Despite processing 15 million clinical charts, Datycs doesn't use this data for model training. Their agreements explicitly respect that data belongs to the patient and the client—an ethical choice that prevents them from building large, aggregated language models from customer data.
Datycs proactively converted unstructured data into FHIR resources long before clients were ready to use them. This future-proofed their platform, positioning them ahead of the curve when interoperability regulations finally mandated such standards, eliminating the need for custom APIs.
The CEO of Datycs applies lessons from standardizing wireless networks in the 90s to today's healthcare challenges. He compares siloed EHRs to old proprietary cell towers, highlighting how open standards like FHIR can solve a problem that the telecom industry conquered decades ago.
Despite the hype, Datycs' CEO finds that even fine-tuned healthcare LLMs struggle with the real-world complexity and messiness of clinical notes. This reality check highlights the ongoing need for specialized NLP and domain-specific tools to achieve accuracy in healthcare.
CEO Srini Rawl explains that while many companies focused on structured healthcare data, Datycs targeted complex, unstructured documents. This challenging niche became their competitive advantage, creating a significant data and experience moat after processing over 15 million clinical charts.
Datycs' initial product, a patient chart summarizer for physicians, faced slow adoption from health systems. The company found a more viable business model by pivoting to solve an urgent problem for payers: processing massive volumes of unstructured documents for back-office operations.
Faced with endless potential use cases, Datycs' CEO reveals their prioritization strategy: they wait for a new feature request, such as for social determinants of health, to mature and be echoed by two or three other customers before investing significant resources in building it.