Novellia Secures $18M to Scale NLP-Driven Patient Data Synthesis

Novellia, a New York-based health technology firm, has closed an $18 million Series A funding round to expand its platform for aggregating and normalizing fragmented patient health data. The investment, led by Spark Capital with participation from Khosla Ventures, Acrew Capital, Bling Capital, and TMV, brings the company’s total capital raised to $28 million as of June 17, 2026.

The current pharmaceutical research sector relies on a $50 billion real-world data market dominated by claims brokers and hospital record aggregators. These existing infrastructures often fail to capture the nuance of patient health histories because they rely on fragmented, incomplete datasets. Novellia addresses this by providing a patient-facing interface that unifies up to 20 years of clinical records from more than 50,000 disparate healthcare systems. The platform uses proprietary natural language processing models to parse unstructured clinical text, including physician notes, lab narratives, and diagnostic reports. By extracting signals from these high-fidelity sources, the system generates a longitudinal, continuously updated dataset that exceeds the granularity of traditional claims-based records. This approach allows biopharma researchers to perform more precise analyses across oncology, rare disease, and cardiometabolic therapeutic areas. The company has already secured seven-figure, multi-year contracts with several of the top 10 global pharmaceutical firms. Shashi Shankar, cofounder and CEO of Novellia, emphasizes that the platform operates on a model where patient consent is the primary driver of data availability. The system ensures that all data utilized for research purposes is de-identified and anonymized, maintaining strict compliance while providing researchers with high-quality evidence. As more patients contribute their complete medical histories to the platform, the resulting dataset grows in both depth and statistical power, allowing researchers to observe longitudinal trends that are invisible in smaller, isolated samples. This iterative process of data collection and refinement ensures that the intelligence generated by the platform remains highly relevant to the changing regulatory and research requirements of clinical development teams. By continuously integrating new patient records, the system improves its ability to map complex disease progressions across diverse patient populations. This accumulation of high-fidelity longitudinal data directly addresses the industry’s historical reliance on static, point-in-time snapshots of health information.

The technical challenge of health data interoperability remains a significant bottleneck in clinical development, as most electronic health record systems operate in silos. Novellia’s architecture functions by normalizing these heterogeneous data streams into a structured format suitable for machine learning applications. By prioritizing the extraction of information from unstructured clinical narratives, the model captures clinical context that is frequently lost in standard structured data exports. This capability is critical for researchers who require a comprehensive longitudinal view of patient outcomes to validate drug efficacy and safety profiles. The shift toward patient-centric data collection represents a departure from the industry standard of purchasing third-party claims data. By placing the patient at the center of the data lifecycle, Novellia mitigates the fragmentation issues inherent in the current clinical data infrastructure. This strategic focus on high-quality, patient-authorized data provides a competitive advantage in an industry where data integrity is the primary determinant of research success. The reliance on direct patient input ensures that the data reflects actual clinical experiences rather than the administrative proxies often found in insurance claims databases. This methodology allows for a more accurate representation of patient health trajectories, which is essential for developing targeted therapies and understanding real-world treatment outcomes. Researchers can now access a more granular view of how specific interventions impact patient health over extended periods, providing a clearer picture of long-term therapeutic performance.

The company plans to deploy the new capital to scale its platform infrastructure and deepen its existing life sciences partnerships. Future milestones will focus on increasing the volume of patient participants and refining the underlying NLP models to handle increasingly complex diagnostic reports. The long-term viability of this model depends on maintaining the balance between patient engagement and the rigorous data requirements of pharmaceutical clinical teams.