Emerging Role of Portable Accelerometers for Assessment of Gait Instability in Idiopathic Normal Pressure Hydrocephalus: A Systematic Review

Abstract

Background

Idiopathic normal pressure hydrocephalus (iNPH) is a reversible cause of gait disturbance, cognitive decline, and urinary incontinence in older adults. Early detection with shunt surgery can significantly improve symptoms.

Methods

A systematic review following PRISMA guidelines evaluated the role of portable and smartphone-based accelerometers in quantifying gait parameters in iNPH.

Results

Across 32 studies, 22 different accelerometers were used, ranging from wearable inertial sensors to smartphone applications. The Three-Dimensional Pose Tracker for Gait Test (TDPT-GT) app was the most frequently utilized, enabling detection of pathological gait features such as shuffling, short steps, and wide-based walking. The G-Walk device was commonly used in clinical assessments, including before and after CSF tap tests. Several studies integrated accelerometry with machine-learning models, improving differentiation of iNPH from other neurological gait disorders.

Conclusion

Portable accelerometry offers an accessible, objective tool for diagnosis, monitoring, and prognostication in iNPH, with growing potential for routine clinical integration.

Keywords

idiopathic normal pressure hydrocephalus normal pressure hydrocephalus accelerometry portable accelerometer gait analysis

Introduction

Idiopathic normal pressure hydrocephalus (iNPH) is a treatable form of communicating hydrocephalus mainly seen in elderly adults affecting men and women equally. Few studies have been conducted to determine its exact prevalence, currently estimated to lie between 10 and 22 per 100 000 in the general population, 1.3% in those over 65, and 5.9% in those over 80.^1,2 Although the etiology and pathophysiology remain to be fully elucidated, iNPH is characterized by an increase in cerebrospinal fluid (CSF) volume, speculated to be due to a disruption of CSF flow dynamics and impaired CSF drainage and absorption.¹ This ultimately leads to ventricular enlargement, mechanical stress on brain parenchyma, hypoperfusion, neuroinflammation, and subsequent brain atrophy, despite normal intracranial pressure (ICP).¹

iNPH is described by a classic triad of gait instability, cognitive decline, and urinary incontinence. Diagnosis is often difficult, as these symptoms may be present in numerous other conditions of varying etiology such as dementia, cerebrovascular disease, and Parkinson’s disease (PD).³ In order to make a diagnosis, current guidelines recommend using a combination of clinical history, brain imaging, and physical exam findings to categorize patients into “probable,” “possible,” and “unlikely” categories.^3,4 Gait instability is typically the initial presentation, often described as “magnetic” and characterized by shuffling movements, difficulty in step initiation, and bradykinesia. Patients are generally placed in the “possible” category if they have more than one symptom of the triad not explained by any other neurological or non neurological disease, with evidence of ventriculomegaly on brain imaging.³ Diagnostic criteria for each category are described in detail by the iNPH guidelines and Japanese Society of Normal Pressure Hydrocephalus.^3,4

For detailed gait analysis, existing laboratories utilize cameras, force plates, and other instruments to quantify various gait parameters. However, these methods are expensive, cumbersome, and reserved for research studies. In most clinical settings gait is assessed through visual evaluation in the form of the timed-up-and-go (TUG) test, 10-meter walk test, and 2-minute walk test. The TUG test involves standing up from a chair, walking 3 meters, turning around, walking back to the chair, and sitting down.⁵ If iNPH is suspected, these studies are often performed after lumbar puncture (LP), often referred to as a CSF tap test (CSF-TT), a diagnostic tool used to predict patient responsiveness to ventriculoperitoneal (VP) shunt placement.⁶ Prior studies have shown discrepancies between objective and subjective evaluators of gait in iNPH patients.⁷ Given that 70-90% of patients diagnosed with iNPH show clinical improvement following CSF drainage, early diagnosis and treatment are essential.⁶

Accelerometers, present in commonly used electronic devices such as cell phones, tablets, smart watches, cameras, and video game controllers are used to measure linear acceleration. Accelerometry has been increasingly used to measure gait patterns and accurately quantifies gait, a variable historically defined by subjective and qualitative interpretation. These devices are small, portable, inexpensive in comparison to current gait-analyzing technologies, and do not interfere with patients’ inherent gait patterns.⁸ Numerous studies have shown that portable accelerometers are equally effective, and potentially more effective, at quantifying gait parameters as compared to the gold standard laboratory gait assessment. Not only this, but because of their portable nature, gait analysis is possible both in the clinic and real-world settings.⁹ A substantial amount of research has provided evidence that portable accelerometers can be used as a tool to enhance diagnostic accuracy, monitor disease progression, stratify risk, and evaluate therapeutic response in movement disorders such as PD, ataxia, and dementia.¹⁰ It has been reported that portable accelerometry can even detect prodromal PD years before clinical diagnosis better than other methods such as genetics, lifestyle, biochemistry, and reported symptoms.¹¹ Given their affordability and utility they may be more accessible to iNPH patients and have the potential to measure multiple gait parameters including speed, cadence, stride length, step time, stance, swing, double support, lateral step variability, and foot strike angle.¹²

Gait improvement in iNPH remains the principal clinical outcome guiding both treatment decisions and postoperative follow-up, making changes in gait performance one of the most sensitive and clinically meaningful markers of therapeutic response. Contemporary reviews highlight the multidimensional nature of gait disturbance in iNPH, encompassing not only reduced speed and shortened stride length, but also abnormalities in balance, variability, turning, and gait initiation. This complexity suggests that brief, in-clinic assessments may fail to capture the full spectrum and temporal dynamics of gait impairment. Wearable accelerometry is therefore uniquely positioned to provide real-life, multiparametric, and continuous monitoring, extending beyond daytime walking to potentially include nocturnal or sleep-related movement patterns and smartphone-based assessments.^13,14

Despite these advantages and growing evidence in other movement disorders, data on the use of portable accelerometers in iNPH remain limited. The purpose of this review is to identify how portable accelerometers can be used to quantify gait patterns in iNPH patients to facilitate early detection, accurate diagnosis, selection for surgery, and treatment.

Methods

The Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA) guidelines were followed during the literature search.¹⁵ This systematic review was prospectively registered in PROSPERO (Registration No. CRD420251249620) prior to the initiation of data extraction.

Search Strategy

A review of the literature was conducted using PubMed, Embase, Ovid, and Web of Science databases in July 2025. The search terms included: accelerometry, accelerometer, gait analysis, software, and normal pressure hydrocephalus. Terms included in either title, abstract, or key words were used to identify all relevant literature. Detailed search strategy for each data base is reported in Supplemental File 1.

Eligibility Criteria

Articles included were (i) any observational studies (cross-sectional, case–control, prospective, or retrospective cohorts), interventional trials (randomized, non-randomized, proof-of-concept), case-series, and case reports, (ii) published in a peer-reviewed journal, (iii) reported original data on gait assessment using portable or smartphone based accelerometry, (iv) in people with idiopathic normal pressure hydrocephalus (iNPH) only, or in mixed cohorts where iNPH patients were included, with or without a comparison to a control group. Articles were excluded if they (i) were published before 2008 or written in a language other than English, (ii) primary investigators did not use portable or smartphone-based accelerometers to analyze gait, (iii) or iNPH was not included in the primary clinical focus.

Selection Process

Results of the literature search were imported to reference manager software (EndNote20). After removing duplicates, references were imported to the Rayyan web platform to proceed with the screening and the first selection. Titles and abstracts were independently screened by two reviewers (RD, EB). Any disagreements were blindly screened by a third reviewer (JA). The same process was repeated for full text review of the papers that past first selection.

Quality Assessment

Following the screening process, thirty-two articles that adhered to all inclusion criteria were assessed for quality and potential bias using the Joanna Briggs Institute (JBI) Critical Appraisal Tools. Each article was evaluated by using the appropriate JBI checklist and further classified as low risk of bias (>70%), moderate risk of bias (50-70%), or high risk of bias (<50%). After discussion amongst researchers, all thirty-two articles were deemed as low risk of bias (>70%) and included in this systematic review (Supplemental File 2).

Results

Using the search criteria, a total of 3894 articles were identified. After deduplication, a total of 660 were screened for abstract and title, of which 58 were chosen for full-text review. Following full-text review, 32 papers were selected to be included in the review (Figure 1). In total 22 different accelerometers were used across studies with findings in patients with iNPH ranging from decreased gait speed, increased step time, and decreased trunk acceleration (Figure 2). These studies were categorized into five groups based on their use of accelerometers for iNPH management: (1) diagnosing abnormal gait patterns in iNPH patients, (2) smartphone-based accelerometry, (3) prolonged gait monitoring in real-world environments, (4) prognostication of fall risk, and (5) prognostication of response to shunt placement.

Figure 1.

Prisma flow Chart.

Figure 2.

Portable and software-based accelerometers and findings in iNPH Fig 2. Line diagram created in BioRender by Martin R (https://www.biorender.com).

Using Accelerometers to Diagnose Abnormal Gait Patterns in iNPH Patients

Eight studies looked at the use of portable accelerometers to diagnose iNPH and/or distinguish abnormal gait patterns in this condition. Out of these eight articles, four were proof-of-concept studies,^16-19 two were cross-sectional studies,^20,21 one was a case-control study,²² and one was a prospective cohort study.²³ All studies used different accelerometers to measure gait parameters (Table 1).

Table 1.

Studies Using Accelerometers to Diagnose Abnormal Gait Patterns in iNPH Patients.

Reference	Objective	Study design	Device used	Location	Major findings
Gaglani et al. (2015)²⁰	To assess whether activity monitors can be used to accurately measure gait patterns in patients with confirmed iNPH	Cross-sectional	Omron step Counter HJ-113, new lifestyles 2000, Nike Fuelband, Fitbit Ultra	Wrist	Fitbit Ultra was the most accurate monitor to assess step count in iNPH patients
Panciani et al. (2018)²²	To test inertial sensor performance in quantifying impairment in gait parameters in iNPH	Case-control	G-walk, BTS Bioengineering	Hip	Inertial sensor system is reliable and easy-to-use for early detection of iNPH in any clinical setting
Bäcklund et al. (2020)¹⁶	To validate an in-house-developed gait analysis system, based on optical and inertial sensors and a novel method for stride detection, for measuring step-width during the swing phase of gait and temporal parameters in iNPH	Proof-of-concept	Striton	Leg	Striton is a reliable wearable system for quantifying step width and temporal gait parameters, which differ between elderly individuals and iNPH patients, as well as before and after iNPH surgery
Atbas et al. (2024) ¹⁷	To introduce a wearable device that can be used in aiding the diagnosis of iNPH	Proof-of-concept	Movella DOT	Wrist	Wearable sensors are reliable and accurate at measuring gait parameters
Taishaku et al. (2024)²¹	To identify gait characteristics including whole body information, related to iNPH, PD, and cervical myelopathy (CM), specifically, the relationship between motion of the upper and lower limbs	Cross-sectional study	Three-Dimensional Pose Tracker for gait test (TDPT-GT) app	Smartphone	The strongest upper and lower limb correlations occurred in hip and elbow angles on the sagittal plane, with reduced swings in iNPH and PD patients.
Magni et al. (2025)¹⁸	To develop a computational pipeline to extract gait cycle parameters from wearable sensors including accelerometer, gyroscope, and foot pressure sensors embedded in shoes	Proof-of-concept	IEE’s ActiSense	Insole	Combining wearable sensors with ML algorithms trained on gait features shows promise for objectively distinguishing gait patterns between PD and NPH patients
Rydja et al (2025)²³	To quantitatively evaluate gait in iNPH patients before and after shunt surgery and compare it to healthy individuals (HI) using an inertial sensor system	Prospective cohort study	RehaGait®	Legs	Patients demonstrated reduced hip flexion, heel strike, and toe-off angles, along with shorter strides, slower velocity, and prolonged swing and single-support phases compared to HI
Lee et al. (2025)¹⁹	To propose a deep neural network (DNN) model to classify disease-specific gait patterns in older adults using commercialized instrumented insoles	Proof-of-concept	GDCA-MD smart insoles	Insole	The proposed DNN model can reliably classify gait abnormalities in older adults using simple instrumented insoles

Using Smartphone-Based Accelerometry as a Diagnostic Tool

Nine studies looked at the use of smartphone-based accelerometers as a tool to diagnose conditions with anomalous gait patterns including iNPH. Three out of these nine studies were proof-concept studies,^24-26 two were case control,^27,28 other two were prospective cohort studies,^29,30 one was a case-series,³¹ and another one was a cross-sectional study.³² To measure their outcomes four studies used the iOS application Three-Dimension Pose Tracker (TDPT), three studies used the iPhone-Senior Quality App, and the other two remaining applications used were iPhone-QS Access app and iPhone-Watkins and Watkins 2.0 apps (Table 2).

Table 2.

Studies Using Smartphone-Based Accelerometry as a Diagnostic Tool.

Reference	Objective	Study design	Device used	Location	Major findings
Ishikawa et al. (2019)²⁷	To determine whether smartphone data can reliably be used to assess gait metrics during a TUG test	Case-control	iPhone-Senior quality app	Smartphone	Smartphones are useful gait-measuring tools in subjects who perform a TUG test
Yamada et al. (2019)²⁹	To determine whether iTUG (measured via smartphones) could accurately quantify gait parameters in iNPH patients before and after CSF-TT	Prospective cohort	iPhone-Senior quality app	Smartphone	iTUG data recorded on iPhones provide a fast, reliable assessment of gait in iNPH patients
Yamada et al. (2021)³⁰	To determine whether 3-D acceleration of the trunk could be quantitatively measured to assess abnormal gait in iNPH patients	Prospective cohort	iPhone-Senior quality app	Smartphone	Reduced fluctuation of forward and vertical trunk acceleration is associated with pathologic gait in iNPH patients
Aoyagi et at. (2022)²⁴	To quantitatively assess pathological gait with a novel smartphone application for full-body human motion tracking in real time from video-based images using a smartphone monocular camera and deep learning	Proof-of-concept	iOS application TDPT-GT	Smartphone	Application could reconstruct the full-body human motion efficiently in real time and perform 3D gait assessments quantitatively from a 2D video from a monocular iPhone camera
Sprau et al. (2021)³¹	To use Apple Health data to estimate activity levels in two iNPH patients before and after shunt placement	Case series	iPhone-QS Access app (exports Apple Health data)	Smartphone	Smartphone-based accelerometry can provide reliable data to assess long-term response to shunt placement in iNPH patients
Iseki et al. (2023)²⁵	To obtain a convenient recording with an iPhone and establish an algorithm based on deep learning to distinguish pathological gait patterns	Proof-of-concept	iOS application TDPT-GT	Smartphone	The pathological gait captured by the iPhone could be distinguished by artificial intelligence
Iseki et al. (2023)²⁶	To capture fluctuations in body dynamics and identify their characteristics in patients with iNPH and PD.	Proof-of- concept	iOS application TDPT-GT	Smartphone	The TDPT-GT system measures whole-body movement and temporal variations during walking, with impaired upper- and lower-body fluctuations potentially contributing to gait and balance disorders in patients
Yamada et al. (2023)³²	To assess pathological gaits quantitatively with three-dimensional coordinates estimated with a deep learning model	Cross-sectional study	iOS application TDPT-GT	Smartphone	Two-dimensional body-axis coordinates derived from 3D relative coordinates estimated by the TDPT-GT application enabled quantification of pathological gait features
Tariq et al. (2024)²⁸	To validate a smart phone app capable of measuring gait parameters in patients with iNPH	Case-control	iPhone-Watkins and Watkins 2.0 apps	Smartphone	The app presented is capable of measuring gait parameters in patients with iNPH with high accuracy as compared to clinical observers

Prolonged Gait Monitoring in a Real-World Environment

Three prospective cohort studies focused on investigating prolonged gait monitoring of iNPH in real-world environment.^13,14,33 Two studies used the ZurichMOVE sensor, while another one used the ActiGraph GT3x+ (Table 3).

Table 3.

Studies Looking at Prolonged Gait Monitoring in a Real-World Environment.

Reference	Objective	Study design	Device used	Location	Major findings
Kuruvithadam et al. (2021)¹³	To assess whether prolonged, real-world measurements of gait are useful in identifying movement disorders	Prospective cohort	ZurichMOVE	Wrist	Short-term gait measurement in the lab differs significantly from long-term measurement in real world settings
Dias et al. (2023)¹⁴	To determine whether prolonged gait assessment at home could accurately identify abnormal gait associated with iNPH	Prospective cohort	ZurichMOVE	Wrist	iNPH patients who respond favorably to VP shunt placement have different gait patterns than non-responders
Jusué-Torres et al. (2023)³³	To determine whether activity monitors could objectively measure gait improvement in iNPH patients following VP shunt placement	Prospective cohort	ActiGraph GT3x+	Hip/belt	Activity monitoring provides objective data on gait entropy and may have utility in long-term follow-up of iNPH patients

Using Accelerometry to Prognosticate Fall Risk

Two studies used accelerometry to determine the risk of falls in patients with gait disorders including iNPH. One of the articles a cross-sectional study using MG-M1100-HW gait analyzer, while the other one was a prospective cohort study using SparkFun IMU, ARM Cortex-M4 processor (Table 4).^34,35

Table 4.

Studies Using Accelerometry to Prognosticate Fall Risk.

Reference	Objective	Study design	Device used	Location	Major findings
Nikaido et al. (2019)³⁴	To determine whether gait parameters could be used to predict future fall risk in iNPH patients	Cross-sectional	MG-M1100-HW gait analyzer	Wrist	Gait variability and poor balance function were strongly associated with fall risk
Nouriani et al. (2023)³⁵	To create an algorithm to predict fall risk in patients with movement disorders (PD, iNPH)	Prospective cohort	SparkFun IMU, ARM Cortex-M4 processor	Trunk/legs	Created an algorithm that predicts high risk fall behaviors w/>95% sensitivity, near falls w/>80% sensitivity

Using Accelerometry to Prognosticate Response to Shunt Placement

Ten studies looked at the use of accelerometry to predict response to shunt placement in iNPH. Six articles were cohort studies, five being prospective in design while one was retrospective.^36-42 The remaining four articles were a case-control study, a technical note, a cross-sectional study, and a case report.^39,43-45 Umit et al.⁴² and Mostile et al.⁴⁴ used the G-Walk, BTS Bioengineering as their accelerometer. He et al.³⁸ and Öztop-Cakmak et al.⁴¹ used the ADPM IMU sensor device. Two different papers by Ferrari et al.^37,46 used the mGait system (mHealth Technologies). The remaining four studies used different accelerometers (Table 5).

Table 5.

Studies Using Accelerometry to Prognosticate Response to Shunt Placement.

Reference	Objective	Study design	Device used	Location	Major findings
Yang et al., (2016)³⁶	To accurately identify iNPH patients who will improve after shunt placement using a wearable, wireless, computerized gait analysis device	Prospective cohort	LegSys, Biosensics	Legs	Computerized gait analysis with lumbar drainage predicted shunt responsiveness in 100% of patients and degree of improvement in 96%
Nikaido et al. (2018)⁴³	To use lower extremity accelerometers to measure gait parameters in iNPH patients before and after CSF-TT/shunt placement vs healthy controls	Case-control	MG-M1100-HW gait analyzer	Wrist	iNPH patients exhibited decreased gait velocity and increased step number/time compared to controls
Ferrari et al. (2020)³⁷	To determine which gait parameters are most strongly correlated with motor activity before and after CSF-TT in iNPH patients with and without vascular encephalopathy	Prospective cohort	mGait system (mHealth technologies)	Hip/feet	iNPH patients with and without vascular encephalopathy showed gait improvement following CSF-TT
He et al. (2021)³⁸	To quantify various gait parameters after extended lumbar drainage placement using APDM sensors	Retrospective cohort	ADPM IMU sensor	Wrist/ankle/feet	APDM sensors provide accurate gait measurements and may be useful in determining patients suitable for surgery
Gago et al. (2022)³⁹	Identify gait features predictive of best response to shunt surgery in iNPH.	Technical note	Physilog® sensors (gait Up®, Switzerland)	Feet	Magnitude of gait improvement after CSF-TT, quantified by gait analysis, can be used as an integral variable in the multimodal clinical approach to the prediction of improvement after shunt surgery.
Ferrari et al. (2022)⁴⁶	To assess whether gait parameters measured by motion sensors could predict favorable responsiveness to shunt placement	Prospective cohort	mGait system (mHealth technologies)	Hip/feet	iNPH patients showed significant gait improvement after shunt placement, with key gait variables predicting positive shunt response.
Mostile et al. (2022)⁴⁴	To determine whether patients with iNPH and untreated PD differed in various gait metrics following a timed TUG test	Cross-sectional	G-walk, BTS Bioengineering	Hip	iNPH patients had slower movement and took longer to complete the TUG test compared to untreated PD
Öztop-Cakmak et al. (2023)⁴¹	To observe whether APDM sensors could be used to detect gait changes in iNPH patients following CSF drainage (CSF-TT or shunt surgery)	Prospective cohort	ADPM IMU sensor	Wrist/ankle/feet	iNPH patients demonstrated significantly increased cadence and decreased step time following CSF drainage
Umit et al. (2024)⁴²	To evaluate the effects of CSF-TT procedures on gait, turning, and balance parameters in the first 24 hours following the procedure in patients.	Prospective cohort	G-walk, BTS Bioengineering	Hip	CSF-TT has positive effects on walking, turning activities, balance, and risk of falling in older patients with iNPH in the first 24 hours
Lee et al. (2025)⁴⁵	Use smart insoles to assist in clinical decision making for a patient with iNPH	Case report	GDCA-MD smart insoles	Insole	The smart insoles were able to identify changes in gait parameters pre- and post-CSF drainage that were not identified with conventional methods

Discussion

The purpose of this review was to identify the various means by which portable accelerometers can be used to quantitatively assess gait in iNPH patients. Characterized by broad-based walking, decreased speed, step-length variability, hesitancy, and postural instability, gait is usually the first symptom that patients with iNPH present. Assessing gait in these patients is important not only for differential diagnosis, but also for determining shunt placement eligibility, establishing a baseline for monitoring, and facilitating long-term follow-up of symptom progression. In the past, gait could only be assessed subjectively in a clinical setting, but with the advent of accelerometry, gait can now be objectively and quantitatively measured. However, it was not until the early 2010s that portable accelerometers began to be used in clinical research to diagnose iNPH by measuring gait parameters such as speed, stride length, and turning speed.^22,47 Due to their ubiquity, portability, and ability to measure multiple parameters simultaneously, accelerometers provide a quick and accessible means to evaluate gait in iNPH patients. In addition, they overcome the limitations of poor inter-rater reliability through objective assessment.

This review identified 32 studies that used a variety of portable accelerometers to quantify gait parameters in patients with iNPH. In total, 22 different accelerometers were used across studies, ranging from wearable wristbands to smartphone-based applications. Among the smartphone applications, the TDPT-GT App was the most widely used, implemented in five studies.^21,24-26,48 This application uses a smartphone camera to capture motion by generating 30 Hz coordinates for 27 body points, which are then analyzed for iNPH gait features such as shuffling, short steps, and wide-based walking patterns.²⁶ Among the wearable accelerometers, the G-Walk device from BTS Bioengineering was the most commonly used, reported in three studies^22,42,44 Unlike software-based accelerometers, the G-Walk device is an inertial sensor capable of recording 3D accelerometric and gyroscopic data, most often used before and after diagnostic interventions such as the CSF-TT.⁴⁹ However, recent quantitative gait-analysis studies suggest that gait improvement following CSF-TT may not manifest within the first hours as traditionally assumed, but can emerge gradually over a 48-72-hour window. In this context, wearable accelerometry offers an opportunity for more ecologically valid, real-world monitoring that can capture delayed or progressive gait responses that may be missed during brief in-clinic assessments.^37,50 This suggests that software-based accelerometers could be implemented more regularly in real-life scenarios, where patients do not necessarily need to be in a clinical setting. In contrast, wearable accelerometers with inertial sensor technology may be better suited for clinical environments, where the use of a device other than a smartphone is required and where a greater technological learning curve may be involved.

More specific uses of portable accelerometers in iNPH were mentioned by Sprau et al. (2021), who stated that remote monitoring of patients allows for the capture of unbiased data for the detection of improvements or setbacks. This is particularly relevant for patients following shunt surgery, where clinical gait improvement can typically be maintained for about 5-7 years.⁵¹ Activity monitoring data could be used in these instances to detect changes from a post-shunt baseline to the early stages of symptom recurrence, thus allowing physicians to intervene earlier through revision surgeries and potentially prevent disease regression. Remote monitoring also enables physicians to track patient symptoms without requiring in-person follow-up when gait metrics remain stable.³¹

Currently, portable accelerometers are more widely used in other diseases that affect gait, such as Parkinson’s disease (PD) and stroke. Five studies included in this review used accelerometers to compare iNPH symptoms with those of PD and other gait disorders, as well as to train accelerometer algorithms for diagnosis.^{18,21,26,35,44} Magni et al. combined wearable accelerometers with a machine learning (ML) algorithm to differentiate between PD and iNPH. This also reflects the increasing number of studies investigating the ability of machine and deep learning models to diagnose various neurological conditions that affect gait.^52-54 For example, Iseki et al. developed a deep learning model capable of detecting iNPH using only an iPhone camera.²⁵ The rapid advancement of such technologies has the potential to make diagnosis of these conditions more accurate and precise from the earliest signs of symptoms.

Nevertheless, a possible limitation of portable and software-based accelerometers is the learning curve and comfort associated with wearable sensors, particularly in older patients who represent the primary population affected by iNPH. Furthermore, because one of the hallmark symptoms of iNPH is dementia, many patients may struggle to independently set up or use software and wearable accelerometers. As a result, independent use of these technologies may only be feasible during periods of symptom resolution, or they may instead need to be implemented as preventive strategies with caregiver or clinician assistance.

Limitations

The current body of evidence is limited by the significant heterogeneity across studies. This includes variability in accelerometer models, study populations, and the specific gait parameters measured by each study. This heterogeneity not only makes direct comparison challenging but prevents a formal meta-analysis from being conducted. Establishing a core set of gait parameters, such as gait velocity, stride length, cadence, variability, and symmetry, that should be consistently reported would further facilitate meaningful cross-study synthesis. Adopting a replicable assessment protocol, including standardized walk distance, pace instructions, sensor placement, and recording duration, would help reduce methodological heterogeneity. Incorporating these elements into future studies would substantially strengthen the evidence base and support more reliable clinical integration of accelerometer-derived gait metrics in iNPH. Additionally, most studies also had a relatively small sample size and short follow-up periods, which limit the generalizability of findings. Another limitation is the lack of standardized protocols for data collection and analysis, which introduces bias and reduces reproducibility. Future studies should focus on standardizing their outcome by identifying and measuring specific gait parameters. Larger, multicenter trials with longer follow-up would also strengthen the body of literature.

Conclusion

This review identified 32 studies utilizing 22 different portable accelerometers to quantify gait parameters in patients with iNPH, ranging from wearable wristbands to smartphone-based applications. These devices provide objective and reproducible measures of gait, overcoming the limitations of subjective clinical assessments and poor inter-rater reliability. Remote monitoring with accelerometers not only facilitates early detection of symptom progression following shunt surgery but also reduces the burden of in-person follow-up. Furthermore, accelerometer-based assessments are increasingly being integrated with machine learning and deep learning models, which have shown promise in differentiating iNPH from other movement disorders such as Parkinson’s disease. While these technologies represent a significant advance, further studies must be conducted to standardize the outcomes obtained by these devices and applications to make a diagnosis. Overall, portable accelerometers and smartphone-based applications can represent a practical and accessible tool for the diagnosis, monitoring, and long-term management of iNPH.

Supplemental Material

Supplemental Material - Emerging Role of Portable Accelerometers for Assessment of Gait Instability in Idiopathic Normal Pressure Hydrocephalus: A Systematic Review

Supplemental Material for Emerging Role of Portable Accelerometers for Assessment of Gait Instability in Idiopathic Normal Pressure Hydrocephalus: A Systematic Review by Ryan J. Davis, MS, Emanuella M. Brito, Julia Armstrong, Roberto Martin, Jehan Bista, Bryan A. Lieber, Regina M. Graham in Journal of Geriatric Psychiatry and Neurology

Supplemental Material

Supplemental Material - Emerging Role of Portable Accelerometers for Assessment of Gait Instability in Idiopathic Normal Pressure Hydrocephalus: A Systematic Review

Footnotes

Author Contributions

The study was conceptualized and supervised by Bryan Lieber. Data acquisition and initial writing were done by Emanuella Brito, Ryan Davis, Julia Armstrong, and Roberto Martin. Review and revision were done by Jehan Bista, Regina M. Graham. All authors read and approved the final manuscript.

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by HCA Florida University Hospital for RMG.

ORCID iDs

Ryan J. Davis

Roberto Martin

Supplemental Material

Supplemental material for this article is available online.

References

Wang

Zhang

Ding

Wang

. Pathogenesis and pathophysiology of idiopathic normal pressure hydrocephalus. CNS Neurosci Ther. 2020;26(12):1230-1240. doi:10.1111/cns.13526

Zaccaria

Bacigalupo

Gervasi

, et al. A systematic review on the epidemiology of normal pressure hydrocephalus. Acta Neurol Scand. 2020;141(2):101-114. doi:10.1111/ane.13182

Relkin

Marmarou

Klinge

Bergsneider

Black

. Diagnosing idiopathic normal-pressure hydrocephalus. Neurosurgery. 2005;57(3 Suppl):S4-S16. doi:10.1227/01.neu.0000168185.29659.c5. discussion ii-v.

Nakajima

Yamada

Miyajima

, et al. Guidelines for management of idiopathic normal pressure hydrocephalus (Third Edition): endorsed by the Japanese society of normal pressure hydrocephalus. Neurol Med -Chir. 2021;61(2):63-97. doi:10.2176/nmc.st.2020-0292.61.

Yahalom

Yekutieli

Israeli-Korn

, et al. Smartphone based timed up and go test can identify postural instability in Parkinson’s Disease. Isr Med Assoc J. 2020;22(1):37-42.

Halperin

Kurlan

Schwalb

Cusimano

Gronseth

Gloss

. Practice guideline: idiopathic normal pressure hydrocephalus: response to shunting and predictors of response: report of the guideline development, dissemination, and implementation subcommittee of the American academy of neurology. Neurology. 2015;85(23):2063-2071. doi:10.1212/WNL.0000000000002193

El Ahmadieh

Kafka

, et al. Clinical outcomes of normal pressure hydrocephalus in 116 patients: objective versus subjective assessment. J Neurosurg. 2020;132(6):1757-1763. doi:10.3171/2019.1.JNS181598

Shema-Shiratzky

Beer

Mor

Elbaz

. Smartphone-based inertial sensors technology - validation of a new application to measure spatiotemporal gait metrics. Gait Posture. 2022;93:102-106. doi:10.1016/j.gaitpost.2022.01.024

Shahar

Agmon

. Gait analysis using accelerometry data from a single smartphone: agreement and consistency between a smartphone application and gold-standard gait analysis System. Sensors (Basel). 2021;21(22):7497. doi:10.3390/s21227497

10.

Ellis

Zhu

, et al. A validated smartphone-based assessment of gait and gait variability in Parkinson’s Disease. PLoS One. 2015;10(10):e0141694. doi:10.1371/journal.pone.0141694

11.

Schalkamp

Peall

Harrison

Sandor

. Wearable movement-tracking data identify Parkinson’s disease years before clinical diagnosis. Nat Med. 2023;29(8):2048-2056. doi:10.1038/s41591-023-02440-2

12.

Cakmak

Akar

Youssef

Samanci

Ertan

Vural

. Comparative assessment of gait and balance in patients with Parkinson’s Disease and normal pressure hydrocephalus. Sisli Etfal Hastan Tıp Bul. 2023;57(2):232-237. doi:10.1474/semb.2023.79990

13.

Kuruvithadam

Menner

Taylor

Zeilinger

Stieglitz

Schmid Daners

. Data-driven investigation of gait patterns in individuals affected by normal pressure hydrocephalus. Sensors (Basel). 2021;21(19):6451. doi:10.3390/s21196451

14.

Dias

Graf

Jehli

, et al. Gait pattern analysis in the home environment as a key factor for the reliable assessment of shunt responsiveness in patients with idiopathic normal pressure hydrocephalus. Front Neurol. 2023;14:1126298. doi:10.3389/fneur.2023.1126298

15.

Page

Moher

Bossuyt

, et al. PRISMA 2020 explanation and elaboration: updated guidance and exemplars for reporting systematic reviews. BMJ. 2021;372:n160. doi:10.1136/bmj.n160

16.

Bäcklund

Öhberg

Johansson

Grip

Sundström

. Novel, clinically applicable method to measure step-width during the swing phase of gait. Physiol Meas. 2020;41(6):065005. doi:10.1088/1361-6579/ab95ed

17.

Atbas

Gaydecki

Callaghan

. A wearable gait-analysis device for idiopathic normal-pressure hydrocephalus (INPH) monitoring. Article. Biomedical Physics and Engineering Express. 2024;10(6):2. doi:10.1088/2057-1976/ad2a1a

18.

Magni

Bremm

Verros

, et al. Machine learning differentiation of Parkinson’s disease and normal pressure hydrocephalus using wearable sensors capturing gait impairments. medRxiv. 2025;24318198:1-36. doi:10.1101/2025.01.07.24318198.

19.

Lee

Hyun

Choi

, et al. Distinguishing pathologic gait in older adults using instrumented insoles and deep neural networks. IEEE J Biomed Health Inform. 2025;29(7):4758-4768. doi:10.1109/JBHI.2025.3549454. Article in Press.

20.

Gaglani

Moore

Haynes

Hoffberger

Rigamonti

. Using commercial activity monitors to measure gait in patients with suspected iNPH: implications for ambulatory monitoring. Cureus. 2015;7(11):e382. doi:10.7759/cureus.382

21.

Taishaku

Yamada

Iseki

, et al. Development of a gait analysis application for assessing upper and lower limb movements to detect pathological gait. Sensors (Basel). 2024;24(19):6329. doi:10.3390/s24196329

22.

Panciani

Migliorati

Muratori

Gelmini

Padovani

Fontanella

. Computerized gait analysis with inertial sensor in the management of idiopathic normal pressure hydrocephalus. Eur J Phys Rehabil Med. 2018;54(5):724-729. doi:10.23736/s1973-9087.18.04949-3

23.

Rydja

Pohl

Eleftheriou

Lundin

. Gait characteristics in idiopathic normal pressure hydrocephalus: a controlled study using an inertial sensor system. Article. PLoS One. 2025;20(2 February):e0317901. doi:10.1371/journal.pone.0317901

24.

Aoyagi

Yamada

Ueda

, et al. Development of smartphone application for markerless three-dimensional motion capture based on deep learning model. Sensors (Basel). 2022;22(14):5282. doi:10.3390/s22145282

25.

Iseki

Hayasaka

Yanagawa

, et al. Artificial intelligence distinguishes pathological gait: the analysis of markerless motion capture gait data acquired by an iOS application (TDPT-GT). Sensors (Basel). 2023;23(13):6217. doi:10.3390/s23136217

26.

Iseki

Suzuki

Fukami

, et al. Fluctuations in upper and lower body movement during walking in normal pressure hydrocephalus and Parkinson’s Disease assessed by motion capture with a smartphone application, TDPT-GT. Sensors (Basel). 2023;23(22):9263. doi:10.3390/s23229263

27.

Ishikawa

Yamada

Yamamoto

Aoyagi

. Gait analysis in a component timed-up-and-go test using a smartphone application. Article. J Neurol Sci. 2019;398:45-49. doi:10.1016/j.jns.2019.01.023

28.

Tariq

Thorne

Toma

Watkins

. ‘Watkins’ & ‘Watkins 2.0’: smart phone applications (Apps) for gait-assessment in normal pressure hydrocephalus and decompensated long-standing overt ventriculomegaly. Article. Acta Neurochir. 2024;166(1):386. doi:10.1007/s00701-024-06275-9

29.

Yamada

Aoyagi

Yamamoto

Ishikawa

. Quantitative evaluation of gait disturbance on an instrumented timed up-and-go test. Aging Dis. 2019;10(1):23-36. doi:10.14336/AD.2018.0426

30.

Yamada

Aoyagi

Ishikawa

Yamaguchi

Yamamoto

Nozaki

. Gait assessment using three-dimensional acceleration of the trunk in idiopathic normal pressure hydrocephalus. Front Aging Neurosci. 2021;13:653964. doi:10.3389/fnagi.2021.653964

31.

Sprau

Basil

Eliahu

, et al. Using smartphone-based accelerometers to gauge postoperative outcomes in patients with NPH: implications for ambulatory monitoring. Surg Neurol Int. 2021;12:464. doi:10.25259/sni_112_2021

32.

Yamada

Ishikawa

Itou

, et al. Importance of oscillating flow of cerebrospinal fluid in idiopathic normal pressure hydrocephalus. Conference abstract. Fluids Barriers CNS. 2023;19:1. doi:10.1186/s12987-022-00397-x

33.

Jusue-Torres

Brown

Pennington

, et al. Objective assessment of patients with idiopathic normal pressure hydrocephalus following ventriculoperitoneal shunt placement using activity-monitoring data: pilot study. Neurosurg Focus. 2023;54(4):E6. doi:10.3171/2023.1.Focus22640

34.

Nikaido

Urakami

Akisue

, et al. Associations among falls, gait variability, and balance function in idiopathic normal pressure hydrocephalus. Clin Neurol Neurosurg. 2019;183:105385. doi:10.1016/j.clineuro.2019.105385

35.

Nouriani

Jonason

Sabal

, et al. Real world validation of activity recognition algorithm and development of novel behavioral biomarkers of falls in aged control and movement disorder patients. Front Aging Neurosci. 2023;15:1117802. doi:10.3389/fnagi.2023.1117802

36.

Yang

Hickman

Tinl

, et al. Quantitative evaluation of changes in gait after extended cerebrospinal fluid drainage for normal pressure hydrocephalus. J Clin Neurosci. 2016;28:31-37. doi:10.1016/j.jocn.2015.11.013

37.

Ferrari

Milletti

Giannini

, et al. The effects of cerebrospinal fluid tap-test on idiopathic normal pressure hydrocephalus: An inertial sensors based assessment. J NeuroEng Rehabil. 2020;17(1):7. doi:10.1186/s12984-019-0638-1

38.

Shao

, et al. Quantitative evaluation of gait changes using APDM inertial sensors after the external lumbar drain in patients with idiopathic normal pressure hydrocephalus. Front Neurol. 2021;12:635044. doi:10.3389/fneur.2021.635044

39.

Gago

Ferreira

Bicho

. Quantitative gait analysis value as a predictor of shunt surgery effectiveness in normal pressure hydrocephalus: a technical note. Clin Neurol Neurosurg. 2022;221:107405. doi:10.1016/j.clineuro.2022.107405

40.

Ferrari

Milletti

Palumbo

, et al. Gait apraxia evaluation in normal pressure hydrocephalus using inertial sensors. Clinical correlates, ventriculoperitoneal shunt outcomes, and tap-test predictive capacity. Fluids Barriers CNS. 2022;19(1):51. doi:10.1186/s12987-022-00350-y

41.

Öztop-Çakmak

Akar

Youssef

, et al. Gait analysis in idiopathic normal pressure hydrocephalus: a single Centre experience. Istanbul Medical Journal. 2023;24(1):36-39. doi:10.4274/imj.galenos.2023.60251

42.

Umit

Nihal

Derya

İlke

Ahmet Turan

. The effects of a cerebrospinal fluid tap test on gait, balance and turning in elderly patients with idiopathic normal pressure hydrocephalus. Int J Gerontol. 2024;18(3):173-178. doi:10.6890/IJGE.202407_18(3).0007

43.

Nikaido

Akisue

Kajimoto

, et al. The effect of CSF drainage on ambulatory center of mass movement in idiopathic normal pressure hydrocephalus. Article. Gait Posture. 2018;63:5-9. doi:10.1016/j.gaitpost.2018.04.024

44.

Mostile

Contrafatto

Terranova

, et al. Turning and sitting in early parkinsonism: differences between idiopathic normal pressure hydrocephalus associated with Parkinsonism and Parkinson’s Disease. Mov Disord Clin Pract. 2023;10(3):466-471. doi:10.1002/mdc3.13638

45.

Lee

Choi

S-I

Chung

Hwang

Kim

. Serial assessment of gait changes after interventions using smart insole in a patient with iNPH: a proof-of-concept case report. Brain & Neuro Rehabilitation. 2025;18(1):e1. doi:10.12786/bn.2025.18.e1

46.

Ferrari

Milletti

Palumbo

47.

Passaretti

Maranzano

Bluett

Rajalingam

Fasano

. Gait analysis in idiopathic normal pressure hydrocephalus: a meta-analysis. Mov Disord Clin Pract. 2023;10(11):1574-1584. doi:10.1002/mdc3.13816

48.

Yamada

Aoyagi

Iseki

, et al. Quantitative gait feature assessment on two-dimensional body axis projection planes converted from three-dimensional coordinates estimated with a deep learning smartphone App. Sensors. 2023;23(2):617. doi:10.3390/s23020617

49.

Lim

Y-H

P-W

Park

K-S

, et al. Quantitative gait analysis and cerebrospinal fluid tap test for idiopathic normal-pressure hydrocephalus. Sci Rep. 2019;9(1):16255. doi:10.1038/s41598-019-52448-3

50.

Schniepp

Trabold

Romagna

, et al. Walking assessment after lumbar puncture in normal-pressure hydrocephalus: a delayed improvement over 3 days. J Neurosurg. 2017;126(1):148-157. doi:10.3171/2015.12.Jns151663

51.

Pujari

Kharkar

Metellus

Shuck

Williams

Rigamonti

. Normal pressure hydrocephalus: long-term outcome after shunt surgery. J Neurol Neurosurg Psychiatry. 2008;79(11):1282-1286. doi:10.1136/jnnp.2007.123620

52.

Zadka

Rabin

Gazit

, et al. A wearable sensor and machine learning estimate step length in older adults and patients with neurological disorders. npj Digit Med. 2024;7(1):142. doi:10.1038/s41746-024-01136-2

53.

Park

Mun

Kim

. Development of machine learning models for gait-based classification of incomplete spinal cord injuries and cauda Equina syndrome. Sci Rep. 2025;15(1):20012. doi:10.1038/s41598-025-04065-6

54.

Sarkar

. Integrating machine learning and deep learning techniques for advanced Alzheimer’s Disease detection through gait analysis. J Bus Manag Stud. 2025;7(1):140-147. doi:10.32996/jbms.2025.7.1.8

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.10 MB

2.15 MB