithms in India and The Netherlands. TRAIN will do this, via a unique and innovative approach, called the Personal Health Train (PHT), which enables machine learning from cancer Big Data without the need for data to leave the individual hospital.
The PHT has important advantages, in terms of privacy and data control concerns, over other Big Data approaches, but requires data to become Findable Accessible Interoperable and Reusable (FAIR) in each participating hospital. Making data FAIR is the major challenge in TRAIN but feasible due to the existing collaboration between partners and prior work in this area.


Finally, TRAIN will perform a prospective clinical trial using the DSSs to show clinical benefit and provide clear path to clinical introduction of Decision Support Systems.

This is a grant from the Dutch Research Council (NWO) in 2019; Indo-Dutch grant (629.002.212).

The Project Lead

The Team

Partners

  • Maastricht University, Maastricht, The Netherlands
  • Centre for Development of Advanced Computing, Maharashtra, India
  • HealthcareGlobal Enterprises limited, Karnataka, India
  • Tata Memorial Centre, Maharashtra, India
  • Philips India Ltd., West Bengal, India
  • Maastro Innovations, Maastricht, The Netherlands

Publications

FAIR-IFICATION OF STRUCTURED CLINICAL DATA

One of the common concerns in clinical research is improving the infrastructure to facilitate the reuse of clinical data and deal with interoperability issues. FAIR (Findable, Accessible, Interoperable and Reusable) Data Principles enables reuse of data by providing us with descriptive metadata, explaining what the data represents and where the data can be found. In addition to aiding scholars, FAIR guidelines also help in enhancing the machine-readability of data, making it easier for machine algorithms to find and utilize the data.