Pangaea Data Limited provides a machine learning based software product which helps its end users, who are scientists from the Biopharmaceutical domain to find patients based on specific phenotypes from electronic health records (EHRs), which is critical for early diagnosis, drug discovery and clinical trials. The product is shown to be 50x faster and 30% more accurate compared to current approaches involving natural language processing (NLP), indexing and semantic searches. For example, in one instance, the product was applied to a dataset of 52,000 patient EHRs and found 5,000 new patients in less than 6 weeks around 200 disease phenotypes, which were missed by searches using ICD codes or technical approaches such as NLP which were applied to the same dataset over the course of 18 months.
Pangaea Data was founded by serial entrepreneurs, who are also highly renowned scientists from Imperial College London and is supported by globally recognised experts from Stanford University. Pangaea works closely with senior executives and decision makers from the life sciences industry and has commercial partnerships with McKinsey and some of the largest contract research organisations in the world.