Development and Clinical Application of Digital Biomarkers for Sarcopenia Authors: Hyunbin Kim, Hyeon Su Kim, Bonsang Gu, Shivam Bhola, Jung-Gil Kim, Jun-Il Yoo Published: 2026 ABSTRACT AbstractSarcopenia, the age-related decline in skeletal muscle mass and function, has become a recognized disease entity with serious health implications. However, traditional diagnostic methods relying on handgrip strength, walking speed, and dual energy X-ray absorptiometry-based muscle mass assessment are limited by subjective components, irregular measurement intervals, and inability to capture muscle quality. This paper comprehensively reviews how digital biomarker technologies including smart watch, smart insoles, surface electromyography, and markerless motion analysis via OpenPose enable objective, continuous, real-world monitoring of sarcopenia. These technologies provide multidimensional insights into muscle strength, physical performance, daily activity patterns, coordination, and fatigue that single-modality assessments cannot achieve. Integration of medical imaging with artificial intelligence-based automatic segmentation models and machine learning-driven multimodal sensor fusion enables construction of personalized digital twin models for surgical planning and rehabilitation optimization. As the Asian Working Group for Sarcopenia 2025 expands the focus to include a life-course muscle health framework, there is growing clinical interest in digital biomarkers for their potential to detect early functional changes in real-life settings. While the integration of digital biomarkers as supplementary indicators for sarcopenia is a promising frontier, their formal inclusion in diagnostic criteria will require further standardization of protocols and large-scale clinical validation. While significant challenges remain including lack of standardized protocols, sensor interoperability issues, elderly patient usability concerns, and data security considerations, digital biomarkers hold the potential to transform sarcopenia management from episodic clinical assessment toward continuous, proactive, data-driven personalized care, provided that clinical validation is further established.Keywords: Biomarkers, Electromyography, Sarcopenia, Wearable electronic devices GRAPHICAL ABSTRACT INTRODUCTION Sarcopenia is the age-related decline in skeletal muscle mass and function, and is now recognized as a disease entity with serious health impacts. Consensus definitions have evolved to accurately identify sarcopenia. The 2014 Asian Working Group for Sarcopenia (AWGS) consensus defined sarcopenia as low muscle mass accompanied by low muscle strength and/or low physical performance.[1] Subsequently, in AWGS 2019, sarcopenia was defined as a combination of low muscle mass and either low muscle strength or physical performance ability, and the concept of “possible sarcopenia” was introduced for early at-risk cases. AWGS 2019 established specific cutoff values (handgrip strength less than 28 kg for males or less than 18 kg for females; walking speed less than 1.0 m/s or chair stand test of 12 seconds or longer for five repetitions) and maintained the previously established appendicular lean mass index (less than 7.0 kg/m2 for males or less than 5.4 kg/m2 for females based on dual energy X-ray absorptiometry [DXA]).[2] These changes reflect a shift in AWGS 2019 toward earlier case finding based on muscle strength or physical performance, while confirmed sarcopenia continued to require low muscle mass together with impaired muscle strength and/or physical performance. Recently, AWGS 2025 has reconceptualized sarcopenia within a life-course muscle health model. Compared to 2019, AWGS 2025 further simplified the diagnostic algorithm, reverting to simultaneously requiring both low muscle mass and muscle strength, while the updated consensus extended diagnostic consideration to middle-aged adults (50–64 years old), applied validated cutoff values, and simplified the algorithm to require both low muscle mass and muscle strength simultaneously (physical performance ability is used as an outcome measure rather than a diagnostic criterion). Additionally, it emphasized the importance of muscle health for healthy longevity.[3] These changes align with global initiatives and consider the interactions between muscle and other systems (bone, fat, immune system, etc.) during the aging process. Unifying definitions is important for regulatory agency recognition like the 2016 adoption of the International Classification of Disease, Tenth Revision code for sarcopenia and for promoting the development of therapeutic methods. Clear diagnostic criteria can open the path to increased clinical awareness of sarcopenia and the development of new drug therapies (Table 1).[4] Table 1Changing definitions and diagnostic criteria for sarcopenia in AsiaCharacteristicAWGS 2014AWGS 2019AWGS 2025Diagnosis processAssessment of low muscle mass with low muscle strength and/or low physical performanceCommunity and hospital/research pathways; possible sarcopenia introducedSimplified diagnosis requiring low muscle mass and low muscle strength; physical performance used as an outcome measureRecommended screening toolNot specifically addressedSARC-F recommendedSARC-F or SARC-CalF recommendedEssential elements for diagnosisLow muscle mass and low muscle strength and/or low physical performance Possible sarcopenia: low muscle strength or low physical performanceSarcopenia: low muscle mass and low muscle strength and/or low physical performanceLow muscle mass and low muscle strengthGrip standard criteriaMale <26 kg, Female <18 kgMale <28 kg, Female <18 kg ≥65 years: male <28 kg, female <18 kg50–64 years: male <34 kg, female <20 kgGait standard criteria<0.8 m/s<1.0 m/sNot a diagnostic criterion; ≥65 years <1.0 m/s; 50–64 years <1.2 m/s, used for functional/outcome assessmentMuscle standard criteria DXA: male <7.0 kg/m2, female <5.4 kg/m2BIA: male <7.0 kg/m2, female <5.7 kg/m2 DXA: male <7.0 kg/m2, female <5.4 kg/m2BIA: male <7.0 kg/m2, female <5.7 kg/m2 ≥65 years- DXA: male <7.0 kg/m2, female <5.4 kg/m2- BIA: male <7.0 kg/m2, female <5.7 kg/m250–64 years- DXA: male <7.2 kg/m2, female <5.5 kg/m2- BIA: male <7.6 kg/m2, female <5.7 kg/m2Age rangeOlder adults, generally ≥60 or ≥65 years depending on countryOlder adults, generally ≥60 or ≥65 years depending on countryOlder adults, with specific diagnostic criteria extended to adults aged 50–64 yearsAWGS, Asian Working Group for Sarcopenia; SARC-F, strength, assistance in walking, rising from a chair, climbing stairs, falls; SARC-CalF, SARC-F combined with calf circumference; DXA, dual energy X-ray absorptiometry; BIA, bioelectrical impedance analysis. Changing definitions and diagnostic criteria for sarcopenia in Asia Possible sarcopenia: low muscle strength or low physical performance Possible sarcopenia: low muscle strength or low physical performance Sarcopenia: low muscle mass and low muscle strength and/or low physical performance Sarcopenia: low muscle mass and low muscle strength and/or low physical performance ≥65 years: male <28 kg, female <18 kg ≥65 years: male <28 kg, female <18 kg 50–64 years: male <34 kg, female <20 kg 50–64 years: male <34 kg, female <20 kg DXA: male <7.0 kg/m2, female <5.4 kg/m2 DXA: male <7.0 kg/m2, female <5.4 kg/m2 BIA: male <7.0 kg/m2, female <5.7 kg/m2 BIA: male <7.0 kg/m2, female <5.7 kg/m2 DXA: male <7.0 kg/m2, female <5.4 kg/m2 DXA: male <7.0 kg/m2, female <5.4 kg/m2 BIA: male <7.0 kg/m2, female <5.7 kg/m2 BIA: male <7.0 kg/m2, female <5.7 kg/m2 ≥65 years- DXA: male <7.0 kg/m2, female <5.4 kg/m2- BIA: male <7.0 kg/m2, female <5.7 kg/m2 ≥65 years - DXA: male <7.0 kg/m2, female <5.4 kg/m2 - DXA: male <7.0 kg/m2, female <5.4 kg/m2 - BIA: male <7.0 kg/m2, female <5.7 kg/m2 - BIA: male <7.0 kg/m2, female <5.7 kg/m2 50–64 years- DXA: male <7.2 kg/m2, female <5.5 kg/m2- BIA: male <7.6 kg/m2, female <5.7 kg/m2 50–64 years - DXA: male <7.2 kg/m2, female <5.5 kg/m2 - DXA: male <7.2 kg/m2, female <5.5 kg/m2 - BIA: male <7.6 kg/m2, female <5.7 kg/m2 - BIA: male <7.6 kg/m2, female <5.7 kg/m2 AWGS, Asian Working Group for Sarcopenia; SARC-F, strength, assistance in walking, rising from a chair, climbing stairs, falls; SARC-CalF, SARC-F combined with calf circumference; DXA, dual energy X-ray absorptiometry; BIA, bioelectrical impedance analysis. Despite the evolution of these consensus statements, sarcopenia diagnosis faces significant challenges. Standard assessment relies on muscle mass estimates (DXA for lean mass or bioelectrical impedance analysis) and muscle function tests such as handgrip strength, walking speed, chair stand time, or a single physical performance battery. However, these measurements are generally conducted at irregular intervals in controlled environments and often involve subjective components dependent on user effort or examiner factors (patient effort during strength testing or examiner timing). For example, while handgrip strength is considered the standard for muscle strength, it primarily reflects upper extremity function and may fail to capture lower extremity deficits.[5] DXA-based lean mass is a two-dimensional projection and can cause confusion in specific populations (such as patients with implanted artifacts).[6] Moreover, low muscle mass alone is not always equated with functional decline, and muscle quality (fat infiltration, fiber composition) plays a role that DXA cannot identify. Advanced imaging such as computed tomography (CT) or magnetic resonance imaging (MRI) can assess muscle cross-sectional area and intramuscular fat. In fact, CT-based muscle attenuation (density) is an indicator of fat infiltration and is related to muscle function.[7] These imaging studies demonstrate that low muscle density (increased fat infiltration) is associated with reduced strength and performance ability. However, routine use of CT/MRI for sarcopenia screening is impractical due to cost, radiation exposure (in the case of CT), and resource constraints. Physical performance tests such as walking speed or chair stand test are useful outcomes but require in-person assessment and can be affected by patient effort, pain, or examiner skill. Therefore, traditional diagnosis provides only a “snapshot” of the patient’s condition and often involves subjective or crude assessment. Thus, conventional questionnaires and timed tests introduce variability and subjectivity, and quantitative indicators for sarcopenia assessment are incomplete. This highlights the necessity for digital biomarkers that can provide objective information about muscle function and patient status.[8] The key metrics, clinical use cases, and validation stages for each measurement technology are summarized in Table 2. Table 2Digital technologies for the assessment and monitoring of sarcopeniaMeasurement technologyBiomarker categoryClinical pathwayKey metricsClinical use caseValidation stageSmart watchDiagnostic (quality) & monitoringPathophysiological analysis & recoverySteps, daily activity, sleep patterns, heart rate variabilityScreening & monitoringReal-world validation against AWGS criteriaSmart insolePrognostic (fall risk)Risk stratification & balanceGait speed, cadence, center of pressure, balanceMonitoring & risk stratification (fall risk)Clinical validation in community-dwelling elderlyMarkerless motion captureDiagnostic & monitoringClinical assessment & severityJoint angles, SPPB components (sit-to-stand), gait kinematicsDiagnostic support & severity assessmentProof-of-concept; validation against lab-based motion capturesEMGDiagnostic (quality) & monitoringPathophysiological analysis & recoveryMuscle firing rate, fatigue index, power spectrumMonitoring & pathophysiological analysisExploratory studies; pilot clinical trialsAI-based imaging (X-ray/CT)DiagnosticDiagnostic support (opportunistic)Muscle cross-sectional area, intramuscular fat infiltrationOpportunistic screening & diagnostic supportInternal/external validation of AI modelsAWGS, Asian Working Group for Sarcopenia; SPPB, short physical performance battery; sEMG, surface electromyography; AI, artificial intelligence; CT, computed tomography. Digital technologies for the assessment and monitoring of sarcopenia AWGS, Asian Working Group for Sarcopenia; SPPB, short physical performance battery; sEMG, surface electromyography; AI, artificial intelligence; CT, computed tomography. THE EMERGENCE OF DIGITAL BIOMARKERS IN SARCOPENIA Digital biomarkers are defined as objective and quantifiable physiological and behavioral data collected and measured through digital devices, which reflect normal or pathological processes or responses to intervention.[9] These are measurements obtained from sensors, and similar to traditional blood or imaging biomarkers, they serve as indicators of health outcomes obtained through digital health technology. For example, smart watch sensors can continuously track walking speed or daily activity, providing information over time rather than a single measurement taken in the clinic. Digital biomarkers can complement and resolve the limitations of traditional sarcopenia measurement methods by providing continuous, highly objective real-world monitoring. In particular, a wearable surface electromyography (sEMG) system combining controlled electrical stimulation with signal-based modeling has been used to acquire stimulated muscle contraction signals and derive features related to muscle mass, muscle strength, and physical performance for sarcopenia assessment.[10] This is particularly relevant to sarcopenia, which is a multifaceted geriatric syndrome wherein muscle mass, strength, balance, and endurance are all interconnected. Compared to conventional laboratory biomarkers requiring sampling and standardized laboratory processing (such as blood tests for inflammatory cytokines or hormones), digital biomarkers are collected by sensors in the patient’s environment or within the body, and often occur in real-time.[11] With advances in sensor technology and analysis, the reliability and validity of digital measurements are also improving. Regulatory agencies and consortiums are actively discussing how to integrate digital assessment indicators into clinical trials, considering the potential of digital evaluation metrics to detect changes earlier or with smaller sample sizes. For example, projects such as Mobilise-D are working to validate wearable-derived gait measures for use in clinical research.[12] In other words, with digital biomarkers, muscle function and behavior can be measured continuously and objectively in daily life rather than being measured temporarily in the clinic.[13] DIGITAL BIOMARKER TECHNOLOGIES There exist digital sensor technologies considered particularly suitable for sarcopenia assessment. These include markerless motion analysis, smart insoles, sEMG, and smart watch devices. Each technology provides distinct but mutually complementary information about muscle strength and physical performance in alignment with the AWGS framework in Figure 1. Four digital technologies for assessing sarcopenia. IMU, inertial measurement unit; COP, center of pressure; sEMG, surface electromyography; RMS, root mean square; PPG, photoplethysmography. Risk stratification and gait stability assessment: Markerless motion analysis and smart insoles Quantitative gait and motion analysis for assessing physical performance ability has long been used in research and clinical biomechanics laboratories. Traditional marker-based motion analysis systems are widely used in research but require costly equipment, dedicated environments, and substantial time and technical expertise. Markerless motion analysis using computer vision has emerged as a more affordable and portable alternative for clinical movement assessment.[14] For sarcopenia, OpenPose-based kinematic analysis enables objective measurement of gait and balance parameters that reflect muscle function. The principal spatiotemporal gait measurements that can be obtained include walking speed, stride length, cadence (steps per minute), gait cycle phases (stance/swing, double support time), and variability in step timing.[15] Slow walking speed (less than 0.8 m/s) and frequent gait variability indicate frailty and fall risk. In patients with sarcopenia, markerless video analysis can reveal subtle compensatory patterns, such as asymmetrical stride length or forward trunk inclination that may not be apparent in brief clinical gait tests. Low gait speed has also been shown to be a useful indicator for identifying the risk of future functional disability in older adults.[16] Smart insoles for plantar pressure and balance analysis Another key technology for sarcopenia monitoring is smart insoles, which are shoe inserts embedded with pressure sensors and motion sensors. Since elderly individuals with sarcopenia often have impaired physical performance, specifically balance and gait stability, due to reduced muscle strength, measuring how weight and pressure are distributed when standing or walking is highly beneficial. Smart insoles are basically thin devices placed inside shoes, containing an array of foot pressure sensors (typically 8–16 sensors distributed across the forefoot, midfoot, and heel) and accelerometers/gyroscopes that sense movement. These insoles record in real-time the pressure exerted by the foot at multiple points and typically transmit the data to a smartphone or computer.[17,18] Smart insoles provide objective measures of plantar pressure, gait symmetry, and temporal gait parameters that may reflect sarcopenia-related impairments in gait and physical performance.[15,17,18] However, the discriminative value of smart-insole variables alone remains limited.[17] Therefore, their current role should be considered complementary functional monitoring rather than standalone diagnosis of sarcopenia. sEMG for muscle activation While motion capture and insoles focus on external movement and force, sEMG provides insight into the internal muscle activity that generates such movement. sEMG involves attaching electrodes to the skin over muscles to record the electrical signals generated when muscle fibers contract.[19] In sarcopenia, muscle strength deficiency is a characteristic phenomenon, and EMG can quantify how well muscles are activated and how quickly they fatigue. sEMG setup typically uses adhesive electrodes and strap-type sensor modules over the major muscle groups of interest. For lower extremity function, recordings can be made from the quadriceps, knee flexors, tibialis anterior, and gastrocnemius. These include the key muscle groups involved in human gait.[20] The EMG signal reflects the sum of muscle fiber action potentials, from which several measurements can be derived. (1) Activation level (amplitude): often expressed as the root mean square voltage of the EMG, which indicates how strongly the muscle is contracting. This can be normalized as a percentage of maximum voluntary contraction for comparison between individuals or sessions. For a given task, lower EMG amplitude may indicate insufficient muscle recruitment in sarcopenic muscle. (2) Muscle fatigue: as muscle fibers fatigue, the frequency content of the EMG signal shifts (slow-twitch fibers become predominant, lowering the median frequency [MDF]). Measurements such as the MDF or mean frequency of the EMG power spectrum tend to decrease with fatigue. This rate of decrease can be quantified as a fatigue index. (3) Coordination and timing: by measuring multiple muscles, patterns such as coactivation or muscle activation timing can be assessed.[21] Smart watch for activity trackers and smart rings that measure vital signs The final key technology for acquiring digital biomarkers in sarcopenia is wearable activity monitoring using smartwatches or smart rings. While motion capture, insoles, and EMG focus on specific functional tasks or brief assessments, smart watch devices continuously track an individual’s overall activity patterns throughout the day, sleep, and even cardiovascular metrics. Sarcopenia is not merely about how well someone performs on clinical tests but about how physically active and engaged they are in daily life, which consequently affects muscle health. Smartwatches (such as Apple Watch, Samsung Galaxy Watch, Garmin, Fitbit, etc.) and smart rings (such as Oura Ring, Galaxy Ring) are equipped with sensors such as accelerometers and gyroscopes, optical heart rate sensors, and sometimes additional sensors for body temperature, blood oxygen, or electrocardiogram. These devices are worn 24 hr a day on the wrist or finger, making them ideal for long-term monitoring.[22] Although these are not direct measurements of muscle, they are related to overall fitness and can inform problems that affect exercise endurance.[23] INTEGRATION WITH IMAGING AND DIGITAL TWIN MODELING FOR PLANNING Beyond direct patient monitoring, digital biomarkers are fused with advanced imaging techniques and simulation to facilitate clinical decision-making and planning. The concept of a “digital twin,” a virtual replica of a patient, is receiving attention in the medical field. In sarcopenia and orthopedic fields, a digital twin can be a detailed musculoskeletal model built based on a patient’s imaging information (X-ray, CT, MRI) and personalized with the patient’s functional data (muscle strength level, gait patterns, etc.). These models can be used to simulate and plan surgical or rehabilitative interventions. As discussed earlier, CT and MRI can provide detailed measurements of muscle size and quality. Latest segmentation models can automatically quantify muscle cross-sectional area and fat infiltration from scan images. For example, muscle area of the psoas major or thigh muscles from abdominal CT, or muscle volume of the thigh from MRI can be extracted and tracked. These imaging biomarkers are used together with digital functional biomarkers to paint a comprehensive picture of sarcopenia. Recent work applied a UNETR-based model to 72 whole-thigh CT scans from patients with hip fractures to segment 27 individual muscles and automatically quantify muscle volumes. The model achieved an average Dice score of 0.84 and an average correlation of 0.968 between predicted and ground-truth muscle volumes, supporting the feasibility of automated CT-based muscle quantification for sarcopenia assessment.[24] The greatest limitation of imaging-based sarcopenia assessment is accessibility and cost. Traditional DXA is accurate but requires specialized equipment and radiologist interpretation, making it difficult to be widely used in primary care settings. Park et al. [25] developed and validated artificial intelligence (AI)-based automatic segmentation models for muscle quantification using 351 lower-extremity long-axis X-ray images from 157 patients. Among the evaluated architectures, U-Net++ achieved the best segmentation performance, with an intersection over union of 0.93 and a Dice coefficient of 0.95. AI-derived muscle estimates showed a moderate-to-strong positive correlation with DXA-derived skeletal muscle index (r=0.72, P<0.0001). These findings suggest that X-ray-based muscle assessment may provide an accessible alternative for estimating muscle mass in primary care settings (Fig. 2).[26] Medical imaging-based sarcopenia biomarker. CT, computed tomography. MULTIMODAL SENSOR INTEGRATION: RATIONALE AND METHODS Each digital technology presented provides unique insights into patient function. However, sarcopenia is inherently multidimensional, simultaneously impairing strength, balance, endurance, and daily functioning. Relying on a single sensor can miss important aspects. Therefore, biomarkers for sarcopenia should consider multimodal integration across multiple sensors. For example, a wrist wearable device can tell you that overall activity is low but not why. A gait camera can show slow walking speed but not whether it is due to obesity or low muscle strength. Insoles can indicate asymmetrical weight bearing but cannot tell whether it is due to neurological deficit or simply habit. Complementary measurement modalities may support comprehensive sarcopenia assessment. Electrically stimulated sEMG demonstrated feasibility for muscle-function estimation in 98 healthy adults (r=0.89).[26] In sarcopenia research, feature-level fusion through machine learning is primarily used to combine data collected from multiple sensors. This approach extracts data from each sensor into features (walking speed from video, muscle activation level from EMG, daily activity from wearables) and then integrates these features into a single integrated feature vector for analysis.[27] For example, researchers can create an integrated feature set consisting of 20 variables including measurements of gait, balance, muscle function, cardiovascular status, and daily activity. By fusing this data, the goal is to capture a comprehensive “signature” of sarcopenia for each individual, which is subsequently used to train a classifier that outputs a composite “sarcopenia risk score.” Such a multimodal index could potentially support earlier detection, more granular risk stratification, and objective monitoring of intervention responses by integrating complementary information from multiple sensor-derived measures. The implementation of multimodal sensor integration through machine learning represents a paradigm shift in sarcopenia assessment. As outlined in the comprehensive development pipeline, this process progresses through five integrated phases: multimodal data collection, data preprocessing and fusion, biomarker discovery and development, clinical validation and regulatory approval, and ultimately clinical application and implementation. The pipeline begins with simultaneous data acquisition from complementary sensors including OpenPose for kinematic analysis, smart insoles for plantar pressure assessment, sEMG for muscle activation, and smartwatches for continuous activity monitoring. However, the heterogeneity of these sensor data, which differ in sampling rates, time scales, and measurement units, requires rigorous preprocessing. Synchronization and normalization of these diverse data streams establish a unified foundation for feature extraction. Following data preprocessing, feature learning extracts meaningful features from each sensor modality. OpenPose provides walking speed, stride length, and joint angle changes; smart insoles yield plantar pressure asymmetry and gait stability metrics; and sEMG quantifies muscle activation level, fatigue, and symmetry. Dimensionality reduction techniques then compress these diverse features into an integrated feature vector that comprehensively captures each individual’s sarcopenia profile across multiple dimensions of muscle function, gait, balance, cardiovascular status, and daily activity. Under the Software-as-a-Medical-Devices framework, clinical utility and safety must be demonstrated through clinical trials and regulatory approval (FDA/CE). Prospective longitudinal studies focusing on the correlation between AI-based risk scores and established outcomes, such as fall-related injuries or hospitalizations, can be used as validation methods. Upon clinical integration, digital biomarkers facilitate diagnosis through objective assessment, enable continuous disease monitoring, and support personalized treatment planning based on individual risk profiles and functional deficits. The development of digital biomarkers represents a comprehensive translational pathway from sensor technology to clinical practice (Fig. 3). By integrating multiple data streams through sophisticated analytical approaches including temporal and spatial alignment, multidimensional feature extraction, and AI-driven multimodal fusion, this pipeline generates sensitive and specific risk assessment tools that reflect each individual’s unique sarcopenia signature. This systematic approach enables earlier detection and more personalized intervention strategies, establishing the foundation for continuous, data-driven management of sarcopenia in real-world clinical settings. Beyond sensor-based monitoring, the integration of structural muscle assessment through advanced imaging further enhances sarcopenia evaluation and personalized treatment planning within this integrated framework. Integrated development pipeline for digital biomarkers in sarcopenia. CLINICAL IMPLICATIONS AND FUTURE PERSPECTIVES OF DIGITAL BIOMARKERS Objective and continuous functional data collected using four digital measurement technologies enable comprehensive understanding not only of muscle mass but also of the multidimensional characteristics of sarcopenia such as muscle strength, physical performance, daily activity level, coordination, and fatigue. Therefore, in the future, digital biomarkers such as “habitual walking speed in real-life settings” or “daily activity volume” could eventually be considered as key supplementary indicators in the “possible sarcopenia” stage or for classifying disease severity, and may be explored for inclusion in official diagnostic frameworks once standardized protocols are achieved. Additionally, these digital tools will enable personalized precision rehabilitation and remote monitoring, thereby establishing themselves as essential elements in tracking patient prognosis outside the clinic and quantitatively evaluating the effectiveness of interventions. Ultimately, digital biomarkers may offer a pathway to transform sarcopenia management from one-time assessment into a continuous and proactive data-driven approach. Despite the potential that digital biomarkers possess, important challenges and limitations exist that must be addressed for complete integration into sarcopenia diagnosis and clinical practice. First, current digital biomarkers lack standardized protocols and reference ranges comparable to conventional tests such as blood tests. Because data collection methods, algorithms, and output units differ for each sensor, there are difficulties in cross-comparing and interpreting results between research and clinical settings. Such standardized consensus is essential for regulatory agencies (FDA, European Medicines Agency, etc.) to recognize digital measurements as official diagnostic criteria or as final evaluation endpoints in clinical trials. Second, the multimodal approach is complex; future efforts must define what constitutes a ‘clinically meaningful change’ in these integrated digital twins. Securing interoperability between sensors and establishing consensus on data fusion algorithms for regulatory submission are essential next steps. Additionally, technical standards for signal processing technology that extracts valid information from raw data have not yet been completely established. Third, there is a practical problem that digital devices may be complex to set up or difficult for elderly patients to use, and compliance with continuous wearing and use may be low. Finally, privacy protection and data security for the sensitive health and behavioral data of individuals that is continuously collected are very important, and clear guidelines regarding patient consent procedures and ethical use of data must be established beforehand. CONCLUSION There is a vision that in the near future, digital biomarkers will become an essential part of both clinical practice and research for sarcopenia. It is envisioned that patients could eventually be evaluated through a combination of AWGS standards and validated digital metrics, which may support individualized training programs and long-term monitoring through digital risk scores to trigger timely interventions. Large-scale research initiatives and consortiums are actively working to standardize this approach and validate its efficacy. As evidence increases and improves, digital biomarkers hold the promise of enabling continuous, personalized, and preventive management of sarcopenia, ultimately helping elderly individuals maintain mobility and quality of life. This vision of data-driven individualized muscle health management represents a significant evolution that could herald a new era in the fight against sarcopenia. Footnotes Funding This research was supported by a grant of the Korean AR-PA-H Project through the Korea Health Industry Development Institute (KHIDI), funded by the Ministry of Health & Welfare, Republic of Korea (Grant No. RS-2024-00507256). Ethics approval and consent to participate Not applicable. Conflict of interest The authors declare no potential conflict of interest relevant to this article. REFERENCES