Digital biomarkers are objective, quantifiable measures of physiological or behavioural state collected by digital tools—wearables, smartphones, ambient sensors, or connected medical devices—and often processed with algorithms or AI. Examples include step count, heart rate variability, sleep structure, voice or gait analysis, and app-based cognitive or motor tasks. They can reflect disease risk, progression, or response to treatment and may be used for screening, monitoring, or as endpoints in clinical trials. Validation and regulatory acceptance are advancing: some digital biomarkers have received regulatory clearance or qualification for specific contexts of use.
The technology addresses the gap between episodic clinic visits and continuous, real-world assessment. For chronic diseases, digital biomarkers can enable remote monitoring and early warning; for drug development, they can provide dense, objective data in naturalistic settings. AI and machine learning are used to derive signatures from raw sensor or usage data, though interpretability and generalisability remain areas of attention. Privacy, equity of access, and alignment with clinical workflows are important considerations.
Challenges include standardisation of collection and analysis, evidence of clinical utility, and integration into care pathways and reimbursement. As evidence accumulates and regulators issue more guidance, digital biomarkers are likely to be increasingly used in prevention, diagnosis, and management, complementing traditional laboratory and imaging biomarkers.