ESTRO 2022 - Abstract Book

S897

Abstract book

ESTRO 2022

Conclusion Our study was the first research work dedicated to assess access to Radiation therapy in Lebanon. It showed a lack of access compared to the international average. We succeeded to pinpoint the main obstacles to access. Since the major barrier to optimal treatment is financial, the access will be severely threatened secondary to the recent economic crisis. Short term solutions through financial support is an urgent need to maintain sustainability and quality of care.

PO-1062 Privacy-preserving dashboard for clinical data using open-source federated learning infrastructure

V. Gouthamchand 1 , G. K 2 , R. Subramanian 2 , A. Choudhury 1 , L. Wee 1 , A. Dekker 1 , S. Sinha 3 , S. Ghosh Laskar 3 , L. Reddy 4

1 University of Maastricht, Faculty of Health, Medicine and Life Sciences, Maastricht, The Netherlands; 2 Centre for Development of Advanced Computing (C-DAC), IT, Pune, India; 3 Tata Memorial Hospital, Oncology, Mumbai, India; 4 HealthCare Global Enterprises Limited, Oncology, Bangalore, India Purpose or Objective Research on real-world cancer data requires vast volumes of geographically distributed multi-institutional data to be analyzed. It is not always feasible to centralize the data for statistical analysis. “Distributed methods” are now making inroads into oncology research, however confidentiality of individual human subjects remains the highest priority. In keeping with Findable-Accessible-Interoperable-Reusable (FAIR) data management principles, data schema and idiosyncratic coding should be publishable as metadata, even though the data content itself is kept private. Additional privacy is made possible by only exchanging statistical summaries, but never any individual-level data. In this way, privacy is maintained without relying exclusively on encryption or privacy-obscuring mechanisms (such as differential privacy). This is known as privacy by design, since re-identification of individual subjects from statistical summaries is a very low risk, even where encryption and differential mechanisms are breached. Materials and Methods We demonstrated a semi-automated method for FAIR-ification of structured clinical data in four simulated geographically- separated “data nodes”, each with one of four open-access HNSCC datasets. The datasets were independent and had distinct schema with coding dictionaries. Individual patient-level data were processed as Resource Descriptor Format (RDF). The anonymous schema were separately extracted as an Ontology Web Language (OWL) resource. Each OWL resource was shared, then individually mapped to a common oncology ontology using an extra annotations layer above the raw data. In the same manner, three private HNSCC datasets - from two Indian centers and one Dutch clinic - were processed in the same manner. An interactive dashboard was created in Python to exploit the Vantage6 distributed learning infrastructure. Distributed algorithms for descriptive statistics of cohorts were transmitted through the Vantage6 infrastructure. Results The results from the algorithms provided the summary statistics of the seven datasets, without combining any individual patient-level data into a single repository (see Fig. 1). The interactive display allows manual exploration of the data at deeper levels by the researcher. The prototype visualization is interactive and can be easily adapted to other use cases.