Building an Enterprise Graph Database POC For a Major Healthcare Organization
About the Client
The client is an American pharmaceutical company. It manufactures prescription pharmaceuticals and biologics in several therapeutic areas, including cancer, HIV/AIDS, cardiovascular disease, diabetes, hepatitis, rheumatoid arthritis, and psychiatric disorders.
The client wanted a rapid development of a POC around experimental and analytical techniques to provide information on biological components (cell, tissue, disease, gene, protein, drug response, and pathway, etc.) and their functions.
Create a database and a solution that could help the client to capture, analyze, visualize, and predict heterogeneous data at scale.
Connect genomics, RWE, and other related information about the patients
Help with mapping and storing the complex relationships among heterogeneous data, as well as exploring new relationships and patterns
Brillio analyzed the need for a graph DB against a data lake and RDBMS, based on licensing, scalability, graph models, schema models, query method, platforms, consistency, and availability – taking into consideration everything from cloud, extensibility, to visualization and data support.
Compared leading knowledge graph databases AWS Neptune and Neo4j
Developed the proof of concept for leveraging statistical methods using EHR, disease & mutation data, and created a demo for understanding patient behavior and disease progression using graph DB
Connected the dots to provide a client implementable solution approach
Brillio has successfully delivered the requirements by bringing in the right talent and technology.
We managed to boost productivity by significantly improving data correlations and insight generation. Brillio enabled the client to uncovered new avenues by analyzing data sets that can provide information about potential patient risk factors.
Helped researchers investigate genomic information of patient’s cohorts to derive new knowledge
Enabled clinicians to explore, visualize, and use real-world data sources to generate insights and make data-driven decisions for patients
Helped patients identify risk factors before they turn into diseases, as well as find similar cases with alternatives for therapies