Human Digital Twins (HDTs) can be a practical route to continuous, personalised health management outside traditional care settings. By linking real-world data from wearables, smartphones and ambient sensors to computational representations of individuals, HDTs aim to monitor health states, anticipate risk and support timely interventions in daily life. A scoping review of implemented HDT systems identified how pervasive sensing is being assembled with modelling and feedback to move from one-off snapshots toward longitudinal, context-aware care. The evidence maps which entities are twinned, how data are captured and transformed and what actions are taken, while drawing out the technical and ethical hurdles that must be addressed for reliable deployment.
Implementation Landscape and Use Cases
Implemented HDT systems cluster around five purposes: continuous health monitoring, disease tracking and diagnosis, rehabilitation support, treatment planning and optimisation and simulation or prediction. Across these aims, the twinned “entity” ranges from disease-specific patients and organs to general adult cohorts, older adults and professional roles such as nurses, caregivers and therapists. Clinical conditions frequently modelled include cancer and cardiovascular disease, alongside musculoskeletal function and metabolic control.
Must Read: Building Immune Digital Twins for Precision Care
HDTs are deployed across varied environments. Hospital settings feature intensive care, operating rooms and inpatient beds for high-acuity monitoring and decision support. Home-based deployments support longer-term management of chronic conditions and ageing-related needs. Rehabilitation spaces use HDTs to guide exercises, assistive robotics and gait correction. Additional contexts include fitness facilities, controlled lab simulations and virtual platforms such as web dialogue systems or metaverse-based services. This breadth reflects a shift from strictly organ-centred, encounter-based views of health towards models that incorporate behaviour and surroundings across everyday life.
Sensing, Mapping and Acting Across Settings
A common workflow underpins implemented systems: sensing, mapping and acting. Sensing spans physiological, behavioural and environmental inputs. Wearable sensors are most commonly used to capture vital signs, movement and activity. Camera-based imaging extracts body pose and motion, while microphones and voice capture inform stress or emotion in mental health contexts. Smartphones and environmental sensors add location, routines and ambient conditions. Some systems integrate clinical or laboratory data, including medical records and imaging, as indirect sensing channels.
Mapping translates raw inputs into low- and high-level representations. Low-level features parameterise signals such as waveforms, landmarks or oxygen levels. High-level features abstract to health concepts including disease severity, gait sequences, medication behaviour or diagnostic scores. Acting then closes the loop through three routes. Simulation forecasts risks or treatment outcomes. Visualisation presents internal states and model outputs for human interpretation. Intervention delivers real-time prompts, alerts or adjustments to support behaviour change and clinical response. These routes can operate independently, yet many systems combine them to deliver Just-in-Time Adaptive Interventions (JITAI), deciding when to help, what to recommend and how to adapt as context evolves.
The review also highlights a directional change. Earlier work relied on discrete clinical snapshots for diagnosis and risk estimation. Increasingly, HDTs integrate Patient-Generated Health Data (PGHD) from wearables, mobile sensing and ambient devices to enable preventive, longitudinal management in naturalistic settings. This expansion extends digital twinning beyond organs and tissues to encompass behaviour and environment, broadening the scope of personalisation.
Challenges and Directions for Real-World Deployment
Real-world adoption hinges on resolving several constraints. Privacy and security are central, as HDTs process sensitive, granular personal data that could be misused or inferred beyond intended scope. Robust safeguards across collection, storage and processing, with clear information for users and appropriate legal and institutional frameworks, are required. Data quality is equally pivotal. Unreliable signals, missingness or poorly validated measures undermine trust and utility, motivating methods to monitor, curate and assure data streams.
Wearability and long-term usability remain practical barriers, particularly for older adults or those with chronic conditions. Discomfort, inconvenience or stigma reduce adherence and continuity, calling for unobtrusive, user-friendly sensing and options like on-device processing and user-controlled sharing to build confidence. Technical integration presents further challenges. Heterogeneous data from multiple sensors and records demand harmonisation, standardised schemas and interoperable interfaces. Governance questions persist around where integrated data reside, ownership, stewardship and the pathway for HDTs to fit within existing clinical systems and workflows.
Future work should broaden disease coverage and underrepresented populations, including mental health and behavioural targets. Sensing should expand to ambient and interactional modalities that reduce burden, while mapping can advance through multimodal fusion with explainable approaches to sustain clinical transparency. Finally, longitudinal field evaluations across diverse settings are needed to test effectiveness and clarify the roles of patients, caregivers and clinicians in routine use.
HDTs coupled with pervasive sensing are progressing from concept to implementation across hospitals, homes and rehabilitation contexts. By structuring systems around sensing, mapping and acting and by incorporating PGHD for continuous context, they enable monitoring, personalised interpretation and timely intervention in everyday life. Delivering this potential at scale will depend on trustworthy data practices, interoperable integration and careful workflow design. Current implementation patterns indicate where HDTs are already adding value and where targeted investment, governance and evaluation can accelerate safe, effective adoption.
Source: Health Informatics Journal
Image Credit: iStock
