Solution
BrainChild is building a national infrastructure that allows pediatric cancer data to be collected, structured, and shared in a secure and interoperable way. Our platform will enable researchers and clinicians to gain meaningful insights from large-scale datasets that were previously fragmented, inaccessible, or underutilized.
What we are building
A technical foundation for better research and smarter care
The BrainChild platform is being designed to support long-term collaboration across healthcare, research, and industry. Its core function is to make high-quality data usable across institutions, systems, and study types. This includes clinical records, genomic sequences, imaging data, lab results, and treatment outcomes — all securely linked and structured according to international standards.
Our infrastructure is aimed to be modular and scalable. It will allow participating centers to contribute and access data in ways that respect legal, ethical, and practical boundaries. Where appropriate, the platform will also support cross-border collaboration and AI-enhanced analytics.
How it works
From raw data to reusable insight
The platform will include components for secure data transfer, automated data validation, harmonization, and analytics. Data is cleaned and standardized upon entry using rule-based tools and AI-driven models. It is then structured into a research-ready format that can support multiple use cases — from exploratory studies to decision support systems about how to construct clinical cohorts.
Where institutions use different technologies or formats, BrainChild aims to adapt through flexible pipelines, including support for multiple transfer methods (e.g. S3, APIs) and visualization frameworks.
Leveraging the strength of existing infrastructure
BrainChild is not starting from scratch — it builds on the exceptional work already done by key national infrastructures. Initiatives such as the Swedish Childhood Tumor Biobank (Barntumörbanken), Genomic Medicine Sweden (GMS), and the Swedish Childhood Cancer Quality Registry (SBCR) have laid critical groundwork in data collection, analysis, and ethical governance.
Barntumörbanken provides a comprehensive, ethically managed collection of tissue and blood samples from children with solid tumors, including whole genome and transcriptome sequencing (WGS/WTS), diagnostic information, and imaging data. It is a vital national resource enabling molecular-level research on pediatric cancers, and collaborates closely with SciLifeLab and the National Genomics Infrastructure.
SBCR serves as a central tool for monitoring the quality of pediatric cancer care in Sweden. With data collected since the 1970s and contributions from all six childhood cancer centers, the registry offers detailed insights into diagnosis, treatment, outcomes, and late effects. It is also a key enabler of clinical research.
By connecting and building upon these established structures, BrainChild aims to enhance interoperability, enable new discoveries, and amplify the collective impact of Sweden’s existing pediatric cancer ecosystem.
Proof-of-concept (POC)
A working prototype across all Swedish childhood cancer centers
In 2024, BrainChild completed a proof-of-concept (POC) in collaboration with the GMS childhood cancer study and the data from 500 pediatric cancer patients across all six clinical centers in Sweden. This demonstration project showed that it is technically feasible to create a shared infrastructure that enables researchers to interact with sensitive clinical and genomic data securely and efficiently.
The POC includes:
- A Microsoft Fabric–based data backbone.
- A rule-based validation engine for ensuring data quality.
- An OpenAI-powered tool for standardizing unstructured clinical text.
- A user interface that allows researchers to explore trends, test queries, and visualize correlations.
The interactive dashboard showcases real-world use of the platform — including basic clinical parameters, data on precision diagnostics and treatment, mutation status, gene fusion tracking, copy number variation exploration, and clinical-lab data correlation.
What’s next
From pilot to platform
In 2025, BrainChild will begin scaling beyond the proof-of-concept. The first stage of expansion will focus on integrating additional data types — such as digital pathology, biobank samples, treatments given and survival trajectories — while expanding participation among clinical and academic partners.
Work is also ongoing to establish long-term governance, hosting, and security frameworks, as well as to align with national and international data-sharing protocols.