Wearable Technologies for Clinical Assessment and Monitoring in Pediatric Neurodevelopmental Disorders : A Narrative Review

Abstract

Pediatric neurodevelopmental disorders (NDDs) as defined by the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition, Text Revision : intellectual disabilities, communication disorders, attention-deficit/hyperactivity disorder (ADHD), autism spectrum disorder (ASD), specific learning disorder, and motor disorders together with closely associated pediatric neurological conditions such as cerebral palsy (CP) and epilepsy, present with symptom fluctuations and variable performance, making longitudinal assessment challenging. Conventional clinical scales, questionnaires, and caregiver diaries are largely subjective and offer limited temporal coverage. This review examines wearable technologies applicable to these conditions, focusing on measurable digital biomarkers and real-world clinical application. A narrative review was performed. Wearable technologies were categorized into motion trackers (accelerometers/gyroscopes, inertial measurement units [IMUs]), physiological monitors (photoplethysmography-derived heart rate and heart rate variability, and electrodermal activity), wearable electroencephalography (EEG)/brain sensors, smart glasses/cameras, and textile-based wearables. IMU and accelerometer based systems achieved posture and gait classification accuracies of approximately 85%, distinguished children with ADHD from controls with 80–90% accuracy, and detected stereotyped behaviors in children with ASD with sensitivities exceeding 90%. Wearable EEG demonstrated high performance for ASD classification (accuracy, 96%; sensitivity, 100%) and for automated sleep staging in epilepsy(accuracy, 80.8–83.5%), and also enabled seizure-related signal monitoring through detection of ictal fast activity. For tonic-clonic seizures, wearable devices achieved sensitivities of 89–94%, whereas tonic-seizure detection performance remains limited. Smartwatch vibrotactile stimulation reduced stereotyped behaviors and improved task performance, and wrist-worn biosignals enabled prediction of aggressive behavior approximately three minutes before onset. Augmented reality based smart glasses supported social-communication coaching in children with ASD. In children with CP, smart-shoe systems classified daily activities with high accuracy and quantified gait asymmetry. Wearables enable objective and continuous assessment of symptoms and behaviors in children with NDDs and related pediatric neurological conditions across real-world environments and are expanding toward personalized intervention. Remaining challenges include integration into clinical workflows (electronic medical record [EMR] linkage, clinician-friendly data summarisation, and reimbursement pathways), ethical and privacy considerations, and standardisation of validation protocols.

INTRODUCTION

Pediatric neurodevelopmental disorders (NDDs) are a clinically relevant nosographic group that affect a child’s brain development. According to the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition, Text Revision (DSM-5-TR), NDDs encompass intellectual disabilities, communication disorders (language disorder, speech-sound disorder, childhood-onset fluency disorder, and social communication disorder), autism spectrum disorder (ASD), attention-deficit/hyperactivity disorder (ADHD), specific learning disorder, and motor disorders (developmental coordination disorder, stereotypic movement disorder, and tic disorders) [29]. In this review we focus on these DSM-5-TR defined NDDs, while also including cerebral palsy (CP) and epilepsy two prevalent pediatric neurological conditions that, although not classified as NDDs within the DSM-5-TR framework, frequently co-occur with NDDs and share similar longitudinal-assessment challenges that are directly relevant to wearable-technology applications. Rapidly progressive conditions such as Krabbe disease, fluctuating disorders such as myasthenia gravis, and fatigue-dependent neuromuscular disorders such as spinal muscular atrophy and Duchenne muscular dystrophy are not NDDs under DSM-5-TR and are therefore outside the primary scope of this review, although they are occasionally referenced for methodological context.

NDDs interfere with typical neurodevelopment and are closely associated with motor, cognitive, and language impairments that substantially affect children’s daily functioning and quality of life. Motor impairments restrict mobility and behavior, whereas language deficits hinder the acquisition of cognitive and communication skills [11]. Management requires a multidimensional approach, and comprehensive individualized strategies are essential for both assessment and rehabilitation [12].

NDDs are chronic in nature and impose considerable burden on caregivers and society [29]. A defining challenge is accurate longitudinal assessment in the context of symptom fluctuations and performance variability [35]. Traditionally, assessment has relied on clinical rating scales, questionnaires, and caregiver diaries however, these methods are subjective, have limited reliability, and capture only a snapshot at a single time point. Repeated clinical visits to compensate for this limitation place substantial burden on children, caregivers, and healthcare providers.

Wearable sensors integrated into everyday items such as shoes, glasses, and clothing now enable non-invasive collection of physiological signals, movement patterns, and behavioral data in real-world environments [35]. These sensors offer continuous monitoring and are increasingly recognised as complementary tools to conventional clinical assessment [12,21]. This review examines the types of wearable technologies applicable to pediatric NDDs, the digital biomarkers they can measure, and the evidence for their clinical utility across key application domains. Beyond summarising technical evidence, we also address the translational gap between raw wearable outputs and routine clinical decision-making including integration into electronic medical records (EMRs), clinician-friendly data summarisation strategies, and reimbursement considerations which are essential for bringing these technologies to the bedside.

WEARABLE TECHNOLOGIES AND MEASURED BIOMARKERS IN PEDIATRIC NDDs

Wearable technologies enable continuous monitoring of motor, physiological, neural, and behavioral biomarkers that are difficult to assess in conventional clinical settings. They can be grouped into motion trackers, physiological monitors, wearable electroencephalography (EEG)/brain sensors, smart glasses/cameras, and smart textiles (Fig. 1). Table 1 summarises device categories, representative examples, measured biomarkers, and clinical applications.

Fig. 1.

Representative wearable technologies and measurable digital biomarkers in pediatric neurodevelopmental disorders. Artificial intelligence (ChatGPT, OpenAI) was used to assist in generating the graphical layout of the figure. The authors reviewed and edited the content for accuracy. IMU : inertial measurement unit, HRV : heart rate variability, PPG : photoplethysmography, ECG : electrocardiography, EDA : electrodermal activity, EEG : electroencephalography.

Table 1.

Overview of wearable technologies and digital biomarkers in pediatric neurodevelopmental disorders

Wearable technology category	Device examples	Sensors/data sources	Key biomarkers or features	Clinical applications in pediatric NDD
Motion trackers	Wrist/ankle IMU, wearable accelerometer, smart band	Accelerometer, gyroscope	Activity level, gait speed, stride variability, repetitive movement frequency, step symmetry	ADHD hyperactivity quantification, ASD stereotyped-behavior detection, motor function monitoring, real-world functional assessment (CP)
Physiological monitors	Smartwatch, chest band, physiological wristband	PPG, ECG, EDA, pupillometer	HR, HRV (SDNN, rMSSD, LF/HF ratio), SCL, SCR, autonomic arousal index	Stress and self-regulation monitoring, emotional reactivity assessment, longitudinal intra-individual tracking (ADHD, ASD)
Wearable EEG/brain sensors	Mobile EEG headset, dry-electrode EEG headband (e.g., Dreem 3)	EEG signal (frequency-domain, connectivity, and nonlinear features)	Spectral power, functional connectivity, sleep-spindle density, seizure-related fast activity	ASD classification support, sleep staging, seizure monitoring (epilepsy), developmental biomarker assessment
Smart glasses & cameras*	AR smart glasses (e.g., Google Glass), wearable camera	Video, audio, gaze tracking, facial recognition	Gaze direction, facial expression, pre-meltdown gesture patterns	Social-interaction coaching, pre-meltdown gesture detection, real-time caregiver alerting (ASD, ADHD)
Textile-based wearables	Smart clothing, smart shoes (e.g., Pediatric SmartShoe)	Conductive textile ECG electrodes, plantar pressure sensors (FSRs), tri-axial accelerometer	ECG, HRV, gait-symmetry index, posture stability, plantar-pressure distribution	Continuous physiological monitoring, gait assessment in CP, improved wearability for sensory-sensitive children

^*The home smart-camera system functions as a fixed home-monitoring system rather than a body-worn wearable (see Section 2.4).

NDD : neurodevelopmental disorder, IMU : inertial measurement unit, ADHD : attention-deficit/hyperactivity disorder, ASD : autism spectrum disorder, CP : cerebral palsy, PPG : photoplethysmography, ECG : electrocardiography, EDA : electrodermal activity, HR : heart rate, HRV : heart rate variability, SDNN : standard deviation of normal-to-normal intervals, rMSSD : root mean square of successive differences, LF : low-frequency, HF : high-frequency, SCL : skin-conductance level, SCR : skinconductance response, EEG : electroencephalography, AR : augmented reality, FSR : force-sensitive resistor

Motion-tracking wearables

Accelerometer and gyroscope based motion trackers allow continuous, noninvasive quantification of motor variables [16] and are used to assess activity levels, sleep wake patterns, and repetitive behaviors in children with NDDs [3,8]. Inertial measurement units (IMUs) collect and analyse movement data to evaluate posture and gait. When combined with supervised machine-learning techniques such as support vector machines, random forests, and extreme gradient boosting, IMU based systems classify posture and gait patterns with an accuracy of approximately 85% [17].

Accelerometer-based wearables can distinguish children with ADHD from typically developing controls with accuracies of 80–90%, and in some convolutional neural network (CNN) based studies using acceleration-image features, accuracies approaching 93% have been reported [22]. In children with ASD, deep-learning models combining IMU data with CNNs and long short-term memory (LSTM) networks detect stereotyped behaviors including hand flapping, body rocking, and combined hand body rocking with sensitivities exceeding 90% [25]. These findings suggest that wearables can objectively quantify hyperactivity, impulsivity, and repetitive motor behaviors that are difficult to assess accurately in brief clinical settings.

Studies comparing gait parameters between clinical environments and daily life have documented inconsistencies in gait speed, step length, and gait variability [10], indicating that clinic-based assessments may not adequately reflect real-world functional performance. Continuous recording of everyday behavioral patterns and motor signals is therefore essential for comprehensive assessment of children with NDDs.

Physiological monitoring wearables

Physiological monitoring utilises heart rate (HR) and heart rate variability (HRV) indices derived from photoplethysmography (PPG), and electrodermal activity (EDA) measures including skin conductance level (SCL) and skin conductance response (SCR) which reflect sympathetic-nervous-system arousal [7]. Wearable electrocardiography (ECG) and impedance cardiography enable comprehensive analysis of autonomic function. Changes in respiratory sinus arrhythmia and pre-ejection period can indicate autonomic responses to emotional stimuli in children [13]. EDA components (SCL and SCR) reflect sympathetic activity and stress responses [19].

In children with NDDs, physiological indicators such as HR, HRV, EDA, and pupil diameter serve as objective biomarkers of arousal and self-regulation. Studies measuring pupil size in children with ADHD and ASD found no significant group differences across conditions, but documented intra-individual associations between pupil size and task performance suggesting that physiological monitoring is more informative for tracking change within individuals over time than for detecting fixed group-level differences [6]. Wearable-based physiological monitoring can therefore serve as an important objective assessment tool for children with NDDs in whom self-regulation and stress responses are difficult to evaluate behaviorally.

Wearable EEG and brain sensors

Wearable EEG sensors enable acquisition of neurophysiological biomarkers in children with NDDs. Wireless wearable EEG systems using dry electrodes overcome several limitations of conventional clinical EEG including long preparation times, low compliance, and hospital-centered measurement and allow recording of brain activity in daily-life settings [19].

By extracting frequency-domain and nonlinear features from wearable EEG signals and applying machine-learning classifiers, ASD has been distinguished from typical development with an accuracy of 96% and sensitivity of 100% [2]. It should be emphasised, however, that conventional clinical EEG is primarily used to detect interictal epileptiform discharges and to evaluate background activity in patients with epilepsy, rather than for ASD classification. If wearable EEG is applied to ASD screening, its performance must be compared rigorously with established diagnostic instruments such as the Autism Diagnostic Observation Schedule and the Autism Diagnostic Interview-Revised, and the clinical relevance of any wearable EEG abnormality in children with NDDs but without epilepsy must be explicitly validated before routine use can be recommended.

Wearable EEG systems are also expanding into brain computer interface applications, enabling reliable neural signal assessment and repeated measurement in pediatric NDD populations [27]. For children with epilepsy, wearable EEG has substantial potential for seizure monitoring : ictal fast activity, interictal epileptiform discharges, and reduction of slow-wave background are characteristic EEG signatures that can be detected by reduced-channel wearable systems [24]. Wearable EEG-based seizure detection, particularly when combined with motion and autonomic signals, is therefore an actively developing clinical application (see Section 3.4 for clinical validation data). Technical challenges remain, including electrode-contact stability, susceptibility to movement artifacts, battery life, and device size [33].

Smart glasses and camera-based systems

Augmented reality (AR)-based smart glasses utilise embedded cameras, microphones, and IMUs to help children interpret social cues including facial expressions, gaze, and tone of voice and to direct attention toward relevant social targets. The Empowered Brain system, implemented on Google Glass-based AR smart glasses, supports communication and social interaction in children and adolescents with ASD [31].

A home smart-camera system (EZVIZ C6W Wi-Fi camera) has also been described in the literature for detecting pre-meltdown behaviors. Strictly speaking, this fixed camera system is a home-based monitoring technology rather than a body-worn wearable device, and we include it here only for contextual completeness of the ambulatory-monitoring ecosystem; it falls outside the formal definition of wearables adopted in the remainder of this review [1]. The system uses OpenCV-based real-time image processing to recognise pre-meltdown gestures such as crying, screaming, head-holding, and ear-covering and notifies caregivers when atypical behaviors are detected, achieving approximately 19 frames per second in real-world environments [1].

Textile-based wearables

Textile-based wearables, such as smart shoes and smart clothing, measure physiological and biomechanical signals via sensors in close contact with the body. Shoe-based wearables use plantar pressure sensors (force-sensitive resistors [FSRs]) and inertial sensors to classify activity types and quantify gait parameters. In children with CP, the Pediatric SmartShoe equipped with multiple FSR pressure sensors (at the heel, metatarsals, and hallux) and a tri-axial accelerometer quantifies gait, sitting, standing, heel-strike and toe-off timing, gait interval, and postural symmetry [14].

Clothing-based wearables with silver-coated conductive textile electrodes at the chest enable continuous dry-electrode ECG recording, from which HRV indices are derived : time-domain measures (average normal-to-normal [NN] interval, standard deviation of NN intervals, root mean square of successive differences, pNN50, mean HR) and frequency-domain measures (low-frequency [LF] power, high-frequency [HF] power, and normalised LF/HF ratio), which serve as digital biomarkers of autonomic balance [20]. Textile wearables generally show relatively high acceptance among children with NDDs because they provide minimal additional tactile stimulation; however, children with severe sensory sensitivities may still experience difficulty with wear, underscoring the need for individualised adaptation protocols.

CLINICAL APPLICATIONS OF WEARABLES IN PEDIATRIC NDDs

Wearable devices are increasingly used to monitor and analyse symptoms and behaviors of children with NDDs in real time, complementing the limitations of periodic clinical assessment. They are gaining importance as clinical decision-support tools that enhance the precision and continuity of care (Table 2).

Table 2.

Clinical applications of wearable technologies in pediatric neurodevelopmental disorders

Clinical application domain	Wearable devices	Sensors/data sources	Key parameters/biomarkers	Clinical utility
Behavioral monitoring	Smartwatch, wristband biosensor (e.g., Empatica E4)	ACC, EDA, PPG	Activity level, repetitive-behavior frequency, physiological-arousal patterns, pre-aggressive biosignal changes	Objective quantification of repetitive and aggressive behaviors; early prediction (minutes before onset); evaluation of intervention response; personalized behavioral-intervention design
Sleep monitoring	EEG headband (e.g., Dreem 3), actigraphy watch, IMU sensor	EEG, IMU, accelerometer	Sleep stages (NREM, REM), WASO, SFI, sleep-spindle density, body-position change frequency	Home-based longitudinal sleep assessment; identification of neurophysiological sleep biomarkers (e.g., reduced spindle density in ASD); complement to PSG; reduced caregiver burden
Emotion recognition and social interaction support	AR smart glasses, wearable EEG headband, smartwatch	Camera, EEG, PPG	Gaze patterns, facial-emotion-recognition accuracy, HR changes related to emotional arousal	Real-time social coaching; emotional-awareness enhancement; closed-loop neurofeedback; support for communication and self-regulation (ASD)
Seizure detection monitoring	Wrist-worn seizure-detection devices (e.g., NightWatch, Embrace2), multimodal patch (e.g., Plug 'n Patch), behind-the-ear EEG sensors	PPG, ACC, EEG, EMG, ECG	Autonomic changes (HR, EDA), movement abnormalities, seizure-associated EEG fast activity, EMG contraction patterns	Early nocturnal seizure detection; caregiver alert; SUDEP-risk reduction; reduced caregiver stress; tonic-seizure detection remains limited
Assessment of motor and physical function	Smart shoes (e.g., Pediatric SmartShoe), wrist-worn activity trackers	IMU, plantar pressure sensors (FSRs), accelerometer	Gait cycle, stance/swing time, step symmetry, cadence, upper-limb use asymmetry, MACS-level correlation	Real-world motor-performance assessment in CP; quantification of functional limb use; monitoring of rehabilitation outcomes in daily life

ACC : accelerometry, EDA : electrodermal activity, PPG : photoplethysmography, EEG : electroencephalography, IMU : inertial measurement unit, NREM : non-rapid eye movement, REM : rapid eye movement, WASO : wake after sleep onset, SFI : Sleep fragmentation index, ASD : autism spectrum disorder, PSG : polysomnography, AR : augmented reality, HR : heart rate, EMG : electromyography, ECG : electrocardiography, SUDEP : sudden unexpected death in epilepsy, FSR : force-sensitive resistor, MACS : Manual Ability Classification System, CP : cerebral palsy

Behavioral monitoring

In children with ASD and intellectual disability, HF vibrotactile stimulation delivered via a smartwatch reduced stereotyped behaviors (leg shaking, body rocking, and vocalizations) during intervention sessions. Fine-motor precision during task performance simultaneously improved, and the frequency of imitative vocalizations increased [28]. These findings demonstrate that wearables can simultaneously measure repetitive behaviors and evaluate intervention responses.

Wrist-worn wearables (Empatica E4) have been used to collect blood-volume pulse, EDA, and accelerometer data from children with ASD. Machine-learning analysis of these biosignals enabled prediction of aggressive behavior approximately three minutes before onset [18], suggesting that wearable biosignals may serve as predictive biomarkers of problematic behaviors. In children with ADHD, wearable devices continuously quantify on-task versus off-task behavior, activity levels, and usage patterns, enabling real-time behavioral measurement, supporting self-awareness, and facilitating personalised intervention design [4]. Wrist-worn actigraphy has also been used to quantify spontaneous behavioral patterns and upper-limb use in children with hemiplegic CP [5].

Sleep monitoring

Children with NDDs frequently experience sleep disturbances that negatively affect attention and cognition. Wearable-based sleep monitoring reduces the burden associated with in-hospital polysomnography (PSG) and allows analysis of sleep duration, sleep stages, and nocturnal posture and movement [9].

Using an EEG headband (Dreem 3), sleep stages including non-rapid eye movement (NREM) and rapid eye movement (REM) sleep as well as wake after sleep onset (WASO), Sleep fragmentation index (SFI), and neurophysiological markers such as sleep spindles and slow oscillations can be assessed. Wearable EEG has revealed reduced sleep-spindle density in children with ASD, potentially reflecting abnormalities in thalamocortical circuits and serving as a candidate digital biomarker [15].

A comparison of nocturnal posture changes using IMUs in children with CP and typically developing children found that children with severe CP (Gross Motor Function Classification System [GMFCS] IV-V) exhibited significantly fewer posture changes per hour (mean 0.49/hour) and longer posture-maintenance durations than typically developing children or those with mild CP (GMFCS I-II) [9]. These findings objectively demonstrate limited spontaneous postural adjustment during sleep in children with severe motor impairment, with implications for long-term risks such as postural deformities and musculoskeletal complications.

In children with drug-resistant epilepsy, validation of wearable actigraphy against continuous PSG noting that video-EEG (VEEG) is primarily a tool for seizure detection rather than routine sleep assessment showed high agreement for total sleep time, nocturnal wake time, and daytime sleep proportion, but limited accuracy for WASO [26]. Comparisons between actigraphy and caregiver sleep diaries have likewise demonstrated acceptable agreement for sleep onset, sleep offset, and daytime sleep, but insufficient agreement for actual night-time sleep and WASO with caregivers tending to misestimate night-time sleep duration relative to actigraphy [30], highlighting the importance of objective wearable monitoring for precise sleep-quality assessment.

A deep-learning-based automatic sleep-staging algorithm (SeqSleepNet), using only three channels (EEG C4-A1, electrooculography, and chin electromyography [EMG]), achieved accuracies of 83.5% in controlled epilepsy and 80.8% in drug-resistant epilepsy, with particularly high agreement for REM-sleep duration [23]. This indicates that channel-reduced wearable EEG can provide hypnogram level sleep analysis in children who cannot easily undergo repeated PSG.

Emotion recognition and social interaction support

A study of children with ASD using a wearable EEG headband and a machine-learning-based closed-loop neurofeedback system across 60 intervention sessions reported improvements in expressive language and social-cognitive domains [34], suggesting that wearable brain sensors may extend beyond monitoring to function as digital intervention devices.

AR-based smart glasses provide real-time social-communication coaching during interactions, using camera-based facial and emotion recognition with gamified feedback to encourage attention to faces and emotion inference [31]. Wearables designed to support emotional regulation in children with ADHD link physiological signals such as HR to the child’s subjective interpretation of emotional states, promoting reflective emotional awareness. However, visible device notifications may raise concerns regarding stigmatisation or social embarrassment, particularly in school settings.

Seizure detection monitoring

Seizure detection and safety monitoring are particularly critical clinical needs in pediatric epilepsy, where prompt response is essential.

In a 2-month home study using NightWatch a multimodal wearable integrating PPG-based HR and accelerometer signals overall seizure sensitivity was 89%, with individual-level sensitivity reaching 100% in some participants. Sensitivity was 94% for tonic-clonic seizures, 83% for hypermotor seizures, and 91% for other major motor seizures. Caregiver stress levels decreased after device use; however, the false-alarm rate was 0.04 per hour, and false positives were higher in children with comorbid intellectual disability [32].

In children with Lennox-Gastaut syndrome and drug-resistant epilepsy, a multimodal wearable system (Plug 'n Patch) integrating behind-the-ear EEG, surface EMG, ECG, accelerometer, and gyroscope was evaluated alongside VEEG for detection of tonic seizures. Among 99 tonic seizures, overall sensitivity was 41% and positive predictive value (PPV) was 9%, indicating limited performance for this seizure type. However, for seizures during sleep lasting ≥20 seconds, performance improved substantially (F1=0.71; sensitivity, 62%; PPV, 82%). Characteristic biosignal patterns during tonic seizures including paroxysmal fast activity on EEG, sustained EMG contraction, brief abrupt movement followed by immobility, and HR increases were consistently observed [24].

Seizure-detection wearables have important clinical implications for nocturnal seizure monitoring and for reducing the risk of sudden unexpected death in epilepsy (SUDEP). Limitations include discomfort from skin-attached electrodes, signal loss, and the current absence of real-time online alert functions in some systems [24]. From a translational perspective, wearable-derived seizure logs should ideally be integrated into EMRs through standardised application-programming interfaces, with automated summarisation tools that present clinicians with per-day event counts, severity indices, and time-to-detection metrics rather than raw waveforms. Regulatory progress is being made in this domain for example, the Empatica Embrace2 seizure-detection device has received U.S. Food and Drug Administration (FDA) 510(k) clearance, providing one pathway toward insurance reimbursement; broader uptake will require prospective health-technology-assessment studies demonstrating clinical utility and cost-effectiveness.

Assessment of motor and physical function

In the rehabilitation of children with CP, conventional gait assessments often rely on fixed laboratory equipment and may not reflect real-world performance, partly because of learned non-use in which motor ability in clinical settings exceeds spontaneous use in daily life.

Shoe-based wearables combine plantar pressure sensors and inertial sensors to distinguish daily activities (sitting, standing, walking) and to derive longitudinal gait parameters. The Pediatric SmartShoe classifies these activities with high accuracy in children with CP (primarily GMFCS I-II) and in typically developing children, and quantifies gait cycle, stance and swing time, step time, single-limb support time, and cadence. Asymmetry indices of posture and gait derived from wearable data show good agreement with direct measurements [14].

Continuous monitoring of upper-limb use with actigraphy wristbands has demonstrated significant differences in usage patterns between children with unilateral CP and typically developing children, with upper-limb activity asymmetry differing significantly across Manual Ability Classification System (MACS) levels [5].

DISCUSSION

This review examined the characteristics of wearable technologies, measurable digital biomarkers, and clinical applications in pediatric NDDs and closely associated pediatric neurological conditions (CP and epilepsy). Wearables complement the temporal and spatial limitations of traditional clinical assessments and enable objective analysis of function and symptoms in daily life [12,21].

Objective quantification and intra-individual tracking

Motion sensors and physiological wearables are valuable not only for group-level differentiation but also for tracking change within individuals over time. Accelerometer-based data distinguish children with ADHD from controls with 80–90% accuracy [22], and IMU data combined with CNN/LSTM models detect stereotyped behaviors in children with ASD with high sensitivity [25]. Physiological monitoring HR, HRV, and EDA may more reliably reflect within-individual autonomic change than fixed between-group differences [6].

Ecological validity

Inconsistencies between hospital and home gait parameters [10] underscore that clinic-based evaluations may not represent real-world functioning. Smart shoes can classify and quantify gait and activity in children with CP [14], and asymmetry in bilateral upper-limb activity in children with unilateral CP correlates with MACS levels [5].

Wearables as intervention tools

Wearables are expanding beyond monitoring to function as intervention tools. Smartwatch vibration reduced stereotyped behaviors and improved functional performance [28], and wrist-worn biosignals predicted aggressive behavior before onset [18]. AR smart glasses providing communication coaching and real-time detection of pre-meltdown gestures highlight the potential of integrated platforms combining observation, feedback, and training [1,31]. Closed-loop neurofeedback using wearable EEG further illustrates this expansion [34].

Sleep and seizure monitoring

Sleep and seizure monitoring represent areas of particularly high clinical need. Wearable EEG measures not only sleep stages but also neurophysiological markers such as spindle density in children with ASD [15], and IMU-based posture monitoring provides clinically relevant data on nocturnal postural adjustment in children with severe CP [9]. For seizure monitoring, high sensitivity and low false-alarm rates with NightWatch support its value in home settings [32]; the limited tonic-seizure detection performance of the Plug 'n Patch system highlights the need for seizure-type-specific multimodal approaches [24].

Clinical translation : pathways and barriers

A key concern raised across the literature is how wearable-derived data can be translated into routine clinical workflows. Three complementary implementation models are emerging : 1) automated data-summarisation dashboards that translate continuous wearable outputs into clinician-interpretable metrics (e.g., daily seizure counts, weekly sleep-quality indices, behavior-frequency trends); 2) threshold-based alert systems that notify clinicians or caregivers only when biosignals exceed predefined abnormal ranges, thereby reducing information overload; and 3) structured EMR integration modules that link wearable-derived biomarkers to standardised clinical-assessment scales (e.g., Vineland Adaptive Behavior Scales, GMFCS level) to support longitudinal monitoring. Insurance and reimbursement frameworks for wearable monitoring remain underdeveloped in most healthcare systems. Some progress is visible in epilepsy monitoring: several seizure-detection wearables (e.g., Empatica Embrace2) have received FDA 510(k) clearance, establishing one regulatory pathway toward reimbursement. Broader adoption across NDD indications will require prospective evidence of clinical utility and cost-effectiveness from health-technology assessment studies.

Challenges and limitations

Children with NDDs and related pediatric neurological conditions can benefit substantially from continuous, objective monitoring through wearables [12,21], supporting assessment and intervention across behavior, sleep, seizures, and motor function [1,18,28,31,32]. However, challenges remain regarding integration into clinical workflows, children’s acceptance, and ethical and privacy considerations [24,33]. Technical issues including electrode stability, signal-artifact susceptibility, battery life, and data security must be addressed. Social stigma associated with visible devices and the potential for notification-induced embarrassment are further barriers, particularly among older children and adolescents. Systematic clinical-validation studies within multidisciplinary frameworks are required before widespread adoption.

CONCLUSION

Wearable technologies provide objective, continuous, and real- world measurement of motor, physiological, behavioral, and neural signals in children with NDDs, CP, and epilepsy. This review demonstrates that wearables are expanding beyond diagnostic support and symptom monitoring toward evaluation of treatment responses and direct intervention.

Wearable-derived data capture everyday functioning and changes not accessible through conventional assessments, providing a foundation for personalised treatment, prognosis prediction, and digital-biomarker development. Successful clinical translation will require evidence-based EMR integration strategies, standardised data summarisation, appropriate regulatory and reimbursement frameworks, and careful resolution of ethical, privacy, and equity considerations. Systematic multidisciplinary validation studies are the essential next step toward establishing wearable technologies as core digital-health tools in pediatric neurology and rehabilitation.