Patient-powered data: a new engine for healthcare AI
Patient-powered data platforms in healthcare AI are systems that let individuals collect, control, and share their own health information so that anonymized, real-world patient data can be used to train medical algorithms, support drug development, and improve care while preserving privacy and consent. For years, healthcare AI training has leaned on clinical trials and insurance or hospital records, which are often narrow, fragmented, and missing context from everyday life. That limits how well algorithms perform on diverse patient populations and real-world conditions. By contrast, patient data healthcare AI platforms start with the person, not the institution: patients pull together their medical records, test results, and histories, then choose whether those data feed into research and model development. This shift is starting to change how companies design real-world data platforms and how quickly patient-powered research can move from insight to impact.
Inside Novellia’s USD 18 million (approx. RM82.8 million) patient data bet
Novellia is one of the clearest examples of this new model. The company has raised USD 18 million (approx. RM82.8 million) in Series A funding led by Spark Capital, bringing its total funding to USD 28 million (approx. RM128.8 million) as it scales a patient-powered, real-world data platform for life sciences companies. Rather than assembling datasets from scattered insurance claims and hospital feeds, Novellia helps individuals securely gather medical records from hospitals, clinics, and laboratories into a single, unified health history. According to Novellia, the current real-world data ecosystem costs more than USD 50 billion (approx. RM230.0 billion) a year yet still leaves researchers with incomplete pictures of patients’ lives and conditions. By securing multi-year, seven-figure contracts with major pharmaceutical and diagnostics firms, the startup is positioning patient-consented, real-world data as a core asset for healthcare AI training and drug development.
From fragmented records to real-world data platforms
Traditional real-world evidence in medicine has grown up around billing systems, claims warehouses, and siloed electronic records. The result is a patchwork view of each patient, with missing lab histories, disconnected specialist notes, and little sense of long-term outcomes. Novellia’s patient-powered research approach aims to reverse that. Its platform can help patients consolidate up to 20 years of medical data in about 30 seconds, turning decades of scattered records into a usable, longitudinal dataset. The company then uses natural language processing models to extract structure from unstructured text such as physician notes, pathology narratives, and diagnostic reports. That process converts messy charts into research-ready datasets that can be used to develop and validate healthcare AI systems. For pharmaceutical and diagnostics partners, this promises faster access to cleaner, more complete real-world data platforms built around the full arc of a patient’s care.
Training healthcare AI on diverse, real-world patient journeys
As healthcare AI training moves beyond controlled trial data, the value of real-world patient journeys grows. Algorithms that will be used in everyday clinics must learn from the full mix of ages, co-morbidities, treatment paths, and adherence patterns that define routine care. Patient data healthcare AI platforms offer exactly that: longitudinal histories that show what happens between specialist visits, which drugs are stopped or switched, and how lab values shift over time. For pharmaceutical companies, this detail can surface new patient subgroups and more realistic safety and effectiveness signals. For AI developers, deidentified, patient-consented datasets can support everything from predictive risk models to decision-support tools. As Novellia expands its AI-powered data processing, it is helping turn raw patient histories into structured training material that may lower bias, improve generalization, and better mirror how medicine plays out in real life.
Privacy, consent, and the future of patient-powered research
Patient-powered research can only scale if people trust how their information is handled. Novellia was founded to place patients at the center of medical research, giving each person tools to access, organize, and manage their own health histories. Individuals decide whether to contribute anonymized and deidentified data to studies, and the company frames this as creating a new category of patient-consented data that is more complete and more useful for research. According to TMV partner Emma Silverman, nearly 70% of patients are open to contributing data for medical breakthroughs, but they expect control, transparency, and privacy safeguards. Novellia’s new mobile application extends its web platform, making it easier for patients to review their records and adjust preferences on the go. If this privacy-first model succeeds, it could set a template for future real-world data platforms and help align healthcare AI innovation with patients’ expectations.
