{"paper_id":"3dbfd97f-c9bd-4002-a9f5-009f73641ccf","body_text":"Clinical validity and reliability of the pose estimation-based system to determine spatio-temporal gait parameters in older adults: A pilot study | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Clinical validity and reliability of the pose estimation-based system to determine spatio-temporal gait parameters in older adults: A pilot study Gamze Yalcinkaya Colak, Umut Can Colak, İlke Kara Oz, Punhan Sailov, and 1 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7987537/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Background: The spatio-temporal gait characteristics reflect physical function, fall risk, and rehabilitation processes in older adults. Recent advancements in artificial intelligence have led to the development of more practically accessible gait analysis systems. The purpose of this study was to examine clinical validity and test-retest reliability of the YOLOv11 algorithm-based system for the assessment of spatio-temporal gait parameters in community-dwelling older adults. Methods: Thirty-one community-dwelling older adults (mean Age: 66.94 years, mean BMI: 26.7) were recruited for the study. Participants were instrumented with a 4-meter gait test using the G-walk system, and video recording simultaneously. The collected videos were processed using the YOLOv11 algorithm to extract the spatio-temporal gait parameters. The consistency of the two methods was analysed using the Pearson correlation coefficient ( r ), and agreement was assessed using Bland-Altman Plots. The test-retest reliability was assessed using intra-class correlation coefficients (ICCs). Results: Primary findings revealed that the YOLOv11 algorithm-based system exhibited strong to excellent consistency with the G-walk system for gait parameters such as cadence, gait speed, and stride length, gait cycle duration, and gait phases distribution ( r =0.60-0.95; p<0.001). Moderate agreement was observed between the two systems for all spatio-temporal variables, except for the first double support, which showed poor agreement. Additionally, reliability was excellent for cadence (ICC=0.92) and gait speed (ICC=0.91), good for gait cycle duration (ICC=0.89) and stride length (ICC=0.90), and moderate for gait phase distributions (ICC=0.59-0.61). Conclusions: The findings of this study suggest that the YOLOv11 algorithm-based system can be a potential alternative for gait analysis in older adults, providing a more accessible and affordable option for clinicians and researchers. Future research is warranted due to the cautious interpretation of stance, swing, and single/double support phases. Trial registration: The study protocol was registered with ClinicalTrials.gov (identifier number: NCT07119944). spatio-temporal gait parameters human pose estimation older adults validity reliability Figures Figure 1 Figure 2 Figure 3 Background Worldwide, people aged 60 and over numbered 962 million in 2017, and it is expected that this number will reach 2.1 billion by 2050 ( 1 , 2 ). As societies age, the implications for public health, healthcare systems, and economies become increasingly complex. In light of the global aging phenomenon, the World Health Organization (WHO) emphasizes the determinants of healthy aging, including integrated healthcare models and consistent monitoring of intrinsic capacity ( 2 , 3 ). The notion of intrinsic capacity covers five domains: locomotor, cognitive, psychological, sensory, and vitality. Within this framework, gait of older adults emerges as an essential element of locomotion, and gait speed is recognized as the sixth vital sign due to its high predictive value for mortality, dementia, frailty, and other negative health outcomes associated with aging ( 4 , 5 ). Additionally, gait parameters are the primary target in geriatric rehabilitation to protect functional independence ( 6 , 7 ). In objective gait analysis, floor sensors, wearable sensors, and image processing are generally reliable approaches for determining spatiotemporal gait characteristics ( 8 ). Specifically, the lab-based optical motion capture systems such as the GAITrite system, were considered the gold standard over the last decade and has been studied in both healthy and various populations with Parkinson’s disease, musculoskeletal problems, neurodegenerative diseases, etc. ( 9 ). However, the initial cost is quite high, and analysis requires a specific laboratory environment with time-consuming measurements. On the other hand, wearable inertial measurement unit (IMU)-based systems were also found to be valid and reliable in defining spatio-temporal gait characteristics ( 10 , 11 ). Systematic reviews and validation studies demonstrated moderate agreement between single IMU-based systems and the GAITrite system ( 10 – 13 ). Viteckova et. al found that the wearable G-walk system is valid and reliable in healthy adults and patients with Parkinson’s disease ( 12 ). In parallel, a single 3D-accelerometer-based gait analysis procedure was also found to be reliable and correlated with fall risk in older adults ( 14 ). At the National Institute of Aging workshop, Boyer et al. highlighted a significant limitation in gait research among older adults, the underrepresentation of individuals with slowed gait speed and incident mobility limitations, who are prevalent in the community but seldom included in traditional laboratory studies. ( 7 ). The authors identified knowledge gaps for future studies regarding the real-world measurement of gait in older adults ( 7 ). In this manner, remote and minimalistic gait monitoring can provide a broader solution for both older adults and health practitioners. Recent advancements in artificial intelligence have enabled to the development of practically applicable gait analysis systems using pose estimation algorithms ( 15 – 20 ). Stenum et al. reported significant agreement between video-based gait analysis using the OpenPose algorithm and 3D motion capture systems in determining spatio-temporal gait parameters ( 19 ). In parallel, Mehdizadeh et al. evaluated the concurrent validity of three pose estimation algorithms (AlphaPose, OpenPose, and Detectron)-based gait analysis and showed that AlphaPose and OpenPose had the highest correlations with a 3D motion capture system for detecting step time and cadence in eleven older adults ( 18 ). Further, Ripic et al. showed excellent agreement with another algorithm from the Kinatrax markerless system and 3D motion analysis in determining spatio-temporal gait parameters for both older adults and individuals with Parkinson’s disease ( 17 ). In general, the mentioned pose estimation models were trained using the same open-source Common Objects in Context (COCO) key point datasets ( 20 ). More recently, the YOLO (You Only Look Once) versions 7 and 8 pose estimation algorithm were used for both gait analysis and fall detection ( 21 , 22 ). According to Rana’s master thesis, YOLOv8, VitPose, and her proposed algorithm have sufficient precision, recall, and F1 scores to capture both kinematic and spatio-temporal analysis of gait ( 22 ). The critical gap in the literature is the lack of a direct comparison between a low-cost, markerless model obtained from a single camera and portable sensor systems widely used by clinicians in geriatric rehabilitation. Therefore, the purpose of this study was 1) to examine the YOLOv11 algorithm’s clinical validity through agreement and consistency analysis with the G-walk system, 2) to investigate the test-retest reliability of the YOLOv11 algorithm for the assessment of spatio-temporal gait parameters in community-dwelling older adults. We hypothesized that the YOLOv11 algorithm would be a valid and reliable gait assessment tool in older adults. Methods Participants Participants aged 60 years or over were selected from the database of volunteers at the Department of Physiotherapy and Rehabilitation, X University. Potential participants underwent an interview-based screening process to identify any exclusion criteria. Individuals were excluded if they exhibited Mini-Mental State Examination ≤ 23/30, were unable to comprehend and execute the test instructions, could not walk 20 meters independently, or had significant neurological, cardiorespiratory, or psychiatric conditions, or cerebrovascular accidents resulting in locomotor disability. While co-morbidities were not automatically disqualifying, acute or uncontrolled conditions served as exclusion factors. Study design and measurements Participants were assessed using the Mini-Mental State Examination (MMSE) and the Timed Up and Go (TUG) test after collecting sociodemographic data and medical history ( 23 ). In terms of medical history, yes/no answers ranging from 0 to 11 were collected according to participants’ chronic diseases such as hypertension, chronic heart diseases, diabetes mellitus, chronic lung disease, chronic joint disorders, chronic metabolic diseases (thyroid dysfunction, anemia), psychiatric disease, cancer history, rheumatoid diseases, cerebrovascular diseases, and eye diseases. For the TUG test, participants were asked to stand from an armchair with armrests, walk 3 meters at a normal pace, turn around, return, and sit down ( 24 ). The time taken was recorded in seconds. The MMSE and TUG tests were used to screen for basic mental ability, mobility, and functional balance, and to ensure sufficient safety of gait trials. Participants took part in two measurement sessions. In the first session, they were evaluated for gait analysis using both an IMU-based G-Walk system and a standard video camera simultaneously (iPhone 15) during a 4-meter walking test. The camera was placed laterally 4.5 meters from the participant (Fig. 1 ). In the second session, scheduled 1–3 days later, participants performed the same gait analysis to assess the reliability analysis of the YOLOv11 algorithm-based system. Before the beginning of the simultaneous trials, the participant familiarized to the procedure 3 times with the researcher accompanying. The last was included in data analysis. The general steps of the study are also summarized in Fig. 1 . Gait analysis procedure with inertial measurement unit (IMU) based G-walk system Inertial measurement unit (IMU)-based gait analysis has been conducted using the BTS G-WALK® (BTS Bioengineering S.p.A., Italy) system, a valid and reliable tool for assessing dynamic spatio-temporal gait parameters (ICC = 0.85–0.99) ( 12 , 13 ). The system includes one wearable IMU sensor (37 g, 70x40x18 mm), one semi-elastic belt, and a notebook for processing the data via the software programme BTS G-Studio (BTS Bioengineering S.p.A., Italy). Wearable IMU sensor composed of a triaxial gyroscope (16 bit/axes), accelerometer (16 bit/axes), and magnetometer (13 bit ± 1,200 mT)( 13 ). Following the recording of demographic data (age, sex, height, weight, and shoe size), the IMU sensor was fixed at the level of the L5 vertebra using the adjustable belt. Participants were instructed to walk a 4-meter straight path marked with starting and finishing points on the floor at their normal pace after the command “Get ready-Attention-Go” and to stop with the “Stop” command. The software programme computed the following gait parameters: gait speed (m/s), cadence (steps/min), stride length (m), gait cycle duration (s), stance phase (% of gait cycle time), swing phase (% of gait cycle time), double support phase (% of gait cycle time), single support phase (% of gait cycle time). The system provides discrete measurement values for both the right and left lower limbs. To create a representative gait pattern, spatio-temporal parameters from both the right and left sides were averaged across participants. However, we provided the findings indicating both sides in the Supplementary File (see Table S1 &2, Figure S1 ). Gait analysis procedure with pose estimation algorithm (YOLOv11) based system Synchronized video recordings taken concurrently with BTS G-WALK assessments were used for pose estimation-based gait analysis using a video camera (iPhone 15) HD, 30 FPS, placed 4.5 meters from the participant. The YOLOv11 algorithm was used for keypoint extractions for spatio-temporal gait variables. YOLOv11 is a fast single-stage algorithm structure based on a “top-down” approach ( 25 ). It detects people in the image using bounding boxes, then estimates separate human-specific keypoints for each box ( 25 ). By tracking the projection of the determined keypoints on the X and Y axes, the movement of the keypoints in two dimensions can be tracked as a time series. In our process, videos were cropped simultaneously according to the command voice from “Get ready-Attention-Go” to “Stop” at first. The YOLOv11 algorithm was used with a default confidence threshold of 0.25. The algorithm detected and tracked 17 key joint landmarks in 2D during walking sequences. The initial contact and toe-off phases of the gait cycle were identified using the X and Y coordinates of the right and left ankle keypoints. Initial contact was determined from the first derivative of the position data, while toe-off was identified from the second derivative (Fig. 2 ). This custom algorithm, developed based on key events, was used to calculate these spatio-temporal gait characteristics. The algorithm extracted the same spatio-temporal gait parameters as the G-walk system for gait speed (m/s), cadence (steps/min), stride length (m), gait cycle duration (s), stance phase (% of gait cycle time), swing phase (% of gait cycle time), double support phase (% of gait cycle time), single support phase (% of gait cycle time). To create a representative gait pattern, spatio-temporal parameters from both the right and left sides were averaged across participants. All algorithm designs and the processes mentioned were implemented using Python 3.8.10. Statistical analysis Demographic data of the participants were presented as means and standard deviations, while categorical data were summarized as frequencies and percentages. The concurrent validity of the YOLOv11 algorithm was assessed by evaluating its consistency and agreement with the IMU-based G-walk system. Pearson correlation analysis was used for consistency, while Bland-Altman plots were used to assess agreement between the two systems. Pearson correlation coefficient was classified as follows: <0.20 poor, 0.20–0.39 weak, 0.40–0.59 moderate, 0.60–0.79 strong, ≥ 0.80 excellent consistency ( 26 ). The agreement between two systems was assessed using 95% limits of agreement (LoA), calculated as the mean difference between the systems plus or minus 1.96 standard deviations. This approach accounts for both systematic error (mean bias) and random error (differences around the mean for each participant). To express the 95% LoA as a percentage (LoA%), the LoA range was divided by each average gait parameter across both systems. Absolute agreement was interpreted using LoA% with categories: excellent (0-4.9%), good (5-9.9%), moderate (10-49.9%), and poor (> 50%) ( 27 , 28 ). Further, intra-class correlation coefficients (ICCs) were calculated to provide reliability statistics for the YOLOv11 algorithm. Test-retest reliability was assessed using a two-way mixed-effects model with absolute agreement ICC( 3 , 1 ), as all measurements were performed twice by the same researcher. The ICC values were interpreted as follows: >0.90 excellent, 0.75–0.90 good, 0.50–0.75 moderate, < 0.50 poor reliability ( 29 ). All analyses were conducted using SPSS IBM 27. The level of significance was set as 0.05. Results A total of 52 participants were assessed for eligibility, of whom 21 were excluded due to (n = 4) stroke-related gait abnormalities, (n = 9) not willing to participate, and (n = 8) scores in MMSE ≤ 23. Thirty-one community-dwelling older adults (21 women and 10 men, aged 66.9 ± 6.4 years, BMI = 26.7 ± 4.6 kg/m 2 ) were included in the study. Table 1 summarizes the descriptive characteristics of the participants, including the mean number of chronic diseases (1.39), MMSE (28.06), and TUG (8.28 s) (Table 1 ). Table 1 Descriptive characteristics of the participants Variable Mean ± SD Age, years 66.94 ± 6.39 Gender, n (%) -women -men 21 (67.7) 10 (32.3) Weight, kg 72.84 ± 13.18 Height, cm 165.32 ± 9.00 BMI, kg/m 2 26.68 ± 4.65 Number of chronic diseases, score 0–11 1.39 ± 1.17 MMSE, score 0–30 28.06 ± 2.51 TUG, s 8.28 ± 2.06 Abbreviations: SD, Standard deviation; BMI, Body mass index; MMSE, Mini-mental state examination; TUG, Timed up and go. Table 2 and Fig. 3 show validity statistics of the YOLO11 algorithm to determine spatio-temporal gait parameters compared to the G-walk system via correlation analysis and Bland- Altman plots. The highest correlations were found in terms of cadence (r = 0.95, p < 0.001), gait speed (r = 0.90, p < 0.001), and gait cycle duration (r = 0.95, p < 0.001). Strong correlations were found for swing phase (r = 0.72, p < 0.001), stance phase (r = 0.68, p < 0.001), first double support (r = 0.67, p < 0.001), stride length (r = 0.70, p < 0.001), and single support (r = 0.60, p = 0.001) parameters. According to Bland Altman plots, LoA% showed moderate agreement (16.75%-47.04%) between the YOLO11 algorithm-based system and the G-walk system to determine swing-stance phases, single support, cadence, gait speed, gait cycle duration, and stride length, while poor agreement (95.60%) was found in terms of first double support (Table 2 & Fig. 3 ). Table 2 Validity statistics of the YOLOv11 Algorithm-based system compared to G-walk system to determine spatio-temporal gait parameters in older adults Parameter G-walk Mean ± SD YOLOv11 Algorithm Mean ± SD r p Consistency LoA% Agreement Swing phase, % GCT 39.38 ± 3.08 40.23 ± 3.26 0.72 < 0.001* Strong 23.34 Moderate Stance phase, % GCT 60.73 ± 3.22 59.76 ± 3.26 0.68 < 0.001* Strong 16.78 Moderate First double support, % GCT 10.36 ± 2.78 10.06 ± 3.26 0.67 < 0.001* Strong 95.60 Poor Single support, % GCT 39.54 ± 2.84 39.98 ± 3.23 0.60 0.001* Strong 27.05 Moderate Cadance, steps/min 105.28 ± 16.00 102.97 ± 14 0.95 < 0.001* Excellent 19.57 Moderate Gait speed, m/s 1.06 ± 0.27 0.94 ± 0.21 0.90 < 0.001* Excellent 47.04 Moderate Gait cycle duration, s 1.16 ± 0.17 1.18 ± 0.16 0.95 < 0.001* Excellent 16.75 Moderate Stride length, m 1.20 ± 0.20 1.24 ± 0.17 0.70 < 0.001* Strong 44.98 Moderate Abbreviations: SD, standard deviation; LoA, Limit of agreement; GCT, Gait cycle time; r, Pearson correlation coefficient; *p<0.05 in Pearson Correlation Analysis. The test-retest reliability analysis of the YOLOv11 algorithm-based system determining spatio-temporal parameters of gait in older adults is summarized in Table 3 . The findings showed excellent reliability for cadence (ICC = 0.92[0.84–0.96], p < 0.001) and gait speed (ICC = 0.91[0.82–0.96], p < 0.001), while good reliability was found for gait cycle duration (ICC = 0.89[0.77–0.95], p < 0.001) and stride length (ICC = 0.90[0.78–0.95], p < 0.001). For the remaining parameters as the phases of gait had moderate reliability (0.59 < ICC < 0.61, p < 0.05) (Table 3 ). Table 3 The reliability analysis of YOLOv11 algorithm-based system determining spatio-temporal characteristics of gait in older adults Parameter YOLOv11 Test Mean ± SD YOLOv11 Retest Mean ± SD 95% CI ICC Interpretation p Lower Upper Swing phase, % GCT 40.23 ± 3.26 40.75 ± 3.08 0.18 0.81 0.61 Moderate 0.007* Stance phase, % GCT 59.76 ± 3.26 59.24 ± 3.08 0.18 0.81 0.61 Moderate 0.007* First double support, % GCT 10.06 ± 3.26 9.57 ± 2.92 0.15 0.80 0.59 Moderate 0.009* Single support, % GCT 39.98 ± 3.23 40.39 ± 2.90 0.14 0.80 0.59 Moderate 0.01* Cadance, steps/min 102.97 ± 14 105.12 ± 13.64 0.84 0.96 0.92 Excellent < 0.001* Gait speed, m/s 0.94 ± 0.21 0.98 ± 0.21 0.82 0.96 0.91 Excellent < 0.001* Gait cycle duration, s 1.18 ± 0.16 1.15 ± 0.14 0.77 0.95 0.89 Good < 0.001* Stride length, m 1.24 ± 0.17 1.27 ± 0.16 0.78 0.95 0.90 Good < 0.001* Abbreviations: SD, standard deviation; CI, confidence interval; ICC, Intra-class correlation coefficient; GCT, Gait cycle time; *p<0.05 in reliability analysis. Detailed analyses for the right and left sides were provided in the Supplementary file (see Table S1 &2, Figure S1 ). Discussion In this study, we established the clinical validity and reliability of the YOLOv11 algorithm-based system by assessing its consistency and agreement with the G-walk system. Our findings indicate that the YOLOv11 algorithm-based system demonstrated excellent to moderate consistency and moderate agreement with the G-walk system, alongside excellent to moderate test-retest reliability. In terms of clinical validity of the YOLOv11 algorithm-based system, we referenced the G-walk system due to general clinical use of current studies, both healthy older adults and other populations such as Parkinson’s disease, dementia, knee osteoarthritis, etc. ( 12 , 30 – 32 ). Our findings revealed that spatio-temporal gait parameters calculated from the single video with the YOLOv11 algorithm-based system have excellent to strong correlation with the G-walk system. However, all spatio-temporal gait parameters had moderate agreement between the systems, except for the first double support with poor agreement. Previous studies regarding the validation of video pose tracking algorithms mainly used OpenPose, AlphaPose, Detectron, and PoseNet algorithms or a customized algorithm for a multiple camera system in community-dwelling older adults ( 15 – 18 ) (Table 4 ). Ripic et al. used an eight-camera based Kinatrax system within a customized algorithm to identify spatio-temporal gait parameters in older adults, and they found excellent agreement with the Nexus motion capture system ( 17 ). Authors expanded their previous report, which has calculated gait phases according to the ankle key-point and retrain their algorithm, including the calcanei key point ( 17 , 33 ). This retraining procedure has led to their findings of better agreement in variables of stride width, step time, stance time, swing time, and double limb support ( 17 ). In our results, we had sufficient correlation scores in terms of consistency, while agreement was moderate for spatio-temporal gait parameters which is obviously lower than the aforementioned study’s findings. The main cause of this discrepancy is that we followed a minimalist methodology with a single video camera to assess clinical validity of the pose estimation-based analysis system. Additionally, Nexus and Kinatrax have a similar technology-based approach for gait analysis ( 17 ), while we had two different approaches IMU based G-walk system and YOLOv11 algorithms-based system. This methodology has offered paradoxical effects encompassing both advantages and disadvantages. The advantage is that this could lead us to a new path for gait analysis in older adults, regardless of location. Further, if the model expands and is adapted through a smartphone application, access to gait analysis, tracking rehabilitation processes, and a clinical easy-to-use system could be possible in elderly care centers, nursing homes, or rehabilitation clinics without a lab-based environment. This kind of approach helps prevent the “inverse care law” phenomenon, where disadvantaged populations are underrepresented in scientific studies compared to healthy individuals, especially for older people ( 34 ). In the other hand, we should underline the disadvantage of high correlation coefficients indicating consistent ranking across G-walk system and YOLOv11 algorithm-based system, while the limits of agreements between absolute values found wider than expected (moderate agreements LoA%=16.75%-47.04% for swing-stance phases, single support, cadence, gait speed, gait cycle duration, and stride length; poor agreement LoA%=95.60% for first double support). This situation points to random error variation rather than systematic bias. The fundamental source of random error lies in the nature of the two technologies being compared. While the G-walk system collects motion information using an inertial measurement sensor, the YOLOv11-based system obtains data through pixel-based geometric estimates, predicting from a single 2D video. However, it is still promising that the algorithm-based system has the ability to capture trends in metrics showing moderate levels of agreement with the G-walk system. Table 4 Validity and reliability studies of pose estimation methodology to conduct gait analyses in older adults Study Subject Test Condition Pose Estimation Algorithm Reference system Parameter Test-retest reliability LoA for gait speed (m/s) LoA for Stride Length (m) LoA for Stance Phase (s or %) Correlations with Reference System Carvalho, 2024 Older people (n = 29), Mean Age = 75.2 years -Qualisys Track Manager with 8 video cameras, synchronized with 3 force plates on 12-meter walking path -Theia3D markerless motion capture NA Kinematic analysis &phases of gait Kinematic analysis ICC:0.53–0.88 Early phases of gait and a point-based estimation ICC < 0.5 Average phase detection ICC > 0.90 NA NA NA NA Liang, 2022 Older people (n = 15), Mean Age = 56.60 years -Single-camera based data collection on 6 meters walking path -Customized algorithm from 2D to 3D kinematic analysis -OpenPose -PoseNet VICON marker based optical system with 6 cameras Kinematic analysis ICC:0.50–0.73 NA NA NA NA Mehdizadeh, 2021 Older people (n = 11), Mean Age = 85.2 years -Two cell phones camera-based data collection with two different placements of “eye-level” and “straight” -13 meters walking path -AlphaPose (AP) -OpenPose (OP) -Detectron (D) Xsens MVN Awinda motion capture system with 7 IMU sensors Spatiotemporal Gait Parameters NA LoA was evaluated to present agreement between pose estimation algorithms. Authors find highest agreement between AP and D to determine spatiotemporal gait variables. Eye-level, front view conditions r, AP& XSens Cadence: 0.99 Step time:0.71 Step width:0.54 r, OP& XSens Cadence: 0.99 Step time:0.71 Step width:0.42 r, D& XSens Cadence: 0.99 Step time:0.71 Step width:0.54 Ripic, 2023 Young (n = 23) & Older Adults (n = 14), and subjects with Parkinson Disease (n = 20) -Kinatrax and Nexus motion capture system -Kinatrax, HumanVersion3 with 8 video cameras on 10-meter walking path -Nexus motion capture system with 8 video cameras on 10-meter walking path Spatiotemporal Gait Parameters NA -0.039,0.039 -0.05,0.048 -0.049,0.046 ICC for relative consistency and absolute Agreement > 0.85 ( Insert Table 4 here .) While the agreement analysis was moderate, the finding about consistency is parallel within the current literature. For instance, Ripic et al. found correlation coefficients greater than 0.85 between the Nexus 3D kinematic analysis system and the pose-estimation based Kinatrax system to determine spatio-temporal gait variables in older adults ( 17 ). Our findings also indicate significant correlations between the YOLOv11 and G-walk systems for cadence, gait speed, and gait cycle duration (r ≥ 0.90); stride length and swing phase (r ≥ 0.70); and stance phase, first double, and single support (r ≥ 0.60). On the other hand, Mehdizadeh et al. examined the correlations separately for AlphaPose, OpenPose, and Detectron with the XSense motion capture system, and found r = 0.99 for cadence, r = 0.71 for step time for these three algorithms under the front view-eye level camera replacement conditions in older adults ( 18 ). These findings are consistent with our results (r = 0.95 for cadence, r = 0.95 for gait cycle duration). The authors also conducted their analyses using a single camera video, while they used two separate cameras to evaluate different gait conditions such as front and back walking, or walking at eye level or straight height ( 18 ). However, we completed a custom-built data extraction process using YOLOv11 keypoints and did not compare it with other algorithms such as AlphaPose or OpenPose. Future research could update this comparison by benchmarking against algorithms with high accuracy, precision, and recall metrics, including both previously used and newly developed ones, such as VitPose and customized models ( 18 , 22 ). Research on conducting psychometric analyses of the pose estimation method is still warranted, particularly regarding the gait analysis in the older population. In terms of test-retest reliability, our findings demonstrate that YOLOv11-based analysis showed excellent to good reliability in detecting cadence (ICC = 0.92), gait speed (ICC = 0.91), stride length (ICC = 0.90), and gait cycle duration (ICC = 0.89). In addition, there is moderate reliability in terms of swing phase (ICC = 0.61), stance phase (ICC = 0.61), first double support (ICC = 0.59), and single support (ICC = 0.59). The lower reliability levels could reflect a methodological limitation known to be challenging for G-walk itself, rather than a flaw inherent to the YOLOv11-based analysis. De Ridder et al. reported excellent reliability scores (ICC = 0.85–0.91) for spatio-temporal gait variables, while the authors found cautious interpretation for agreement analyses of temporal variables with the GaitRite system ( 13 ). Similarly, Viteckova et al. concluded in their comparison of G-walk and GaitRite systems that these phase durations should be used with caution due to systematic error ( 12 ). G-walk system detects foot contact events using the sensor mounted at the L5 level of the spine via acceleration and angular velocity signals ( 12 , 13 ). Thus, it might be demonstrated that the YOLOv11 algorithm-based analysis also faces an inherent difficulty in detecting these phase transitions. In parallel, Carvalho et al. reported that the reliability of the Theia3D markerless motion capture system decreases (ICC < 0.50) at the early phase transitions of gait compared to the general kinematic flow of the gait cycles (ICC > 0.90) in older adults ( 16 ). Additionally, Liang et al. found reliability scores ranging from 0.50 to 0.73 for kinematic gait analysis in older adults using OpenPose and PoseNet algorithm-based systems ( 15 ). Although kinematic gait analysis requires greater precision than spatio-temporal gait analysis, the moderate reliability indicated in the aforementioned studies suggests that pose estimation systems still need improvement. There are several limitations that could affect the present findings and the potential impact of the proposed YOLOv11 algorithm-based spatio-temporal gait analysis. First one is that we used the G-walk system as a reference with its own limitations, such as soft tissue artefacts or sensor drift. However, this criterion was selected due to G-walk’s low cost, widespread use in clinical and rehabilitation settings, and established validity. While this validates the success of the present system in variables such as gait speed, cadence, gait cycle duration, and stride length, an inertial sensor placed on L5, inevitably introduces random error in measurements of gait phases. This may have caused the Bland-Altman agreement limits between the systems to expand, potentially underestimating the actual clinical validity of the YOLOv11-based system. In interpreting our findings, it should be noted that G-walk is a practical criterion but not a perfect gold standard. Secondly, data collection using a single smartphone camera from left lateral view is another limitation of the present study. However, the limitation stems from a conscious methodological choice, as only spatio-temporal parameters are evaluated. The obtained validity and reliability scores cannot be directly generalised to older individuals with gait disorders due to our inclusion criteria. Finally, we opted for a 4-meter gait distance to evaluate the basic parameters of gait in a location-independent, minimal environment in older adults. However, it is important to consider that this abbreviated distance might have affected the comprehensive capture of a stable, steady-state gait cycle compared to the extended paths reported previously. Conclusion Quantitative tools that analyze spatio-temporal gait characteristics can help identify specific gait deviations in older adults. They can also guide physiotherapy and rehabilitation programs, as well as track the effectiveness of these interventions. However, these measurements can be quite expensive and are often inaccessible to many researchers and clinicians. In this study, we used video-based gait analysis using only one smartphone and an open-source algorithm to show the clinical validity and reliability of spatio-temporal gait analysis based on the YOLOv11 algorithm in community-dwelling older adults. Our findings demonstrate that the proposed system captures gait speed, cadence, gait cycle duration, and stride length with excellent to strong consistency, moderate agreement, and excellent to good test-retest reliability, comparable to G-walk, providing its ability to consistently capture these metrics. This may provide significant evidence for the proposed system’s use in rehabilitation monitoring and general follow-up processes, without the need for expensive and complex laboratory setups. However, the agreement analysis and reliability findings observed in the temporal variables of gait indicate that the system’s interchangeability is not yet feasible in these metrics. The source of these limitations may not be only random error arising from a single camera setup, but also methodological difficulties encountered by the reference G-walk system itself. Overall, pose estimation technology is promising in capturing clinically relevant metrics regarding spatial gait parameters in older adults. Future studies should focus on improving and testing the performance of pose estimation-based systems in a clinical population to enhance the clinical validity of this system. Abbreviations WHO World Health Organization IMU Inertial measurement unit COCO Common Objects in Context YOLO You Only Look Once MMSE Mini-Mental State Examination TUG Timed Up and Go LOA Limits of agreement ICCs Intra-class correlation coefficients Declarations Ethics approval and consent to participate : The study was approved by the X University Ethical Committee for Non-Interventional Research (Registration Number: 2025/12 − 08, Date: 09.04.2025) and executed according to principles of the Helsinki Declaration. All patients gave their written informed consent. Author Contribution Conceptualization: GYC, UC, IKO, SO; conducting outcome assessments, GYC, IKO, PS; analyzing and interpreting data, UC, GYC; writing and original draft preparation, GYC, UC; writing, review, and editing, GYC, UC, IKO, SO, PS; approving the final version to be published, GYC, UC, IKO, PS, SO. Funding: No funding was received for this work.Acknowledgment: We thank Necla Kara for granting permission to use her image in this manuscript. Furthermore, this work was honored with the Best Poster Award at the PhysAgeNet & EGRAPA Conference 2025, granted by the EU COST Action CA20104 - Network on Evidence-Based Physical Activity in Old Age (PhsAgeNet), supported by COST (European Cooperation in Science and Technology).Conflict of Interest: The authors declare that they have no conflict of interest related to the publication of this manuscript. Acknowledgement We thank Necla Kara for granting permission to use her image in this manuscript. Furthermore, this work was honored with the Best Poster Award at the PhysAgeNet & EGRAPA Conference 2025, granted by the EU COST Action CA20104 - Network on Evidence-Based Physical Activity in Old Age (PhsAgeNet), supported by COST (European Cooperation in Science and Technology). References Progress report on the United Nations. Decade of Healthy Ageing, 2021–3. Gianfredi V, Nucci D, Pennisi F, Maggi S, Veronese N, Soysal P. Aging, longevity, and healthy aging: the public health approach. Aging Clin Exp Res. 2025;37(1):125. 10.1007/s40520-025-03021-8 . Bautmans I, Knoop V, Amuthavalli Thiyagarajan J, Maier AB, Beard JR, Freiberger E, et al. WHO working definition of vitality capacity for healthy longevity monitoring. Lancet Healthy Longev. 2022;3(11):e789–96. 10.1016/S2666-7568(22)00200-8 . Soltani A, Abolhassani N, Marques-Vidal P, Aminian K, Vollenweider P, Paraschiv-Ionescu A. Real-world gait speed estimation, frailty and handgrip strength: a cohort-based study. Sci Rep. 2021;11(1):18966. 10.1038/s41598-021-98359-0 . Zhou H, Park C, Shahbazi M, York MK, Kunik ME, Naik AD, et al. Digital Biomarkers of Cognitive Frailty: The Value of Detailed Gait Assessment Beyond Gait Speed. Gerontology. 2022;68(2):224–33. 10.1159/000515939 . Lindemann U. Spatiotemporal gait analysis of older persons in clinical practice and research. Z Gerontol Geriatr. 2020;53(2):171–8. 10.1007/s00391-019-01520-8 . Boyer KA, Hayes KL, Umberger BR, Adamczyk PG, Bean JF, Brach JS, et al. Age-related changes in gait biomechanics and their impact on the metabolic cost of walking: Report from a National Institute on Aging workshop. Exp Gerontol. 2023;173:112102. 10.1016/j.exger.2023.112102 . Sethi D, Bharti S, Prakash C. A comprehensive survey on gait analysis: History, parameters, approaches, pose estimation, and future work. Artif Intell Med. 2022;129:102314. 10.1016/j.artmed.2022.102314 . Beauchet O, Allali G, Sekhon H, Verghese J, Guilain S, Steinmetz JP, et al. Guidelines for Assessment of Gait and Reference Values for Spatiotemporal Gait Parameters in Older Adults: The Biomathics and Canadian Gait Consortiums Initiative. Front Hum Neurosci. 2017;11. 10.3389/fnhum.2017.00353 . Prasanth H, Caban M, Keller U, Courtine G, Ijspeert A, Vallery H, et al. Wearable Sensor-Based Real-Time Gait Detection: A Systematic Review. Sensors. 2021;21(8):2727. 10.3390/s21082727 . Prisco G, Pirozzi MA, Santone A, Esposito F, Cesarelli M, Amato F, et al. Validity of Wearable Inertial Sensors for Gait Analysis: A Systematic Review. Diagnostics. 2024;15(1):36. 10.3390/diagnostics15010036 . Vítečková S, Horáková H, Poláková K, Krupička R, Růžička E, Brožová H. Agreement between the GAITRite ® System and the Wearable Sensor BTS G-Walk ® for measurement of gait parameters in healthy adults and Parkinson’s disease patients. PeerJ. 2020;8:e8835. 10.7717/peerj.8835 . De Ridder R, Lebleu J, Willems T, De Blaiser C, Detrembleur C, Roosen P. Concurrent Validity of a Commercial Wireless Trunk Triaxial Accelerometer System for Gait Analysis. J Sport Rehabil. 2019;28(6). 10.1123/jsr.2018-0295 . Bautmans I, Jansen B, Van Keymolen B, Mets T. Reliability and clinical correlates of 3D-accelerometry based gait analysis outcomes according to age and fall-risk. Gait Posture. 2011;33(3):366–72. 10.1016/j.gaitpost.2010.12.003 . Liang S, Zhang Y, Diao Y, Li G, Zhao G. The reliability and validity of gait analysis system using 3D markerless pose estimation algorithms. Front Bioeng Biotechnol. 2022;10. 10.3389/fbioe.2022.857975 . Carvalho A, Vanrenterghem J, Cabral S, Assunção A, Fernandes R, Veloso AP, et al. Markerless three-dimensional gait analysis in healthy older adults: test–retest reliability and measurement error. J Biomech. 2024;174:112280. 10.1016/j.jbiomech.2024 . Ripic Z, Signorile JF, Best TM, Jacobs KA, Nienhuis M, Whitelaw C, et al. Validity of artificial intelligence-based markerless motion capture system for clinical gait analysis: Spatiotemporal results in healthy adults and adults with Parkinson’s disease. J Biomech. 2023;155:111645. 10.1016/j.jbiomech.2023.111645 . Mehdizadeh S, Nabavi H, Sabo A, Arora T, Iaboni A, Taati B. Concurrent validity of human pose tracking in video for measuring gait parameters in older adults: a preliminary analysis with multiple trackers, viewing angles, and walking directions. J Neuroeng Rehabil. 2021;18(1):139. 10.1186/s12984-021-00933-0 . Stenum J, Rossi C, Roemmich RT. Two-dimensional video-based analysis of human gait using pose estimation. PLoS Comput Biol. 2021;17(4):e1008935. 10.1371/journal.pcbi.1008935 . Lin TY, Maire M, Belongie S, Bourdev L, Girshick R, Hays J et al. Microsoft COCO: Common Objects in Context. 2015 Feb 21. Tîrziu E, Vasilevschi AM, Alexandru A, Tudora E. Enhanced Fall Detection Using YOLOv7-W6-Pose for Real-Time Elderly Monitoring. Future Internet. 2024;16(12):472. 10.3390/fi16120472 . Rana MS. Leveraging Markerless Computer Vision for Comprehensive Walking Automated Gait Analysis in Rehabilitation. 2024. Güngen C, Ertan T, Eker E, Yaşar R, Engin F. [Reliability and validity of the standardized Mini Mental State Examination in the diagnosis of mild dementia in Turkish population]. Turk Psikiyatri Derg. 2002;13(4). PMID: 12794644. Podsiadlo D, Richardson S. The Timed Up & Go: A Test of Basic Functional Mobility for Frail Elderly Persons. J Am Geriatr Soc. 1991;39(2):142–8. 10.1111/j.1532-5415.1991.tb01616.x . Khanam R, Hussain M. YOLOv11: An Overview of the Key Architectural Enhancements. 2024 Oct 23. Portney LG. Foundations of clinical research: applications to evidence-based practice. FA Davis; 2020. Olsen S, Rashid U, Barbado D, Suresh P, Alder G, Khan Niazi I, et al. The validity of smartphone-based spatiotemporal gait measurements during walking with and without head turns: Comparison with the GAITRite® system. J Biomech. 2024;162:111899. 10.1016/j.jbiomech.2023.111899 . Godfrey A, Del Din S, Barry G, Mathers JC, Rochester L. Instrumenting gait with an accelerometer: A system and algorithm examination. Med Eng Phys. 2015;37(4):400–7. 10.1016/j.medengphy.2015.02.003 . Koo TK, Li MY. A Guideline of Selecting and Reporting Intraclass Correlation Coefficients for Reliability Research. J Chiropr Med. 2016;15(2):155–63. 10.1016/j.jcm.2016.02.012 . Latajka A, Stefańska M, Woźniewski M, Malicka I. Walking Speed and Risk of Falling Patients Operated for Selected Malignant Tumors. Healthcare. 2023;11(23):3069. 10.3390/healthcare11233069 . Kang CJ, Park SH, Son SM. Improving Gait Ability and Cognition Function through Action Observation Training in Elderly with Dementia. NeuroRehabilitation: An International. Interdisciplinary J. 2025;56(2):234–42. 10.1177/10538135241296772 . Gianzina E, Yiannakopoulos CK, Kalinterakis G, Delis S, Chronopoulos E. Evaluation of the Timed Up and Go Test in Patients with Knee Osteoarthritis Using Inertial Sensors. Int J Translational Med. 2024;5(1):2. 10.3390/ijtm5010002 . Ripic Z, Signorile JF, Kuenze C, Eltoukhy M. Concurrent validity of artificial intelligence-based markerless motion capture for over-ground gait analysis: A study of spatiotemporal parameters. J Biomech. 2022;143:111278. 10.1016/j.jbiomech.2022.111278 . Cookson R, Doran T, Asaria M, Gupta I, Mujica FP. The inverse care law re-examined: a global perspective. Lancet. 2021;397(10276):828–38. 10.1016/S0140-6736(21)00243-9 . Additional Declarations No competing interests reported. Supplementary Files SupplementaryFile.docx Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {\"props\":{\"pageProps\":{\"initialData\":{\"identity\":\"rs-7987537\",\"acceptedTermsAndConditions\":true,\"allowDirectSubmit\":true,\"archivedVersions\":[],\"articleType\":\"Research Article\",\"associatedPublications\":[],\"authors\":[{\"id\":546151752,\"identity\":\"561a8bb1-746b-4ad9-bed8-e733b30a4ad4\",\"order_by\":0,\"name\":\"Gamze Yalcinkaya Colak\",\"email\":\"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAABKUlEQVRIiWNgGAWjYDADPgbGBsYGIIOfgYENJADm4AVsMC2SDXAtzIS0QA02OEBAi7z76eQPHxgOy7OxH257OKPmsLzxjeRjj3kYbGQ3HOA/9gGLFsMzuRsMZzAcNmzjSWw33HDssOG2G2npxjwMacYbDjAzz8CmpSF3QzIPw2HGNgnGNskHbIcZt93IMZMGiiSCtGBzmGH/2w2H/zActodo+XfYfvMMsJb/OLXIS+RubGYAmgnWsrENaLgEWMsBnFoMJN5uZuwxSE8G+qVNcmZfevKMM8/SDecYJBvPPMxsjNWW/tzNH35UWNv2sx9/JtnzDchoTz724E2FnWzf8cbHWG05ACabYXwgQyABJALEOGJSvgFM1cH4QAb/AexKR8EoGAWjYMQCAAsfaH9oRB/bAAAAAElFTkSuQmCC\",\"orcid\":\"\",\"institution\":\"Bozok Universitesi\",\"correspondingAuthor\":true,\"prefix\":\"\",\"firstName\":\"Gamze\",\"middleName\":\"Yalcinkaya\",\"lastName\":\"Colak\",\"suffix\":\"\"},{\"id\":546151753,\"identity\":\"f3e49bcf-8f99-4f5a-a6cc-a0229ede0422\",\"order_by\":1,\"name\":\"Umut Can Colak\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"Ege University\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Umut\",\"middleName\":\"Can\",\"lastName\":\"Colak\",\"suffix\":\"\"},{\"id\":546151756,\"identity\":\"e8ffedb4-7004-4abc-9fcc-ab5a611e31ea\",\"order_by\":2,\"name\":\"İlke Kara Oz\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"Dokuz Eylül University\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"İlke\",\"middleName\":\"Kara\",\"lastName\":\"Oz\",\"suffix\":\"\"},{\"id\":546151757,\"identity\":\"c93f05df-8951-45a6-99c3-dd615105444f\",\"order_by\":3,\"name\":\"Punhan Sailov\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"Dokuz Eylül University\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Punhan\",\"middleName\":\"\",\"lastName\":\"Sailov\",\"suffix\":\"\"},{\"id\":546151761,\"identity\":\"eff8df05-bf45-4344-b884-d8fc47a1889d\",\"order_by\":4,\"name\":\"Seher Ozyurek\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"Dokuz Eylül University\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Seher\",\"middleName\":\"\",\"lastName\":\"Ozyurek\",\"suffix\":\"\"}],\"badges\":[],\"createdAt\":\"2025-10-30 09:23:45\",\"currentVersionCode\":1,\"declarations\":\"\",\"doi\":\"10.21203/rs.3.rs-7987537/v1\",\"doiUrl\":\"https://doi.org/10.21203/rs.3.rs-7987537/v1\",\"draftVersion\":[],\"editorialEvents\":[],\"editorialNote\":\"\",\"failedWorkflow\":false,\"files\":[{\"id\":96364375,\"identity\":\"28a002f3-c83b-425f-80f6-149fe3364a2c\",\"added_by\":\"auto\",\"created_at\":\"2025-11-20 10:09:15\",\"extension\":\"docx\",\"order_by\":0,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"acdc-reference\",\"size\":966265,\"visible\":true,\"origin\":\"\",\"legend\":\"\",\"description\":\"\",\"filename\":\"Manuscriptanonymous.docx\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-7987537/v1/cc1337d7784a748e1894871c.docx\"},{\"id\":96286602,\"identity\":\"69cbbd58-3cdb-468e-9339-58a5525e2d29\",\"added_by\":\"auto\",\"created_at\":\"2025-11-19 12:05:33\",\"extension\":\"json\",\"order_by\":1,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"acdc-reference\",\"size\":8137,\"visible\":true,\"origin\":\"\",\"legend\":\"\",\"description\":\"\",\"filename\":\"db9fd0eb3ff24f22ad7123d400c70bf5.json\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-7987537/v1/7ac38f8e20af6cf205f9881c.json\"},{\"id\":96286608,\"identity\":\"82d2a39a-5a76-4b63-8958-522bcf245578\",\"added_by\":\"auto\",\"created_at\":\"2025-11-19 12:05:34\",\"extension\":\"docx\",\"order_by\":2,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"acdc-reference\",\"size\":315353,\"visible\":true,\"origin\":\"\",\"legend\":\"\",\"description\":\"\",\"filename\":\"SupplementaryFile.docx\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-7987537/v1/b0bc4af2f3316dc6bc75e8d3.docx\"},{\"id\":96286610,\"identity\":\"2b933f43-4258-46e8-b21b-67f63e0cdfab\",\"added_by\":\"auto\",\"created_at\":\"2025-11-19 12:05:34\",\"extension\":\"xml\",\"order_by\":3,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"acdc-reference\",\"size\":120277,\"visible\":true,\"origin\":\"\",\"legend\":\"\",\"description\":\"\",\"filename\":\"db9fd0eb3ff24f22ad7123d400c70bf51enriched.xml\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-7987537/v1/f7fd23387eb65f20783f90e7.xml\"},{\"id\":96286605,\"identity\":\"b8192f1a-6d06-4ba2-bf48-ec1d1977ef22\",\"added_by\":\"auto\",\"created_at\":\"2025-11-19 12:05:33\",\"extension\":\"png\",\"order_by\":7,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"acdc-reference\",\"size\":68620,\"visible\":true,\"origin\":\"\",\"legend\":\"\",\"description\":\"\",\"filename\":\"Onlinefloatimage1.png\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-7987537/v1/0516db6cad788fab979f1939.png\"},{\"id\":96363724,\"identity\":\"ed94664a-1413-4e36-bedb-e159bed4b224\",\"added_by\":\"auto\",\"created_at\":\"2025-11-20 10:07:51\",\"extension\":\"png\",\"order_by\":8,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"acdc-reference\",\"size\":32883,\"visible\":true,\"origin\":\"\",\"legend\":\"\",\"description\":\"\",\"filename\":\"Onlinefloatimage2.png\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-7987537/v1/7cb7670bac9c822544f49944.png\"},{\"id\":96365240,\"identity\":\"68a5036e-83ad-4a7d-bd28-d58d8b8458aa\",\"added_by\":\"auto\",\"created_at\":\"2025-11-20 10:10:08\",\"extension\":\"png\",\"order_by\":9,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"acdc-reference\",\"size\":126249,\"visible\":true,\"origin\":\"\",\"legend\":\"\",\"description\":\"\",\"filename\":\"Onlinefloatimage3.png\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-7987537/v1/05948267f738f716c43fc1e9.png\"},{\"id\":96286613,\"identity\":\"7efa5b4d-7da9-4401-a2e7-ab9c023bb6b4\",\"added_by\":\"auto\",\"created_at\":\"2025-11-19 12:05:34\",\"extension\":\"xml\",\"order_by\":10,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"acdc-reference\",\"size\":116651,\"visible\":true,\"origin\":\"\",\"legend\":\"\",\"description\":\"\",\"filename\":\"db9fd0eb3ff24f22ad7123d400c70bf51structuring.xml\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-7987537/v1/d41fa44cde2fddd786abd7e4.xml\"},{\"id\":96286614,\"identity\":\"c1198c9b-125d-4526-a960-a5f74445a344\",\"added_by\":\"auto\",\"created_at\":\"2025-11-19 12:05:34\",\"extension\":\"html\",\"order_by\":11,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"acdc-reference\",\"size\":127054,\"visible\":true,\"origin\":\"\",\"legend\":\"\",\"description\":\"\",\"filename\":\"earlyproof.html\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-7987537/v1/9d640442c708db6f700c3290.html\"},{\"id\":96286604,\"identity\":\"f41fddba-0d27-4ec3-a130-fbba12f7c334\",\"added_by\":\"auto\",\"created_at\":\"2025-11-19 12:05:33\",\"extension\":\"png\",\"order_by\":1,\"title\":\"Figure 1\",\"display\":\"\",\"copyAsset\":false,\"role\":\"figure\",\"size\":326645,\"visible\":true,\"origin\":\"\",\"legend\":\"\\u003cp\\u003eData collection and variable extraction steps of the study\\u003c/p\\u003e\",\"description\":\"\",\"filename\":\"floatimage1.png\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-7987537/v1/a5299cc03e5ba7e24d4592e0.png\"},{\"id\":96286603,\"identity\":\"59c2574d-94f3-4c00-a64f-0beb1517eb93\",\"added_by\":\"auto\",\"created_at\":\"2025-11-19 12:05:33\",\"extension\":\"png\",\"order_by\":2,\"title\":\"Figure 2\",\"display\":\"\",\"copyAsset\":false,\"role\":\"figure\",\"size\":231164,\"visible\":true,\"origin\":\"\",\"legend\":\"\\u003cp\\u003eThe algorithm design methodology\\u003c/p\\u003e\",\"description\":\"\",\"filename\":\"floatimage2.png\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-7987537/v1/294cdaa8db53edfaa046c30f.png\"},{\"id\":96286609,\"identity\":\"7d67d9bd-55ea-4854-9ec1-d589dea448bd\",\"added_by\":\"auto\",\"created_at\":\"2025-11-19 12:05:34\",\"extension\":\"jpeg\",\"order_by\":3,\"title\":\"Figure 3\",\"display\":\"\",\"copyAsset\":false,\"role\":\"figure\",\"size\":764993,\"visible\":true,\"origin\":\"\",\"legend\":\"\\u003cp\\u003eThe agreement analysis between the YOLOv11 algorithm-based system and the G-walk system\\u003c/p\\u003e\",\"description\":\"\",\"filename\":\"floatimage3.jpeg\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-7987537/v1/244f5d771a04bba47129c667.jpeg\"},{\"id\":102620432,\"identity\":\"d2a8deb9-7989-4da9-8eaf-1ed07d262cfe\",\"added_by\":\"auto\",\"created_at\":\"2026-02-13 16:41:28\",\"extension\":\"pdf\",\"order_by\":0,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"manuscript-pdf\",\"size\":2256889,\"visible\":true,\"origin\":\"\",\"legend\":\"\",\"description\":\"\",\"filename\":\"manuscript.pdf\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-7987537/v1/c66d822b-f81e-450d-bbc6-0b58a7068f75.pdf\"},{\"id\":96364882,\"identity\":\"eeb3bbbb-05fe-445a-8545-ff18373cd086\",\"added_by\":\"auto\",\"created_at\":\"2025-11-20 10:09:45\",\"extension\":\"docx\",\"order_by\":0,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"supplement\",\"size\":315353,\"visible\":true,\"origin\":\"\",\"legend\":\"\",\"description\":\"\",\"filename\":\"SupplementaryFile.docx\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-7987537/v1/27d77360ed7e1fcbca1c089e.docx\"}],\"financialInterests\":\"No competing interests reported.\",\"formattedTitle\":\"Clinical validity and reliability of the pose estimation-based system to determine spatio-temporal gait parameters in older adults: A pilot study\",\"fulltext\":[{\"header\":\"Background\",\"content\":\"\\u003cp\\u003eWorldwide, people aged 60 and over numbered 962\\u0026nbsp;million in 2017, and it is expected that this number will reach 2.1\\u0026nbsp;billion by 2050 (\\u003cspan citationid=\\\"CR1\\\" class=\\\"CitationRef\\\"\\u003e1\\u003c/span\\u003e, \\u003cspan citationid=\\\"CR2\\\" class=\\\"CitationRef\\\"\\u003e2\\u003c/span\\u003e). As societies age, the implications for public health, healthcare systems, and economies become increasingly complex. In light of the global aging phenomenon, the World Health Organization (WHO) emphasizes the determinants of healthy aging, including integrated healthcare models and consistent monitoring of intrinsic capacity (\\u003cspan citationid=\\\"CR2\\\" class=\\\"CitationRef\\\"\\u003e2\\u003c/span\\u003e, \\u003cspan citationid=\\\"CR3\\\" class=\\\"CitationRef\\\"\\u003e3\\u003c/span\\u003e). The notion of intrinsic capacity covers five domains: locomotor, cognitive, psychological, sensory, and vitality. Within this framework, gait of older adults emerges as an essential element of locomotion, and gait speed is recognized as the sixth vital sign due to its high predictive value for mortality, dementia, frailty, and other negative health outcomes associated with aging (\\u003cspan citationid=\\\"CR4\\\" class=\\\"CitationRef\\\"\\u003e4\\u003c/span\\u003e, \\u003cspan citationid=\\\"CR5\\\" class=\\\"CitationRef\\\"\\u003e5\\u003c/span\\u003e). Additionally, gait parameters are the primary target in geriatric rehabilitation to protect functional independence (\\u003cspan citationid=\\\"CR6\\\" class=\\\"CitationRef\\\"\\u003e6\\u003c/span\\u003e, \\u003cspan citationid=\\\"CR7\\\" class=\\\"CitationRef\\\"\\u003e7\\u003c/span\\u003e).\\u003c/p\\u003e\\u003cp\\u003eIn objective gait analysis, floor sensors, wearable sensors, and image processing are generally reliable approaches for determining spatiotemporal gait characteristics (\\u003cspan citationid=\\\"CR8\\\" class=\\\"CitationRef\\\"\\u003e8\\u003c/span\\u003e). Specifically, the lab-based optical motion capture systems such as the GAITrite system, were considered the gold standard over the last decade and has been studied in both healthy and various populations with Parkinson\\u0026rsquo;s disease, musculoskeletal problems, neurodegenerative diseases, etc. (\\u003cspan citationid=\\\"CR9\\\" class=\\\"CitationRef\\\"\\u003e9\\u003c/span\\u003e). However, the initial cost is quite high, and analysis requires a specific laboratory environment with time-consuming measurements. On the other hand, wearable inertial measurement unit (IMU)-based systems were also found to be valid and reliable in defining spatio-temporal gait characteristics (\\u003cspan citationid=\\\"CR10\\\" class=\\\"CitationRef\\\"\\u003e10\\u003c/span\\u003e, \\u003cspan citationid=\\\"CR11\\\" class=\\\"CitationRef\\\"\\u003e11\\u003c/span\\u003e). Systematic reviews and validation studies demonstrated moderate agreement between single IMU-based systems and the GAITrite system (\\u003cspan additionalcitationids=\\\"CR11 CR12\\\" citationid=\\\"CR10\\\" class=\\\"CitationRef\\\"\\u003e10\\u003c/span\\u003e\\u0026ndash;\\u003cspan citationid=\\\"CR13\\\" class=\\\"CitationRef\\\"\\u003e13\\u003c/span\\u003e). Viteckova et. al found that the wearable G-walk system is valid and reliable in healthy adults and patients with Parkinson\\u0026rsquo;s disease (\\u003cspan citationid=\\\"CR12\\\" class=\\\"CitationRef\\\"\\u003e12\\u003c/span\\u003e). In parallel, a single 3D-accelerometer-based gait analysis procedure was also found to be reliable and correlated with fall risk in older adults (\\u003cspan citationid=\\\"CR14\\\" class=\\\"CitationRef\\\"\\u003e14\\u003c/span\\u003e).\\u003c/p\\u003e\\u003cp\\u003eAt the National Institute of Aging workshop, Boyer et al. highlighted a significant limitation in gait research among older adults, the underrepresentation of individuals with slowed gait speed and incident mobility limitations, who are prevalent in the community but seldom included in traditional laboratory studies. (\\u003cspan citationid=\\\"CR7\\\" class=\\\"CitationRef\\\"\\u003e7\\u003c/span\\u003e). The authors identified knowledge gaps for future studies regarding the real-world measurement of gait in older adults (\\u003cspan citationid=\\\"CR7\\\" class=\\\"CitationRef\\\"\\u003e7\\u003c/span\\u003e). In this manner, remote and minimalistic gait monitoring can provide a broader solution for both older adults and health practitioners. Recent advancements in artificial intelligence have enabled to the development of practically applicable gait analysis systems using pose estimation algorithms (\\u003cspan additionalcitationids=\\\"CR16 CR17 CR18 CR19\\\" citationid=\\\"CR15\\\" class=\\\"CitationRef\\\"\\u003e15\\u003c/span\\u003e\\u0026ndash;\\u003cspan citationid=\\\"CR20\\\" class=\\\"CitationRef\\\"\\u003e20\\u003c/span\\u003e). Stenum et al. reported significant agreement between video-based gait analysis using the OpenPose algorithm and 3D motion capture systems in determining spatio-temporal gait parameters (\\u003cspan citationid=\\\"CR19\\\" class=\\\"CitationRef\\\"\\u003e19\\u003c/span\\u003e). In parallel, Mehdizadeh et al. evaluated the concurrent validity of three pose estimation algorithms (AlphaPose, OpenPose, and Detectron)-based gait analysis and showed that AlphaPose and OpenPose had the highest correlations with a 3D motion capture system for detecting step time and cadence in eleven older adults (\\u003cspan citationid=\\\"CR18\\\" class=\\\"CitationRef\\\"\\u003e18\\u003c/span\\u003e). Further, Ripic et al. showed excellent agreement with another algorithm from the Kinatrax markerless system and 3D motion analysis in determining spatio-temporal gait parameters for both older adults and individuals with Parkinson\\u0026rsquo;s disease (\\u003cspan citationid=\\\"CR17\\\" class=\\\"CitationRef\\\"\\u003e17\\u003c/span\\u003e). In general, the mentioned pose estimation models were trained using the same open-source Common Objects in Context (COCO) key point datasets (\\u003cspan citationid=\\\"CR20\\\" class=\\\"CitationRef\\\"\\u003e20\\u003c/span\\u003e). More recently, the YOLO (You Only Look Once) versions 7 and 8 pose estimation algorithm were used for both gait analysis and fall detection (\\u003cspan citationid=\\\"CR21\\\" class=\\\"CitationRef\\\"\\u003e21\\u003c/span\\u003e, \\u003cspan citationid=\\\"CR22\\\" class=\\\"CitationRef\\\"\\u003e22\\u003c/span\\u003e). According to Rana\\u0026rsquo;s master thesis, YOLOv8, VitPose, and her proposed algorithm have sufficient precision, recall, and F1 scores to capture both kinematic and spatio-temporal analysis of gait (\\u003cspan citationid=\\\"CR22\\\" class=\\\"CitationRef\\\"\\u003e22\\u003c/span\\u003e). The critical gap in the literature is the lack of a direct comparison between a low-cost, markerless model obtained from a single camera and portable sensor systems widely used by clinicians in geriatric rehabilitation. Therefore, the purpose of this study was 1) to examine the YOLOv11 algorithm\\u0026rsquo;s clinical validity through agreement and consistency analysis with the G-walk system, 2) to investigate the test-retest reliability of the YOLOv11 algorithm for the assessment of spatio-temporal gait parameters in community-dwelling older adults. We hypothesized that the YOLOv11 algorithm would be a valid and reliable gait assessment tool in older adults.\\u003c/p\\u003e\"},{\"header\":\"Methods\",\"content\":\"\\u003cdiv id=\\\"Sec3\\\" class=\\\"Section2\\\"\\u003e\\u003ch2\\u003eParticipants\\u003c/h2\\u003e\\u003cp\\u003eParticipants aged 60 years or over were selected from the database of volunteers at the Department of Physiotherapy and Rehabilitation, X University. Potential participants underwent an interview-based screening process to identify any exclusion criteria. Individuals were excluded if they exhibited Mini-Mental State Examination\\u0026thinsp;\\u0026le;\\u0026thinsp;23/30, were unable to comprehend and execute the test instructions, could not walk 20 meters independently, or had significant neurological, cardiorespiratory, or psychiatric conditions, or cerebrovascular accidents resulting in locomotor disability. While co-morbidities were not automatically disqualifying, acute or uncontrolled conditions served as exclusion factors.\\u003c/p\\u003e\\u003c/div\\u003e\\n\\u003ch3\\u003eStudy design and measurements\\u003c/h3\\u003e\\n\\u003cp\\u003eParticipants were assessed using the Mini-Mental State Examination (MMSE) and the Timed Up and Go (TUG) test after collecting sociodemographic data and medical history (\\u003cspan citationid=\\\"CR23\\\" class=\\\"CitationRef\\\"\\u003e23\\u003c/span\\u003e). In terms of medical history, yes/no answers ranging from 0 to 11 were collected according to participants\\u0026rsquo; chronic diseases such as hypertension, chronic heart diseases, diabetes mellitus, chronic lung disease, chronic joint disorders, chronic metabolic diseases (thyroid dysfunction, anemia), psychiatric disease, cancer history, rheumatoid diseases, cerebrovascular diseases, and eye diseases.\\u003c/p\\u003e\\u003cp\\u003eFor the TUG test, participants were asked to stand from an armchair with armrests, walk 3 meters at a normal pace, turn around, return, and sit down (\\u003cspan citationid=\\\"CR24\\\" class=\\\"CitationRef\\\"\\u003e24\\u003c/span\\u003e). The time taken was recorded in seconds. The MMSE and TUG tests were used to screen for basic mental ability, mobility, and functional balance, and to ensure sufficient safety of gait trials.\\u003c/p\\u003e\\u003cp\\u003eParticipants took part in two measurement sessions. In the first session, they were evaluated for gait analysis using both an IMU-based G-Walk system and a standard video camera simultaneously (iPhone 15) during a 4-meter walking test. The camera was placed laterally 4.5 meters from the participant (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig1\\\" class=\\\"InternalRef\\\"\\u003e1\\u003c/span\\u003e). In the second session, scheduled 1\\u0026ndash;3 days later, participants performed the same gait analysis to assess the reliability analysis of the YOLOv11 algorithm-based system. Before the beginning of the simultaneous trials, the participant familiarized to the procedure 3 times with the researcher accompanying. The last was included in data analysis. The general steps of the study are also summarized in Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig1\\\" class=\\\"InternalRef\\\"\\u003e1\\u003c/span\\u003e.\\u003c/p\\u003e\\n\\u003ch3\\u003eGait analysis procedure with inertial measurement unit (IMU) based G-walk system\\u003c/h3\\u003e\\n\\u003cp\\u003eInertial measurement unit (IMU)-based gait analysis has been conducted using the BTS G-WALK\\u0026reg; (BTS Bioengineering S.p.A., Italy) system, a valid and reliable tool for assessing dynamic spatio-temporal gait parameters (ICC\\u0026thinsp;=\\u0026thinsp;0.85\\u0026ndash;0.99) (\\u003cspan citationid=\\\"CR12\\\" class=\\\"CitationRef\\\"\\u003e12\\u003c/span\\u003e, \\u003cspan citationid=\\\"CR13\\\" class=\\\"CitationRef\\\"\\u003e13\\u003c/span\\u003e). The system includes one wearable IMU sensor (37 g, 70x40x18 mm), one semi-elastic belt, and a notebook for processing the data via the software programme BTS G-Studio (BTS Bioengineering S.p.A., Italy). Wearable IMU sensor composed of a triaxial gyroscope (16 bit/axes), accelerometer (16 bit/axes), and magnetometer (13 bit\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;1,200 mT)(\\u003cspan citationid=\\\"CR13\\\" class=\\\"CitationRef\\\"\\u003e13\\u003c/span\\u003e). Following the recording of demographic data (age, sex, height, weight, and shoe size), the IMU sensor was fixed at the level of the L5 vertebra using the adjustable belt. Participants were instructed to walk a 4-meter straight path marked with starting and finishing points on the floor at their normal pace after the command \\u0026ldquo;Get ready-Attention-Go\\u0026rdquo; and to stop with the \\u0026ldquo;Stop\\u0026rdquo; command. The software programme computed the following gait parameters: gait speed (m/s), cadence (steps/min), stride length (m), gait cycle duration (s), stance phase (% of gait cycle time), swing phase (% of gait cycle time), double support phase (% of gait cycle time), single support phase (% of gait cycle time). The system provides discrete measurement values for both the right and left lower limbs. To create a representative gait pattern, spatio-temporal parameters from both the right and left sides were averaged across participants. However, we provided the findings indicating both sides in the Supplementary File (see Table \\u003cspan refid=\\\"MOESM1\\\" class=\\\"InternalRef\\\"\\u003eS1\\u003c/span\\u003e\\u0026amp;2, Figure \\u003cspan refid=\\\"MOESM1\\\" class=\\\"InternalRef\\\"\\u003eS1\\u003c/span\\u003e).\\u003c/p\\u003e\\n\\u003ch3\\u003eGait analysis procedure with pose estimation algorithm (YOLOv11) based system\\u003c/h3\\u003e\\n\\u003cp\\u003eSynchronized video recordings taken concurrently with BTS G-WALK assessments were used for pose estimation-based gait analysis using a video camera (iPhone 15) HD, 30 FPS, placed 4.5 meters from the participant. The YOLOv11 algorithm was used for keypoint extractions for spatio-temporal gait variables. YOLOv11 is a fast single-stage algorithm structure based on a \\u0026ldquo;top-down\\u0026rdquo; approach (\\u003cspan citationid=\\\"CR25\\\" class=\\\"CitationRef\\\"\\u003e25\\u003c/span\\u003e). It detects people in the image using bounding boxes, then estimates separate human-specific keypoints for each box (\\u003cspan citationid=\\\"CR25\\\" class=\\\"CitationRef\\\"\\u003e25\\u003c/span\\u003e). By tracking the projection of the determined keypoints on the X and Y axes, the movement of the keypoints in two dimensions can be tracked as a time series. In our process, videos were cropped simultaneously according to the command voice from \\u0026ldquo;Get ready-Attention-Go\\u0026rdquo; to \\u0026ldquo;Stop\\u0026rdquo; at first. The YOLOv11 algorithm was used with a default confidence threshold of 0.25. The algorithm detected and tracked 17 key joint landmarks in 2D during walking sequences. The initial contact and toe-off phases of the gait cycle were identified using the X and Y coordinates of the right and left ankle keypoints. Initial contact was determined from the first derivative of the position data, while toe-off was identified from the second derivative (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig2\\\" class=\\\"InternalRef\\\"\\u003e2\\u003c/span\\u003e). This custom algorithm, developed based on key events, was used to calculate these spatio-temporal gait characteristics. The algorithm extracted the same spatio-temporal gait parameters as the G-walk system for gait speed (m/s), cadence (steps/min), stride length (m), gait cycle duration (s), stance phase (% of gait cycle time), swing phase (% of gait cycle time), double support phase (% of gait cycle time), single support phase (% of gait cycle time). To create a representative gait pattern, spatio-temporal parameters from both the right and left sides were averaged across participants. All algorithm designs and the processes mentioned were implemented using Python 3.8.10.\\u003c/p\\u003e\\u003cp\\u003e\\u003c/p\\u003e\\u003cdiv id=\\\"Sec7\\\" class=\\\"Section2\\\"\\u003e\\u003ch2\\u003eStatistical analysis\\u003c/h2\\u003e\\u003cp\\u003eDemographic data of the participants were presented as means and standard deviations, while categorical data were summarized as frequencies and percentages. The concurrent validity of the YOLOv11 algorithm was assessed by evaluating its consistency and agreement with the IMU-based G-walk system. Pearson correlation analysis was used for consistency, while Bland-Altman plots were used to assess agreement between the two systems. Pearson correlation coefficient was classified as follows: \\u0026lt;0.20 poor, 0.20\\u0026ndash;0.39 weak, 0.40\\u0026ndash;0.59 moderate, 0.60\\u0026ndash;0.79 strong, \\u0026ge;\\u0026thinsp;0.80 excellent consistency (\\u003cspan citationid=\\\"CR26\\\" class=\\\"CitationRef\\\"\\u003e26\\u003c/span\\u003e). The agreement between two systems was assessed using 95% limits of agreement (LoA), calculated as the mean difference between the systems plus or minus 1.96 standard deviations. This approach accounts for both systematic error (mean bias) and random error (differences around the mean for each participant). To express the 95% LoA as a percentage (LoA%), the LoA range was divided by each average gait parameter across both systems. Absolute agreement was interpreted using LoA% with categories: excellent (0-4.9%), good (5-9.9%), moderate (10-49.9%), and poor (\\u0026gt;\\u0026thinsp;50%) (\\u003cspan citationid=\\\"CR27\\\" class=\\\"CitationRef\\\"\\u003e27\\u003c/span\\u003e, \\u003cspan citationid=\\\"CR28\\\" class=\\\"CitationRef\\\"\\u003e28\\u003c/span\\u003e). Further, intra-class correlation coefficients (ICCs) were calculated to provide reliability statistics for the YOLOv11 algorithm. Test-retest reliability was assessed using a two-way mixed-effects model with absolute agreement ICC(\\u003cspan citationid=\\\"CR3\\\" class=\\\"CitationRef\\\"\\u003e3\\u003c/span\\u003e, \\u003cspan citationid=\\\"CR1\\\" class=\\\"CitationRef\\\"\\u003e1\\u003c/span\\u003e), as all measurements were performed twice by the same researcher. The ICC values were interpreted as follows: \\u0026gt;0.90 excellent, 0.75\\u0026ndash;0.90 good, 0.50\\u0026ndash;0.75 moderate, \\u0026lt;\\u0026thinsp;0.50 poor reliability (\\u003cspan citationid=\\\"CR29\\\" class=\\\"CitationRef\\\"\\u003e29\\u003c/span\\u003e). All analyses were conducted using SPSS IBM 27. The level of significance was set as 0.05.\\u003c/p\\u003e\\u003c/div\\u003e\"},{\"header\":\"Results\",\"content\":\"\\u003cp\\u003eA total of 52 participants were assessed for eligibility, of whom 21 were excluded due to (n\\u0026thinsp;=\\u0026thinsp;4) stroke-related gait abnormalities, (n\\u0026thinsp;=\\u0026thinsp;9) not willing to participate, and (n\\u0026thinsp;=\\u0026thinsp;8) scores in MMSE\\u0026thinsp;\\u0026le;\\u0026thinsp;23. Thirty-one community-dwelling older adults (21 women and 10 men, aged 66.9\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;6.4 years, BMI\\u0026thinsp;=\\u0026thinsp;26.7\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;4.6 kg/m\\u003csup\\u003e2\\u003c/sup\\u003e) were included in the study. Table\\u0026nbsp;\\u003cspan refid=\\\"Tab1\\\" class=\\\"InternalRef\\\"\\u003e1\\u003c/span\\u003e summarizes the descriptive characteristics of the participants, including the mean number of chronic diseases (1.39), MMSE (28.06), and TUG (8.28 s) (Table\\u0026nbsp;\\u003cspan refid=\\\"Tab1\\\" class=\\\"InternalRef\\\"\\u003e1\\u003c/span\\u003e).\\u003c/p\\u003e\\u003cp\\u003e\\u003cdiv class=\\\"gridtable\\\"\\u003e\\u003ctable float=\\\"Yes\\\" id=\\\"Tab1\\\" border=\\\"1\\\"\\u003e\\u003ccaption language=\\\"En\\\"\\u003e\\u003cdiv class=\\\"CaptionNumber\\\"\\u003eTable 1\\u003c/div\\u003e\\u003cdiv class=\\\"CaptionContent\\\"\\u003e\\u003cp\\u003eDescriptive characteristics of the participants\\u003c/p\\u003e\\u003c/div\\u003e\\u003c/caption\\u003e\\u003ccolgroup cols=\\\"2\\\"\\u003e\\u003cdiv align=\\\"left\\\" class=\\\"colspec\\\" colname=\\\"c1\\\" colnum=\\\"1\\\"\\u003e\\u003c/div\\u003e\\u003cdiv align=\\\"char\\\" char=\\\"\\u0026plusmn;\\\" class=\\\"colspec\\\" colname=\\\"c2\\\" colnum=\\\"2\\\"\\u003e\\u003c/div\\u003e\\u003cthead\\u003e\\u003ctr\\u003e\\u003cth align=\\\"left\\\" colname=\\\"c1\\\"\\u003e\\u003cp\\u003eVariable\\u003c/p\\u003e\\u003c/th\\u003e\\u003cth align=\\\"left\\\" colname=\\\"c2\\\"\\u003e\\u003cp\\u003eMean\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;SD\\u003c/p\\u003e\\u003c/th\\u003e\\u003c/tr\\u003e\\u003c/thead\\u003e\\u003ctbody\\u003e\\u003ctr\\u003e\\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e\\u003cp\\u003eAge, years\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"char\\\" char=\\\"\\u0026plusmn;\\\" colname=\\\"c2\\\"\\u003e\\u003cp\\u003e66.94\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;6.39\\u003c/p\\u003e\\u003c/td\\u003e\\u003c/tr\\u003e\\u003ctr\\u003e\\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e\\u003cp\\u003eGender, n (%)\\u003c/p\\u003e\\u003cp\\u003e\\u003cem\\u003e-women\\u003c/em\\u003e\\u003c/p\\u003e\\u003cp\\u003e\\u003cem\\u003e-men\\u003c/em\\u003e\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e\\u003cp\\u003e\\u003c/br\\u003e\\u003c/p\\u003e\\u003cp\\u003e21 (67.7)\\u003c/p\\u003e\\u003cp\\u003e10 (32.3)\\u003c/p\\u003e\\u003c/td\\u003e\\u003c/tr\\u003e\\u003ctr\\u003e\\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e\\u003cp\\u003eWeight, kg\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"char\\\" char=\\\"\\u0026plusmn;\\\" colname=\\\"c2\\\"\\u003e\\u003cp\\u003e72.84\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;13.18\\u003c/p\\u003e\\u003c/td\\u003e\\u003c/tr\\u003e\\u003ctr\\u003e\\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e\\u003cp\\u003eHeight, cm\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"char\\\" char=\\\"\\u0026plusmn;\\\" colname=\\\"c2\\\"\\u003e\\u003cp\\u003e165.32\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;9.00\\u003c/p\\u003e\\u003c/td\\u003e\\u003c/tr\\u003e\\u003ctr\\u003e\\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e\\u003cp\\u003eBMI, kg/m\\u003csup\\u003e2\\u003c/sup\\u003e\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"char\\\" char=\\\"\\u0026plusmn;\\\" colname=\\\"c2\\\"\\u003e\\u003cp\\u003e26.68\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;4.65\\u003c/p\\u003e\\u003c/td\\u003e\\u003c/tr\\u003e\\u003ctr\\u003e\\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e\\u003cp\\u003eNumber of chronic diseases, score 0\\u0026ndash;11\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"char\\\" char=\\\"\\u0026plusmn;\\\" colname=\\\"c2\\\"\\u003e\\u003cp\\u003e1.39\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;1.17\\u003c/p\\u003e\\u003c/td\\u003e\\u003c/tr\\u003e\\u003ctr\\u003e\\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e\\u003cp\\u003eMMSE, score 0\\u0026ndash;30\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"char\\\" char=\\\"\\u0026plusmn;\\\" colname=\\\"c2\\\"\\u003e\\u003cp\\u003e28.06\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;2.51\\u003c/p\\u003e\\u003c/td\\u003e\\u003c/tr\\u003e\\u003ctr\\u003e\\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e\\u003cp\\u003eTUG, s\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"char\\\" char=\\\"\\u0026plusmn;\\\" colname=\\\"c2\\\"\\u003e\\u003cp\\u003e8.28\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;2.06\\u003c/p\\u003e\\u003c/td\\u003e\\u003c/tr\\u003e\\u003c/tbody\\u003e\\u003c/colgroup\\u003e\\u003c/table\\u003e\\u003c/div\\u003e\\u003c/p\\u003e\\u003cp\\u003e\\u003cstrong\\u003e\\u003cem\\u003eAbbreviations:\\u0026nbsp;\\u003c/em\\u003e\\u003c/strong\\u003eSD, Standard deviation; BMI, Body mass index; MMSE, Mini-mental state examination; TUG, Timed up and go.\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cbr\\u003e\\u003c/p\\u003e\\u003cp\\u003eTable\\u0026nbsp;\\u003cspan refid=\\\"Tab2\\\" class=\\\"InternalRef\\\"\\u003e2\\u003c/span\\u003e and Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig3\\\" class=\\\"InternalRef\\\"\\u003e3\\u003c/span\\u003e show validity statistics of the YOLO11 algorithm to determine spatio-temporal gait parameters compared to the G-walk system via correlation analysis and Bland- Altman plots. The highest correlations were found in terms of cadence (r\\u0026thinsp;=\\u0026thinsp;0.95, p\\u0026thinsp;\\u0026lt;\\u0026thinsp;0.001), gait speed (r\\u0026thinsp;=\\u0026thinsp;0.90, p\\u0026thinsp;\\u0026lt;\\u0026thinsp;0.001), and gait cycle duration (r\\u0026thinsp;=\\u0026thinsp;0.95, p\\u0026thinsp;\\u0026lt;\\u0026thinsp;0.001). Strong correlations were found for swing phase (r\\u0026thinsp;=\\u0026thinsp;0.72, p\\u0026thinsp;\\u0026lt;\\u0026thinsp;0.001), stance phase (r\\u0026thinsp;=\\u0026thinsp;0.68, p\\u0026thinsp;\\u0026lt;\\u0026thinsp;0.001), first double support (r\\u0026thinsp;=\\u0026thinsp;0.67, p\\u0026thinsp;\\u0026lt;\\u0026thinsp;0.001), stride length (r\\u0026thinsp;=\\u0026thinsp;0.70, p\\u0026thinsp;\\u0026lt;\\u0026thinsp;0.001), and single support (r\\u0026thinsp;=\\u0026thinsp;0.60, p\\u0026thinsp;=\\u0026thinsp;0.001) parameters. According to Bland Altman plots, LoA% showed moderate agreement (16.75%-47.04%) between the YOLO11 algorithm-based system and the G-walk system to determine swing-stance phases, single support, cadence, gait speed, gait cycle duration, and stride length, while poor agreement (95.60%) was found in terms of first double support (Table\\u0026nbsp;\\u003cspan refid=\\\"Tab2\\\" class=\\\"InternalRef\\\"\\u003e2\\u003c/span\\u003e \\u0026amp; Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig3\\\" class=\\\"InternalRef\\\"\\u003e3\\u003c/span\\u003e).\\u003c/p\\u003e\\u003cp\\u003e\\u003cdiv class=\\\"gridtable\\\"\\u003e\\u003ctable float=\\\"Yes\\\" id=\\\"Tab2\\\" border=\\\"1\\\"\\u003e\\u003ccaption language=\\\"En\\\"\\u003e\\u003cdiv class=\\\"CaptionNumber\\\"\\u003eTable 2\\u003c/div\\u003e\\u003cdiv class=\\\"CaptionContent\\\"\\u003e\\u003cp\\u003eValidity statistics of the YOLOv11 Algorithm-based system compared to G-walk system to determine spatio-temporal gait parameters in older adults\\u003c/p\\u003e\\u003c/div\\u003e\\u003c/caption\\u003e\\u003ccolgroup cols=\\\"8\\\"\\u003e\\u003cdiv align=\\\"left\\\" class=\\\"colspec\\\" colname=\\\"c1\\\" colnum=\\\"1\\\"\\u003e\\u003c/div\\u003e\\u003cdiv align=\\\"char\\\" char=\\\"\\u0026plusmn;\\\" class=\\\"colspec\\\" colname=\\\"c2\\\" colnum=\\\"2\\\"\\u003e\\u003c/div\\u003e\\u003cdiv align=\\\"char\\\" char=\\\"\\u0026plusmn;\\\" class=\\\"colspec\\\" colname=\\\"c3\\\" colnum=\\\"3\\\"\\u003e\\u003c/div\\u003e\\u003cdiv align=\\\"char\\\" char=\\\".\\\" class=\\\"colspec\\\" colname=\\\"c4\\\" colnum=\\\"4\\\"\\u003e\\u003c/div\\u003e\\u003cdiv align=\\\"char\\\" char=\\\".\\\" class=\\\"colspec\\\" colname=\\\"c5\\\" colnum=\\\"5\\\"\\u003e\\u003c/div\\u003e\\u003cdiv align=\\\"left\\\" class=\\\"colspec\\\" colname=\\\"c6\\\" colnum=\\\"6\\\"\\u003e\\u003c/div\\u003e\\u003cdiv align=\\\"char\\\" char=\\\".\\\" class=\\\"colspec\\\" colname=\\\"c7\\\" colnum=\\\"7\\\"\\u003e\\u003c/div\\u003e\\u003cdiv align=\\\"left\\\" class=\\\"colspec\\\" colname=\\\"c8\\\" colnum=\\\"8\\\"\\u003e\\u003c/div\\u003e\\u003cthead\\u003e\\u003ctr\\u003e\\u003cth align=\\\"left\\\" colname=\\\"c1\\\"\\u003e\\u003cp\\u003eParameter\\u003c/p\\u003e\\u003c/th\\u003e\\u003cth align=\\\"left\\\" colname=\\\"c2\\\"\\u003e\\u003cp\\u003eG-walk\\u003c/p\\u003e\\u003cp\\u003eMean\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;SD\\u003c/p\\u003e\\u003c/th\\u003e\\u003cth align=\\\"left\\\" colname=\\\"c3\\\"\\u003e\\u003cp\\u003eYOLOv11 Algorithm\\u003c/p\\u003e\\u003cp\\u003eMean\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;SD\\u003c/p\\u003e\\u003c/th\\u003e\\u003cth align=\\\"left\\\" colname=\\\"c4\\\"\\u003e\\u003cp\\u003er\\u003c/p\\u003e\\u003c/th\\u003e\\u003cth align=\\\"left\\\" colname=\\\"c5\\\"\\u003e\\u003cp\\u003ep\\u003c/p\\u003e\\u003c/th\\u003e\\u003cth align=\\\"left\\\" colname=\\\"c6\\\"\\u003e\\u003cp\\u003eConsistency\\u003c/p\\u003e\\u003c/th\\u003e\\u003cth align=\\\"left\\\" colname=\\\"c7\\\"\\u003e\\u003cp\\u003eLoA%\\u003c/p\\u003e\\u003c/th\\u003e\\u003cth align=\\\"left\\\" colname=\\\"c8\\\"\\u003e\\u003cp\\u003eAgreement\\u003c/p\\u003e\\u003c/th\\u003e\\u003c/tr\\u003e\\u003c/thead\\u003e\\u003ctbody\\u003e\\u003ctr\\u003e\\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e\\u003cp\\u003eSwing phase, % GCT\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"char\\\" char=\\\"\\u0026plusmn;\\\" colname=\\\"c2\\\"\\u003e\\u003cp\\u003e39.38\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;3.08\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"char\\\" char=\\\"\\u0026plusmn;\\\" colname=\\\"c3\\\"\\u003e\\u003cp\\u003e40.23\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;3.26\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e\\u003cp\\u003e0.72\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c5\\\"\\u003e\\u003cp\\u003e\\u0026lt;\\u0026thinsp;0.001*\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e\\u003cp\\u003eStrong\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c7\\\"\\u003e\\u003cp\\u003e23.34\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"left\\\" colname=\\\"c8\\\"\\u003e\\u003cp\\u003eModerate\\u003c/p\\u003e\\u003c/td\\u003e\\u003c/tr\\u003e\\u003ctr\\u003e\\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e\\u003cp\\u003eStance phase, % GCT\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"char\\\" char=\\\"\\u0026plusmn;\\\" colname=\\\"c2\\\"\\u003e\\u003cp\\u003e60.73\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;3.22\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"char\\\" char=\\\"\\u0026plusmn;\\\" colname=\\\"c3\\\"\\u003e\\u003cp\\u003e59.76\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;3.26\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e\\u003cp\\u003e0.68\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c5\\\"\\u003e\\u003cp\\u003e\\u0026lt;\\u0026thinsp;0.001*\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e\\u003cp\\u003eStrong\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c7\\\"\\u003e\\u003cp\\u003e16.78\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"left\\\" colname=\\\"c8\\\"\\u003e\\u003cp\\u003eModerate\\u003c/p\\u003e\\u003c/td\\u003e\\u003c/tr\\u003e\\u003ctr\\u003e\\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e\\u003cp\\u003eFirst double support, % GCT\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"char\\\" char=\\\"\\u0026plusmn;\\\" colname=\\\"c2\\\"\\u003e\\u003cp\\u003e10.36\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;2.78\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"char\\\" char=\\\"\\u0026plusmn;\\\" colname=\\\"c3\\\"\\u003e\\u003cp\\u003e10.06\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;3.26\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e\\u003cp\\u003e0.67\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c5\\\"\\u003e\\u003cp\\u003e\\u0026lt;\\u0026thinsp;0.001*\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e\\u003cp\\u003eStrong\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c7\\\"\\u003e\\u003cp\\u003e95.60\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"left\\\" colname=\\\"c8\\\"\\u003e\\u003cp\\u003ePoor\\u003c/p\\u003e\\u003c/td\\u003e\\u003c/tr\\u003e\\u003ctr\\u003e\\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e\\u003cp\\u003eSingle support, % GCT\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"char\\\" char=\\\"\\u0026plusmn;\\\" colname=\\\"c2\\\"\\u003e\\u003cp\\u003e39.54\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;2.84\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"char\\\" char=\\\"\\u0026plusmn;\\\" colname=\\\"c3\\\"\\u003e\\u003cp\\u003e39.98\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;3.23\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e\\u003cp\\u003e0.60\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c5\\\"\\u003e\\u003cp\\u003e0.001*\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e\\u003cp\\u003eStrong\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c7\\\"\\u003e\\u003cp\\u003e27.05\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"left\\\" colname=\\\"c8\\\"\\u003e\\u003cp\\u003eModerate\\u003c/p\\u003e\\u003c/td\\u003e\\u003c/tr\\u003e\\u003ctr\\u003e\\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e\\u003cp\\u003eCadance, steps/min\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"char\\\" char=\\\"\\u0026plusmn;\\\" colname=\\\"c2\\\"\\u003e\\u003cp\\u003e105.28\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;16.00\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"char\\\" char=\\\"\\u0026plusmn;\\\" colname=\\\"c3\\\"\\u003e\\u003cp\\u003e102.97\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;14\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e\\u003cp\\u003e0.95\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c5\\\"\\u003e\\u003cp\\u003e\\u0026lt;\\u0026thinsp;0.001*\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e\\u003cp\\u003eExcellent\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c7\\\"\\u003e\\u003cp\\u003e19.57\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"left\\\" colname=\\\"c8\\\"\\u003e\\u003cp\\u003eModerate\\u003c/p\\u003e\\u003c/td\\u003e\\u003c/tr\\u003e\\u003ctr\\u003e\\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e\\u003cp\\u003eGait speed, m/s\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"char\\\" char=\\\"\\u0026plusmn;\\\" colname=\\\"c2\\\"\\u003e\\u003cp\\u003e1.06\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.27\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"char\\\" char=\\\"\\u0026plusmn;\\\" colname=\\\"c3\\\"\\u003e\\u003cp\\u003e0.94\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.21\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e\\u003cp\\u003e0.90\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c5\\\"\\u003e\\u003cp\\u003e\\u0026lt;\\u0026thinsp;0.001*\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e\\u003cp\\u003eExcellent\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c7\\\"\\u003e\\u003cp\\u003e47.04\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"left\\\" colname=\\\"c8\\\"\\u003e\\u003cp\\u003eModerate\\u003c/p\\u003e\\u003c/td\\u003e\\u003c/tr\\u003e\\u003ctr\\u003e\\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e\\u003cp\\u003eGait cycle duration, s\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"char\\\" char=\\\"\\u0026plusmn;\\\" colname=\\\"c2\\\"\\u003e\\u003cp\\u003e1.16\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.17\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"char\\\" char=\\\"\\u0026plusmn;\\\" colname=\\\"c3\\\"\\u003e\\u003cp\\u003e1.18\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.16\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e\\u003cp\\u003e0.95\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c5\\\"\\u003e\\u003cp\\u003e\\u0026lt;\\u0026thinsp;0.001*\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e\\u003cp\\u003eExcellent\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c7\\\"\\u003e\\u003cp\\u003e16.75\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"left\\\" colname=\\\"c8\\\"\\u003e\\u003cp\\u003eModerate\\u003c/p\\u003e\\u003c/td\\u003e\\u003c/tr\\u003e\\u003ctr\\u003e\\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e\\u003cp\\u003eStride length, m\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"char\\\" char=\\\"\\u0026plusmn;\\\" colname=\\\"c2\\\"\\u003e\\u003cp\\u003e1.20\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.20\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"char\\\" char=\\\"\\u0026plusmn;\\\" colname=\\\"c3\\\"\\u003e\\u003cp\\u003e1.24\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.17\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e\\u003cp\\u003e0.70\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c5\\\"\\u003e\\u003cp\\u003e\\u0026lt;\\u0026thinsp;0.001*\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e\\u003cp\\u003eStrong\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c7\\\"\\u003e\\u003cp\\u003e44.98\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"left\\\" colname=\\\"c8\\\"\\u003e\\u003cp\\u003eModerate\\u003c/p\\u003e\\u003c/td\\u003e\\u003c/tr\\u003e\\u003c/tbody\\u003e\\u003c/colgroup\\u003e\\u003c/table\\u003e\\u003c/div\\u003e\\u003c/p\\u003e\\u003cp\\u003e\\u003c/p\\u003e\\u003cp\\u003e\\u003cstrong\\u003e\\u003cem\\u003eAbbreviations:\\u003c/em\\u003e\\u003c/strong\\u003e SD, standard deviation; LoA, Limit of agreement; GCT, Gait cycle time; r, Pearson correlation coefficient; *p\\u0026lt;0.05 in Pearson Correlation Analysis.\\u0026nbsp;\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cbr\\u003e\\u003c/p\\u003e\\u003cp\\u003eThe test-retest reliability analysis of the YOLOv11 algorithm-based system determining spatio-temporal parameters of gait in older adults is summarized in Table\\u0026nbsp;\\u003cspan refid=\\\"Tab3\\\" class=\\\"InternalRef\\\"\\u003e3\\u003c/span\\u003e. The findings showed excellent reliability for cadence (ICC\\u0026thinsp;=\\u0026thinsp;0.92[0.84\\u0026ndash;0.96], p\\u0026thinsp;\\u0026lt;\\u0026thinsp;0.001) and gait speed (ICC\\u0026thinsp;=\\u0026thinsp;0.91[0.82\\u0026ndash;0.96], p\\u0026thinsp;\\u0026lt;\\u0026thinsp;0.001), while good reliability was found for gait cycle duration (ICC\\u0026thinsp;=\\u0026thinsp;0.89[0.77\\u0026ndash;0.95], p\\u0026thinsp;\\u0026lt;\\u0026thinsp;0.001) and stride length (ICC\\u0026thinsp;=\\u0026thinsp;0.90[0.78\\u0026ndash;0.95], p\\u0026thinsp;\\u0026lt;\\u0026thinsp;0.001). For the remaining parameters as the phases of gait had moderate reliability (0.59\\u0026thinsp;\\u0026lt;\\u0026thinsp;ICC\\u0026thinsp;\\u0026lt;\\u0026thinsp;0.61, p\\u0026thinsp;\\u0026lt;\\u0026thinsp;0.05) (Table\\u0026nbsp;\\u003cspan refid=\\\"Tab3\\\" class=\\\"InternalRef\\\"\\u003e3\\u003c/span\\u003e).\\u003c/p\\u003e\\u003cp\\u003e\\u003cdiv class=\\\"gridtable\\\"\\u003e\\u003ctable float=\\\"Yes\\\" id=\\\"Tab3\\\" border=\\\"1\\\"\\u003e\\u003ccaption language=\\\"En\\\"\\u003e\\u003cdiv class=\\\"CaptionNumber\\\"\\u003eTable 3\\u003c/div\\u003e\\u003cdiv class=\\\"CaptionContent\\\"\\u003e\\u003cp\\u003eThe reliability analysis of YOLOv11 algorithm-based system determining spatio-temporal characteristics of gait in older adults\\u003c/p\\u003e\\u003c/div\\u003e\\u003c/caption\\u003e\\u003ccolgroup cols=\\\"8\\\"\\u003e\\u003cdiv align=\\\"left\\\" class=\\\"colspec\\\" colname=\\\"c1\\\" colnum=\\\"1\\\"\\u003e\\u003c/div\\u003e\\u003cdiv align=\\\"char\\\" char=\\\"\\u0026plusmn;\\\" class=\\\"colspec\\\" colname=\\\"c2\\\" colnum=\\\"2\\\"\\u003e\\u003c/div\\u003e\\u003cdiv align=\\\"char\\\" char=\\\"\\u0026plusmn;\\\" class=\\\"colspec\\\" colname=\\\"c3\\\" colnum=\\\"3\\\"\\u003e\\u003c/div\\u003e\\u003cdiv align=\\\"char\\\" char=\\\".\\\" class=\\\"colspec\\\" colname=\\\"c4\\\" colnum=\\\"4\\\"\\u003e\\u003c/div\\u003e\\u003cdiv align=\\\"char\\\" char=\\\".\\\" class=\\\"colspec\\\" colname=\\\"c5\\\" colnum=\\\"5\\\"\\u003e\\u003c/div\\u003e\\u003cdiv align=\\\"char\\\" char=\\\".\\\" class=\\\"colspec\\\" colname=\\\"c6\\\" colnum=\\\"6\\\"\\u003e\\u003c/div\\u003e\\u003cdiv align=\\\"left\\\" class=\\\"colspec\\\" colname=\\\"c7\\\" colnum=\\\"7\\\"\\u003e\\u003c/div\\u003e\\u003cdiv align=\\\"char\\\" char=\\\".\\\" class=\\\"colspec\\\" colname=\\\"c8\\\" colnum=\\\"8\\\"\\u003e\\u003c/div\\u003e\\u003cthead\\u003e\\u003ctr\\u003e\\u003cth align=\\\"left\\\" colname=\\\"c1\\\" morerows=\\\"1\\\" rowspan=\\\"2\\\"\\u003e\\u003cp\\u003eParameter\\u003c/p\\u003e\\u003c/th\\u003e\\u003cth align=\\\"left\\\" colname=\\\"c2\\\" morerows=\\\"1\\\" rowspan=\\\"2\\\"\\u003e\\u003cp\\u003eYOLOv11 Test Mean\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;SD\\u003c/p\\u003e\\u003c/th\\u003e\\u003cth align=\\\"left\\\" colname=\\\"c3\\\" morerows=\\\"1\\\" rowspan=\\\"2\\\"\\u003e\\u003cp\\u003eYOLOv11 Retest\\u003c/p\\u003e\\u003cp\\u003eMean\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;SD\\u003c/p\\u003e\\u003c/th\\u003e\\u003cth align=\\\"left\\\" colspan=\\\"2\\\" nameend=\\\"c5\\\" namest=\\\"c4\\\"\\u003e\\u003cp\\u003e95% CI\\u003c/p\\u003e\\u003c/th\\u003e\\u003cth align=\\\"left\\\" colname=\\\"c6\\\" morerows=\\\"1\\\" rowspan=\\\"2\\\"\\u003e\\u003cp\\u003eICC\\u003c/p\\u003e\\u003c/th\\u003e\\u003cth align=\\\"left\\\" colname=\\\"c7\\\" morerows=\\\"1\\\" rowspan=\\\"2\\\"\\u003e\\u003cp\\u003eInterpretation\\u003c/p\\u003e\\u003c/th\\u003e\\u003cth align=\\\"left\\\" colname=\\\"c8\\\" morerows=\\\"1\\\" rowspan=\\\"2\\\"\\u003e\\u003cp\\u003ep\\u003c/p\\u003e\\u003c/th\\u003e\\u003c/tr\\u003e\\u003ctr\\u003e\\u003cth align=\\\"left\\\" colname=\\\"c4\\\"\\u003e\\u003cp\\u003eLower\\u003c/p\\u003e\\u003c/th\\u003e\\u003cth align=\\\"left\\\" colname=\\\"c5\\\"\\u003e\\u003cp\\u003eUpper\\u003c/p\\u003e\\u003c/th\\u003e\\u003c/tr\\u003e\\u003c/thead\\u003e\\u003ctbody\\u003e\\u003ctr\\u003e\\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e\\u003cp\\u003eSwing phase, % GCT\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"char\\\" char=\\\"\\u0026plusmn;\\\" colname=\\\"c2\\\"\\u003e\\u003cp\\u003e40.23\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;3.26\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"char\\\" char=\\\"\\u0026plusmn;\\\" colname=\\\"c3\\\"\\u003e\\u003cp\\u003e40.75\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;3.08\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e\\u003cp\\u003e0.18\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c5\\\"\\u003e\\u003cp\\u003e0.81\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c6\\\"\\u003e\\u003cp\\u003e0.61\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"left\\\" colname=\\\"c7\\\"\\u003e\\u003cp\\u003eModerate\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c8\\\"\\u003e\\u003cp\\u003e0.007*\\u003c/p\\u003e\\u003c/td\\u003e\\u003c/tr\\u003e\\u003ctr\\u003e\\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e\\u003cp\\u003eStance phase, % GCT\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"char\\\" char=\\\"\\u0026plusmn;\\\" colname=\\\"c2\\\"\\u003e\\u003cp\\u003e59.76\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;3.26\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"char\\\" char=\\\"\\u0026plusmn;\\\" colname=\\\"c3\\\"\\u003e\\u003cp\\u003e59.24\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;3.08\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e\\u003cp\\u003e0.18\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c5\\\"\\u003e\\u003cp\\u003e0.81\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c6\\\"\\u003e\\u003cp\\u003e0.61\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"left\\\" colname=\\\"c7\\\"\\u003e\\u003cp\\u003eModerate\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c8\\\"\\u003e\\u003cp\\u003e0.007*\\u003c/p\\u003e\\u003c/td\\u003e\\u003c/tr\\u003e\\u003ctr\\u003e\\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e\\u003cp\\u003eFirst double support, % GCT\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"char\\\" char=\\\"\\u0026plusmn;\\\" colname=\\\"c2\\\"\\u003e\\u003cp\\u003e10.06\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;3.26\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"char\\\" char=\\\"\\u0026plusmn;\\\" colname=\\\"c3\\\"\\u003e\\u003cp\\u003e9.57\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;2.92\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e\\u003cp\\u003e0.15\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c5\\\"\\u003e\\u003cp\\u003e0.80\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c6\\\"\\u003e\\u003cp\\u003e0.59\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"left\\\" colname=\\\"c7\\\"\\u003e\\u003cp\\u003eModerate\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c8\\\"\\u003e\\u003cp\\u003e0.009*\\u003c/p\\u003e\\u003c/td\\u003e\\u003c/tr\\u003e\\u003ctr\\u003e\\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e\\u003cp\\u003eSingle support, % GCT\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"char\\\" char=\\\"\\u0026plusmn;\\\" colname=\\\"c2\\\"\\u003e\\u003cp\\u003e39.98\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;3.23\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"char\\\" char=\\\"\\u0026plusmn;\\\" colname=\\\"c3\\\"\\u003e\\u003cp\\u003e40.39\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;2.90\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e\\u003cp\\u003e0.14\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c5\\\"\\u003e\\u003cp\\u003e0.80\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c6\\\"\\u003e\\u003cp\\u003e0.59\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"left\\\" colname=\\\"c7\\\"\\u003e\\u003cp\\u003eModerate\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c8\\\"\\u003e\\u003cp\\u003e0.01*\\u003c/p\\u003e\\u003c/td\\u003e\\u003c/tr\\u003e\\u003ctr\\u003e\\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e\\u003cp\\u003eCadance, steps/min\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"char\\\" char=\\\"\\u0026plusmn;\\\" colname=\\\"c2\\\"\\u003e\\u003cp\\u003e102.97\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;14\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"char\\\" char=\\\"\\u0026plusmn;\\\" colname=\\\"c3\\\"\\u003e\\u003cp\\u003e105.12\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;13.64\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e\\u003cp\\u003e0.84\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c5\\\"\\u003e\\u003cp\\u003e0.96\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c6\\\"\\u003e\\u003cp\\u003e0.92\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"left\\\" colname=\\\"c7\\\"\\u003e\\u003cp\\u003eExcellent\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c8\\\"\\u003e\\u003cp\\u003e\\u0026lt;\\u0026thinsp;0.001*\\u003c/p\\u003e\\u003c/td\\u003e\\u003c/tr\\u003e\\u003ctr\\u003e\\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e\\u003cp\\u003eGait speed, m/s\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"char\\\" char=\\\"\\u0026plusmn;\\\" colname=\\\"c2\\\"\\u003e\\u003cp\\u003e0.94\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.21\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"char\\\" char=\\\"\\u0026plusmn;\\\" colname=\\\"c3\\\"\\u003e\\u003cp\\u003e0.98\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.21\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e\\u003cp\\u003e0.82\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c5\\\"\\u003e\\u003cp\\u003e0.96\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c6\\\"\\u003e\\u003cp\\u003e0.91\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"left\\\" colname=\\\"c7\\\"\\u003e\\u003cp\\u003eExcellent\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c8\\\"\\u003e\\u003cp\\u003e\\u0026lt;\\u0026thinsp;0.001*\\u003c/p\\u003e\\u003c/td\\u003e\\u003c/tr\\u003e\\u003ctr\\u003e\\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e\\u003cp\\u003eGait cycle duration, s\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"char\\\" char=\\\"\\u0026plusmn;\\\" colname=\\\"c2\\\"\\u003e\\u003cp\\u003e1.18\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.16\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"char\\\" char=\\\"\\u0026plusmn;\\\" colname=\\\"c3\\\"\\u003e\\u003cp\\u003e1.15\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.14\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e\\u003cp\\u003e0.77\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c5\\\"\\u003e\\u003cp\\u003e0.95\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c6\\\"\\u003e\\u003cp\\u003e0.89\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"left\\\" colname=\\\"c7\\\"\\u003e\\u003cp\\u003eGood\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c8\\\"\\u003e\\u003cp\\u003e\\u0026lt;\\u0026thinsp;0.001*\\u003c/p\\u003e\\u003c/td\\u003e\\u003c/tr\\u003e\\u003ctr\\u003e\\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e\\u003cp\\u003eStride length, m\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"char\\\" char=\\\"\\u0026plusmn;\\\" colname=\\\"c2\\\"\\u003e\\u003cp\\u003e1.24\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.17\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"char\\\" char=\\\"\\u0026plusmn;\\\" colname=\\\"c3\\\"\\u003e\\u003cp\\u003e1.27\\u0026thinsp;\\u0026plusmn;\\u0026thinsp;0.16\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c4\\\"\\u003e\\u003cp\\u003e0.78\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c5\\\"\\u003e\\u003cp\\u003e0.95\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c6\\\"\\u003e\\u003cp\\u003e0.90\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"left\\\" colname=\\\"c7\\\"\\u003e\\u003cp\\u003eGood\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"char\\\" char=\\\".\\\" colname=\\\"c8\\\"\\u003e\\u003cp\\u003e\\u0026lt;\\u0026thinsp;0.001*\\u003c/p\\u003e\\u003c/td\\u003e\\u003c/tr\\u003e\\u003c/tbody\\u003e\\u003c/colgroup\\u003e\\u003c/table\\u003e\\u003c/div\\u003e\\u003c/p\\u003e\\u003cp\\u003e\\u003cstrong\\u003e\\u003cem\\u003eAbbreviations:\\u003c/em\\u003e\\u003c/strong\\u003e SD, standard deviation; CI, confidence interval; ICC, Intra-class correlation coefficient; GCT, Gait cycle time; *p\\u0026lt;0.05 in reliability analysis.\\u003c/p\\u003e\\u003cp\\u003eDetailed analyses for the right and left sides were provided in the Supplementary file (see Table \\u003cspan refid=\\\"MOESM1\\\" class=\\\"InternalRef\\\"\\u003eS1\\u003c/span\\u003e\\u0026amp;2, Figure \\u003cspan refid=\\\"MOESM1\\\" class=\\\"InternalRef\\\"\\u003eS1\\u003c/span\\u003e).\\u003c/p\\u003e\"},{\"header\":\"Discussion\",\"content\":\"\\u003cp\\u003eIn this study, we established the clinical validity and reliability of the YOLOv11 algorithm-based system by assessing its consistency and agreement with the G-walk system. Our findings indicate that the YOLOv11 algorithm-based system demonstrated excellent to moderate consistency and moderate agreement with the G-walk system, alongside excellent to moderate test-retest reliability.\\u003c/p\\u003e\\u003cp\\u003eIn terms of clinical validity of the YOLOv11 algorithm-based system, we referenced the G-walk system due to general clinical use of current studies, both healthy older adults and other populations such as Parkinson\\u0026rsquo;s disease, dementia, knee osteoarthritis, etc. (\\u003cspan citationid=\\\"CR12\\\" class=\\\"CitationRef\\\"\\u003e12\\u003c/span\\u003e, \\u003cspan additionalcitationids=\\\"CR31\\\" citationid=\\\"CR30\\\" class=\\\"CitationRef\\\"\\u003e30\\u003c/span\\u003e\\u0026ndash;\\u003cspan citationid=\\\"CR32\\\" class=\\\"CitationRef\\\"\\u003e32\\u003c/span\\u003e). Our findings revealed that spatio-temporal gait parameters calculated from the single video with the YOLOv11 algorithm-based system have excellent to strong correlation with the G-walk system. However, all spatio-temporal gait parameters had moderate agreement between the systems, except for the first double support with poor agreement. Previous studies regarding the validation of video pose tracking algorithms mainly used OpenPose, AlphaPose, Detectron, and PoseNet algorithms or a customized algorithm for a multiple camera system in community-dwelling older adults (\\u003cspan additionalcitationids=\\\"CR16 CR17\\\" citationid=\\\"CR15\\\" class=\\\"CitationRef\\\"\\u003e15\\u003c/span\\u003e\\u0026ndash;\\u003cspan citationid=\\\"CR18\\\" class=\\\"CitationRef\\\"\\u003e18\\u003c/span\\u003e) (Table\\u0026nbsp;\\u003cspan refid=\\\"Tab4\\\" class=\\\"InternalRef\\\"\\u003e4\\u003c/span\\u003e). Ripic et al. used an eight-camera based Kinatrax system within a customized algorithm to identify spatio-temporal gait parameters in older adults, and they found excellent agreement with the Nexus motion capture system (\\u003cspan citationid=\\\"CR17\\\" class=\\\"CitationRef\\\"\\u003e17\\u003c/span\\u003e). Authors expanded their previous report, which has calculated gait phases according to the ankle key-point and retrain their algorithm, including the calcanei key point (\\u003cspan citationid=\\\"CR17\\\" class=\\\"CitationRef\\\"\\u003e17\\u003c/span\\u003e, \\u003cspan citationid=\\\"CR33\\\" class=\\\"CitationRef\\\"\\u003e33\\u003c/span\\u003e). This retraining procedure has led to their findings of better agreement in variables of stride width, step time, stance time, swing time, and double limb support (\\u003cspan citationid=\\\"CR17\\\" class=\\\"CitationRef\\\"\\u003e17\\u003c/span\\u003e). In our results, we had sufficient correlation scores in terms of consistency, while agreement was moderate for spatio-temporal gait parameters which is obviously lower than the aforementioned study\\u0026rsquo;s findings. The main cause of this discrepancy is that we followed a minimalist methodology with a single video camera to assess clinical validity of the pose estimation-based analysis system. Additionally, Nexus and Kinatrax have a similar technology-based approach for gait analysis (\\u003cspan citationid=\\\"CR17\\\" class=\\\"CitationRef\\\"\\u003e17\\u003c/span\\u003e), while we had two different approaches IMU based G-walk system and YOLOv11 algorithms-based system. This methodology has offered paradoxical effects encompassing both advantages and disadvantages. The advantage is that this could lead us to a new path for gait analysis in older adults, regardless of location. Further, if the model expands and is adapted through a smartphone application, access to gait analysis, tracking rehabilitation processes, and a clinical easy-to-use system could be possible in elderly care centers, nursing homes, or rehabilitation clinics without a lab-based environment. This kind of approach helps prevent the \\u0026ldquo;inverse care law\\u0026rdquo; phenomenon, where disadvantaged populations are underrepresented in scientific studies compared to healthy individuals, especially for older people (\\u003cspan citationid=\\\"CR34\\\" class=\\\"CitationRef\\\"\\u003e34\\u003c/span\\u003e). In the other hand, we should underline the disadvantage of high correlation coefficients indicating consistent ranking across G-walk system and YOLOv11 algorithm-based system, while the limits of agreements between absolute values found wider than expected (moderate agreements LoA%=16.75%-47.04% for swing-stance phases, single support, cadence, gait speed, gait cycle duration, and stride length; poor agreement LoA%=95.60% for first double support). This situation points to random error variation rather than systematic bias. The fundamental source of random error lies in the nature of the two technologies being compared. While the G-walk system collects motion information using an inertial measurement sensor, the YOLOv11-based system obtains data through pixel-based geometric estimates, predicting from a single 2D video. However, it is still promising that the algorithm-based system has the ability to capture trends in metrics showing moderate levels of agreement with the G-walk system.\\u003c/p\\u003e\\u003cp\\u003e\\u003cdiv class=\\\"gridtable\\\"\\u003e\\u003ctable float=\\\"Yes\\\" id=\\\"Tab4\\\" border=\\\"1\\\"\\u003e\\u003ccaption language=\\\"En\\\"\\u003e\\u003cdiv class=\\\"CaptionNumber\\\"\\u003eTable 4\\u003c/div\\u003e\\u003cdiv class=\\\"CaptionContent\\\"\\u003e\\u003cp\\u003eValidity and reliability studies of pose estimation methodology to conduct gait analyses in older adults\\u003c/p\\u003e\\u003c/div\\u003e\\u003c/caption\\u003e\\u003ccolgroup cols=\\\"11\\\"\\u003e\\u003cdiv align=\\\"left\\\" class=\\\"colspec\\\" colname=\\\"c1\\\" colnum=\\\"1\\\"\\u003e\\u003c/div\\u003e\\u003cdiv align=\\\"left\\\" class=\\\"colspec\\\" colname=\\\"c2\\\" colnum=\\\"2\\\"\\u003e\\u003c/div\\u003e\\u003cdiv align=\\\"left\\\" class=\\\"colspec\\\" colname=\\\"c3\\\" colnum=\\\"3\\\"\\u003e\\u003c/div\\u003e\\u003cdiv align=\\\"left\\\" class=\\\"colspec\\\" colname=\\\"c4\\\" colnum=\\\"4\\\"\\u003e\\u003c/div\\u003e\\u003cdiv align=\\\"left\\\" class=\\\"colspec\\\" colname=\\\"c5\\\" colnum=\\\"5\\\"\\u003e\\u003c/div\\u003e\\u003cdiv align=\\\"left\\\" class=\\\"colspec\\\" colname=\\\"c6\\\" colnum=\\\"6\\\"\\u003e\\u003c/div\\u003e\\u003cdiv align=\\\"left\\\" class=\\\"colspec\\\" colname=\\\"c7\\\" colnum=\\\"7\\\"\\u003e\\u003c/div\\u003e\\u003cdiv align=\\\"left\\\" class=\\\"colspec\\\" colname=\\\"c8\\\" colnum=\\\"8\\\"\\u003e\\u003c/div\\u003e\\u003cdiv align=\\\"left\\\" class=\\\"colspec\\\" colname=\\\"c9\\\" colnum=\\\"9\\\"\\u003e\\u003c/div\\u003e\\u003cdiv align=\\\"left\\\" class=\\\"colspec\\\" colname=\\\"c10\\\" colnum=\\\"10\\\"\\u003e\\u003c/div\\u003e\\u003cdiv align=\\\"left\\\" class=\\\"colspec\\\" colname=\\\"c11\\\" colnum=\\\"11\\\"\\u003e\\u003c/div\\u003e\\u003cthead\\u003e\\u003ctr\\u003e\\u003cth align=\\\"left\\\" colname=\\\"c1\\\"\\u003e\\u003cp\\u003eStudy\\u003c/p\\u003e\\u003c/th\\u003e\\u003cth align=\\\"left\\\" colname=\\\"c2\\\"\\u003e\\u003cp\\u003eSubject\\u003c/p\\u003e\\u003c/th\\u003e\\u003cth align=\\\"left\\\" colname=\\\"c3\\\"\\u003e\\u003cp\\u003eTest Condition\\u003c/p\\u003e\\u003c/th\\u003e\\u003cth align=\\\"left\\\" colname=\\\"c4\\\"\\u003e\\u003cp\\u003ePose Estimation Algorithm\\u003c/p\\u003e\\u003c/th\\u003e\\u003cth align=\\\"left\\\" colname=\\\"c5\\\"\\u003e\\u003cp\\u003eReference system\\u003c/p\\u003e\\u003c/th\\u003e\\u003cth align=\\\"left\\\" colname=\\\"c6\\\"\\u003e\\u003cp\\u003eParameter\\u003c/p\\u003e\\u003c/th\\u003e\\u003cth align=\\\"left\\\" colname=\\\"c7\\\"\\u003e\\u003cp\\u003eTest-retest reliability\\u003c/p\\u003e\\u003c/th\\u003e\\u003cth align=\\\"left\\\" colname=\\\"c8\\\"\\u003e\\u003cp\\u003eLoA for gait speed (m/s)\\u003c/p\\u003e\\u003c/th\\u003e\\u003cth align=\\\"left\\\" colname=\\\"c9\\\"\\u003e\\u003cp\\u003eLoA for Stride Length (m)\\u003c/p\\u003e\\u003c/th\\u003e\\u003cth align=\\\"left\\\" colname=\\\"c10\\\"\\u003e\\u003cp\\u003eLoA for Stance Phase\\u003c/p\\u003e\\u003cp\\u003e(s or %)\\u003c/p\\u003e\\u003c/th\\u003e\\u003cth align=\\\"left\\\" colname=\\\"c11\\\"\\u003e\\u003cp\\u003eCorrelations with Reference System\\u003c/p\\u003e\\u003c/th\\u003e\\u003c/tr\\u003e\\u003c/thead\\u003e\\u003ctbody\\u003e\\u003ctr\\u003e\\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e\\u003cp\\u003eCarvalho, 2024\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e\\u003cp\\u003eOlder people (n\\u0026thinsp;=\\u0026thinsp;29),\\u003c/p\\u003e\\u003cp\\u003eMean Age\\u0026thinsp;=\\u0026thinsp;75.2 years\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"left\\\" colname=\\\"c3\\\"\\u003e\\u003cp\\u003e-Qualisys Track Manager with 8 video cameras, synchronized with 3 force plates on 12-meter walking path\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"left\\\" colname=\\\"c4\\\"\\u003e\\u003cp\\u003e-Theia3D markerless motion capture\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e\\u003cp\\u003eNA\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e\\u003cp\\u003eKinematic analysis \\u0026amp;phases of gait\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"left\\\" colname=\\\"c7\\\"\\u003e\\u003cp\\u003eKinematic analysis\\u003c/p\\u003e\\u003cp\\u003eICC:0.53\\u0026ndash;0.88\\u003c/p\\u003e\\u003cp\\u003eEarly phases of gait and a point-based estimation\\u003c/p\\u003e\\u003cp\\u003eICC\\u0026thinsp;\\u0026lt;\\u0026thinsp;0.5\\u003c/p\\u003e\\u003cp\\u003eAverage phase detection\\u003c/p\\u003e\\u003cp\\u003eICC\\u0026thinsp;\\u0026gt;\\u0026thinsp;0.90\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"left\\\" colname=\\\"c8\\\"\\u003e\\u003cp\\u003eNA\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"left\\\" colname=\\\"c9\\\"\\u003e\\u003cp\\u003eNA\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"left\\\" colname=\\\"c10\\\"\\u003e\\u003cp\\u003eNA\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"left\\\" colname=\\\"c11\\\"\\u003e\\u003cp\\u003eNA\\u003c/p\\u003e\\u003c/td\\u003e\\u003c/tr\\u003e\\u003ctr\\u003e\\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e\\u003cp\\u003eLiang, 2022\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e\\u003cp\\u003eOlder people (n\\u0026thinsp;=\\u0026thinsp;15),\\u003c/p\\u003e\\u003cp\\u003eMean Age\\u0026thinsp;=\\u0026thinsp;56.60 years\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"left\\\" colname=\\\"c3\\\"\\u003e\\u003cp\\u003e-Single-camera based data collection on 6 meters walking path\\u003c/p\\u003e\\u003cp\\u003e-Customized algorithm from 2D to 3D kinematic analysis\\u003c/p\\u003e \\u003c/td\\u003e\\u003ctd align=\\\"left\\\" colname=\\\"c4\\\"\\u003e\\u003cp\\u003e-OpenPose\\u003c/p\\u003e \\u003cp\\u003e-PoseNet\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e\\u003cp\\u003eVICON marker based optical system with 6 cameras\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e\\u003cp\\u003eKinematic analysis\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"left\\\" colname=\\\"c7\\\"\\u003e\\u003cp\\u003eICC:0.50\\u0026ndash;0.73\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"left\\\" colname=\\\"c8\\\"\\u003e\\u003cp\\u003eNA\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"left\\\" colname=\\\"c9\\\"\\u003e\\u003cp\\u003eNA\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"left\\\" colname=\\\"c10\\\"\\u003e\\u003cp\\u003eNA\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"left\\\" colname=\\\"c11\\\"\\u003e\\u003cp\\u003eNA\\u003c/p\\u003e\\u003c/td\\u003e\\u003c/tr\\u003e\\u003ctr\\u003e\\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e\\u003cp\\u003eMehdizadeh, 2021\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e\\u003cp\\u003eOlder people (n\\u0026thinsp;=\\u0026thinsp;11),\\u003c/p\\u003e\\u003cp\\u003eMean Age\\u0026thinsp;=\\u0026thinsp;85.2 years\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"left\\\" colname=\\\"c3\\\"\\u003e\\u003cp\\u003e-Two cell phones camera-based data collection with two different placements of \\u0026ldquo;eye-level\\u0026rdquo; and \\u0026ldquo;straight\\u0026rdquo;\\u003c/p\\u003e \\u003cp\\u003e-13 meters walking path\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"left\\\" colname=\\\"c4\\\"\\u003e\\u003cp\\u003e-AlphaPose (AP)\\u003c/p\\u003e\\u003cp\\u003e-OpenPose (OP)\\u003c/p\\u003e\\u003cp\\u003e-Detectron (D)\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e\\u003cp\\u003eXsens MVN Awinda motion capture system with 7 IMU sensors\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e\\u003cp\\u003eSpatiotemporal Gait Parameters\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"left\\\" colname=\\\"c7\\\"\\u003e\\u003cp\\u003eNA\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"left\\\" colspan=\\\"3\\\" nameend=\\\"c10\\\" namest=\\\"c8\\\"\\u003e\\u003cp\\u003eLoA was evaluated to present agreement between pose estimation algorithms. Authors find highest agreement between AP and D to determine spatiotemporal gait variables.\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"left\\\" colname=\\\"c11\\\"\\u003e\\u003cp\\u003e\\u003cb\\u003eEye-level, front view conditions\\u003c/b\\u003e\\u003c/p\\u003e\\u003cp\\u003er, AP\\u0026amp; XSens\\u003c/p\\u003e\\u003cp\\u003eCadence: 0.99\\u003c/p\\u003e\\u003cp\\u003eStep time:0.71\\u003c/p\\u003e\\u003cp\\u003eStep width:0.54\\u003c/p\\u003e\\u003cp\\u003er, OP\\u0026amp; XSens\\u003c/p\\u003e\\u003cp\\u003eCadence: 0.99\\u003c/p\\u003e\\u003cp\\u003eStep time:0.71\\u003c/p\\u003e\\u003cp\\u003eStep width:0.42\\u003c/p\\u003e\\u003cp\\u003er, D\\u0026amp; XSens\\u003c/p\\u003e\\u003cp\\u003eCadence: 0.99\\u003c/p\\u003e\\u003cp\\u003eStep time:0.71\\u003c/p\\u003e\\u003cp\\u003eStep width:0.54\\u003c/p\\u003e\\u003c/td\\u003e\\u003c/tr\\u003e\\u003ctr\\u003e\\u003ctd align=\\\"left\\\" colname=\\\"c1\\\"\\u003e\\u003cp\\u003eRipic, 2023\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"left\\\" colname=\\\"c2\\\"\\u003e\\u003cp\\u003eYoung (n\\u0026thinsp;=\\u0026thinsp;23) \\u0026amp; Older Adults (n\\u0026thinsp;=\\u0026thinsp;14), and subjects with Parkinson Disease\\u003c/p\\u003e\\u003cp\\u003e(n\\u0026thinsp;=\\u0026thinsp;20)\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"left\\\" colname=\\\"c3\\\"\\u003e\\u003cp\\u003e-Kinatrax and Nexus motion capture system\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"left\\\" colname=\\\"c4\\\"\\u003e\\u003cp\\u003e-Kinatrax, HumanVersion3 with 8 video cameras on 10-meter walking path\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"left\\\" colname=\\\"c5\\\"\\u003e\\u003cp\\u003e-Nexus motion capture system with 8 video cameras on 10-meter walking path\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"left\\\" colname=\\\"c6\\\"\\u003e\\u003cp\\u003eSpatiotemporal Gait Parameters\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"left\\\" colname=\\\"c7\\\"\\u003e\\u003cp\\u003eNA\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"left\\\" colname=\\\"c8\\\"\\u003e\\u003cp\\u003e-0.039,0.039\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"left\\\" colname=\\\"c9\\\"\\u003e\\u003cp\\u003e-0.05,0.048\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"left\\\" colname=\\\"c10\\\"\\u003e\\u003cp\\u003e-0.049,0.046\\u003c/p\\u003e\\u003c/td\\u003e\\u003ctd align=\\\"left\\\" colname=\\\"c11\\\"\\u003e\\u003cp\\u003eICC for relative consistency and absolute\\u003c/p\\u003e\\u003cp\\u003eAgreement\\u0026thinsp;\\u0026gt;\\u0026thinsp;0.85\\u003c/p\\u003e\\u003c/td\\u003e\\u003c/tr\\u003e\\u003c/tbody\\u003e\\u003c/colgroup\\u003e\\u003c/table\\u003e\\u003c/div\\u003e\\u003c/p\\u003e\\u003cp\\u003e(\\u003cem\\u003eInsert\\u003c/em\\u003e Table\\u0026nbsp;\\u003cspan refid=\\\"Tab4\\\" class=\\\"InternalRef\\\"\\u003e4\\u003c/span\\u003e \\u003cem\\u003ehere\\u003c/em\\u003e.)\\u003c/p\\u003e\\u003cp\\u003eWhile the agreement analysis was moderate, the finding about consistency is parallel within the current literature. For instance, Ripic et al. found correlation coefficients greater than 0.85 between the Nexus 3D kinematic analysis system and the pose-estimation based Kinatrax system to determine spatio-temporal gait variables in older adults (\\u003cspan citationid=\\\"CR17\\\" class=\\\"CitationRef\\\"\\u003e17\\u003c/span\\u003e). Our findings also indicate significant correlations between the YOLOv11 and G-walk systems for cadence, gait speed, and gait cycle duration (r\\u0026thinsp;\\u0026ge;\\u0026thinsp;0.90); stride length and swing phase (r\\u0026thinsp;\\u0026ge;\\u0026thinsp;0.70); and stance phase, first double, and single support (r\\u0026thinsp;\\u0026ge;\\u0026thinsp;0.60). On the other hand, Mehdizadeh et al. examined the correlations separately for AlphaPose, OpenPose, and Detectron with the XSense motion capture system, and found r\\u0026thinsp;=\\u0026thinsp;0.99 for cadence, r\\u0026thinsp;=\\u0026thinsp;0.71 for step time for these three algorithms under the front view-eye level camera replacement conditions in older adults (\\u003cspan citationid=\\\"CR18\\\" class=\\\"CitationRef\\\"\\u003e18\\u003c/span\\u003e). These findings are consistent with our results (r\\u0026thinsp;=\\u0026thinsp;0.95 for cadence, r\\u0026thinsp;=\\u0026thinsp;0.95 for gait cycle duration). The authors also conducted their analyses using a single camera video, while they used two separate cameras to evaluate different gait conditions such as front and back walking, or walking at eye level or straight height (\\u003cspan citationid=\\\"CR18\\\" class=\\\"CitationRef\\\"\\u003e18\\u003c/span\\u003e). However, we completed a custom-built data extraction process using YOLOv11 keypoints and did not compare it with other algorithms such as AlphaPose or OpenPose. Future research could update this comparison by benchmarking against algorithms with high accuracy, precision, and recall metrics, including both previously used and newly developed ones, such as VitPose and customized models (\\u003cspan citationid=\\\"CR18\\\" class=\\\"CitationRef\\\"\\u003e18\\u003c/span\\u003e, \\u003cspan citationid=\\\"CR22\\\" class=\\\"CitationRef\\\"\\u003e22\\u003c/span\\u003e). Research on conducting psychometric analyses of the pose estimation method is still warranted, particularly regarding the gait analysis in the older population.\\u003c/p\\u003e\\u003cp\\u003eIn terms of test-retest reliability, our findings demonstrate that YOLOv11-based analysis showed excellent to good reliability in detecting cadence (ICC\\u0026thinsp;=\\u0026thinsp;0.92), gait speed (ICC\\u0026thinsp;=\\u0026thinsp;0.91), stride length (ICC\\u0026thinsp;=\\u0026thinsp;0.90), and gait cycle duration (ICC\\u0026thinsp;=\\u0026thinsp;0.89). In addition, there is moderate reliability in terms of swing phase (ICC\\u0026thinsp;=\\u0026thinsp;0.61), stance phase (ICC\\u0026thinsp;=\\u0026thinsp;0.61), first double support (ICC\\u0026thinsp;=\\u0026thinsp;0.59), and single support (ICC\\u0026thinsp;=\\u0026thinsp;0.59). The lower reliability levels could reflect a methodological limitation known to be challenging for G-walk itself, rather than a flaw inherent to the YOLOv11-based analysis. De Ridder et al. reported excellent reliability scores (ICC\\u0026thinsp;=\\u0026thinsp;0.85\\u0026ndash;0.91) for spatio-temporal gait variables, while the authors found cautious interpretation for agreement analyses of temporal variables with the GaitRite system (\\u003cspan citationid=\\\"CR13\\\" class=\\\"CitationRef\\\"\\u003e13\\u003c/span\\u003e). Similarly, Viteckova et al. concluded in their comparison of G-walk and GaitRite systems that these phase durations should be used with caution due to systematic error (\\u003cspan citationid=\\\"CR12\\\" class=\\\"CitationRef\\\"\\u003e12\\u003c/span\\u003e). G-walk system detects foot contact events using the sensor mounted at the L5 level of the spine via acceleration and angular velocity signals (\\u003cspan citationid=\\\"CR12\\\" class=\\\"CitationRef\\\"\\u003e12\\u003c/span\\u003e, \\u003cspan citationid=\\\"CR13\\\" class=\\\"CitationRef\\\"\\u003e13\\u003c/span\\u003e). Thus, it might be demonstrated that the YOLOv11 algorithm-based analysis also faces an inherent difficulty in detecting these phase transitions. In parallel, Carvalho et al. reported that the reliability of the Theia3D markerless motion capture system decreases (ICC\\u0026thinsp;\\u0026lt;\\u0026thinsp;0.50) at the early phase transitions of gait compared to the general kinematic flow of the gait cycles (ICC\\u0026thinsp;\\u0026gt;\\u0026thinsp;0.90) in older adults (\\u003cspan citationid=\\\"CR16\\\" class=\\\"CitationRef\\\"\\u003e16\\u003c/span\\u003e). Additionally, Liang et al. found reliability scores ranging from 0.50 to 0.73 for kinematic gait analysis in older adults using OpenPose and PoseNet algorithm-based systems (\\u003cspan citationid=\\\"CR15\\\" class=\\\"CitationRef\\\"\\u003e15\\u003c/span\\u003e). Although kinematic gait analysis requires greater precision than spatio-temporal gait analysis, the moderate reliability indicated in the aforementioned studies suggests that pose estimation systems still need improvement.\\u003c/p\\u003e\\u003cp\\u003eThere are several limitations that could affect the present findings and the potential impact of the proposed YOLOv11 algorithm-based spatio-temporal gait analysis. First one is that we used the G-walk system as a reference with its own limitations, such as soft tissue artefacts or sensor drift. However, this criterion was selected due to G-walk\\u0026rsquo;s low cost, widespread use in clinical and rehabilitation settings, and established validity. While this validates the success of the present system in variables such as gait speed, cadence, gait cycle duration, and stride length, an inertial sensor placed on L5, inevitably introduces random error in measurements of gait phases. This may have caused the Bland-Altman agreement limits between the systems to expand, potentially underestimating the actual clinical validity of the YOLOv11-based system. In interpreting our findings, it should be noted that G-walk is a practical criterion but not a perfect gold standard. Secondly, data collection using a single smartphone camera from left lateral view is another limitation of the present study. However, the limitation stems from a conscious methodological choice, as only spatio-temporal parameters are evaluated. The obtained validity and reliability scores cannot be directly generalised to older individuals with gait disorders due to our inclusion criteria. Finally, we opted for a 4-meter gait distance to evaluate the basic parameters of gait in a location-independent, minimal environment in older adults. However, it is important to consider that this abbreviated distance might have affected the comprehensive capture of a stable, steady-state gait cycle compared to the extended paths reported previously.\\u003c/p\\u003e\"},{\"header\":\"Conclusion\",\"content\":\"\\u003cp\\u003eQuantitative tools that analyze spatio-temporal gait characteristics can help identify specific gait deviations in older adults. They can also guide physiotherapy and rehabilitation programs, as well as track the effectiveness of these interventions. However, these measurements can be quite expensive and are often inaccessible to many researchers and clinicians. In this study, we used video-based gait analysis using only one smartphone and an open-source algorithm to show the clinical validity and reliability of spatio-temporal gait analysis based on the YOLOv11 algorithm in community-dwelling older adults. Our findings demonstrate that the proposed system captures gait speed, cadence, gait cycle duration, and stride length with excellent to strong consistency, moderate agreement, and excellent to good test-retest reliability, comparable to G-walk, providing its ability to consistently capture these metrics. This may provide significant evidence for the proposed system\\u0026rsquo;s use in rehabilitation monitoring and general follow-up processes, without the need for expensive and complex laboratory setups. However, the agreement analysis and reliability findings observed in the temporal variables of gait indicate that the system\\u0026rsquo;s interchangeability is not yet feasible in these metrics. The source of these limitations may not be only random error arising from a single camera setup, but also methodological difficulties encountered by the reference G-walk system itself. Overall, pose estimation technology is promising in capturing clinically relevant metrics regarding spatial gait parameters in older adults. Future studies should focus on improving and testing the performance of pose estimation-based systems in a clinical population to enhance the clinical validity of this system.\\u003c/p\\u003e\"},{\"header\":\"Abbreviations\",\"content\":\"\\u003cdiv class=\\\"DefinitionList\\\"\\u003e\\u003cdiv class=\\\"DefinitionListEntry\\\"\\u003e\\u003cdiv class=\\\"Term\\\"\\u003eWHO\\u003c/div\\u003e\\u003cdiv class=\\\"Description\\\"\\u003e\\u003cp\\u003eWorld Health Organization\\u003c/p\\u003e\\u003c/div\\u003e\\u003c/div\\u003e\\u003cdiv class=\\\"DefinitionListEntry\\\"\\u003e\\u003cdiv class=\\\"Term\\\"\\u003eIMU\\u003c/div\\u003e\\u003cdiv class=\\\"Description\\\"\\u003e\\u003cp\\u003eInertial measurement unit\\u003c/p\\u003e\\u003c/div\\u003e\\u003c/div\\u003e\\u003cdiv class=\\\"DefinitionListEntry\\\"\\u003e\\u003cdiv class=\\\"Term\\\"\\u003eCOCO\\u003c/div\\u003e\\u003cdiv class=\\\"Description\\\"\\u003e\\u003cp\\u003eCommon Objects in Context\\u003c/p\\u003e\\u003c/div\\u003e\\u003c/div\\u003e\\u003cdiv class=\\\"DefinitionListEntry\\\"\\u003e\\u003cdiv class=\\\"Term\\\"\\u003eYOLO\\u003c/div\\u003e\\u003cdiv class=\\\"Description\\\"\\u003e\\u003cp\\u003eYou Only Look Once\\u003c/p\\u003e\\u003c/div\\u003e\\u003c/div\\u003e\\u003cdiv class=\\\"DefinitionListEntry\\\"\\u003e\\u003cdiv class=\\\"Term\\\"\\u003eMMSE\\u003c/div\\u003e\\u003cdiv class=\\\"Description\\\"\\u003e\\u003cp\\u003eMini-Mental State Examination\\u003c/p\\u003e\\u003c/div\\u003e\\u003c/div\\u003e\\u003cdiv class=\\\"DefinitionListEntry\\\"\\u003e\\u003cdiv class=\\\"Term\\\"\\u003eTUG\\u003c/div\\u003e\\u003cdiv class=\\\"Description\\\"\\u003e\\u003cp\\u003eTimed Up and Go\\u003c/p\\u003e\\u003c/div\\u003e\\u003c/div\\u003e\\u003cdiv class=\\\"DefinitionListEntry\\\"\\u003e\\u003cdiv class=\\\"Term\\\"\\u003eLOA\\u003c/div\\u003e\\u003cdiv class=\\\"Description\\\"\\u003e\\u003cp\\u003eLimits of agreement\\u003c/p\\u003e\\u003c/div\\u003e\\u003c/div\\u003e\\u003cdiv class=\\\"DefinitionListEntry\\\"\\u003e\\u003cdiv class=\\\"Term\\\"\\u003eICCs\\u003c/div\\u003e\\u003cdiv class=\\\"Description\\\"\\u003e\\u003cp\\u003eIntra-class correlation coefficients\\u003c/p\\u003e\\u003c/div\\u003e\\u003c/div\\u003e\\u003c/div\\u003e\"},{\"header\":\"Declarations\",\"content\":\"\\u003ch2\\u003e\\u0026nbsp;\\u003c/h2\\u003e\\n\\u003ch2\\u003e\\u003cstrong\\u003eEthics approval and consent to participate\\u003c/strong\\u003e:\\u003c/h2\\u003e\\n\\u003cp\\u003eThe study was approved by the X University Ethical Committee for Non-Interventional Research (Registration Number: 2025/12\\u0026thinsp;\\u0026minus;\\u0026thinsp;08, Date: 09.04.2025) and executed according to principles of the Helsinki Declaration. All patients gave their written informed consent.\\u003c/p\\u003e\\n\\u003ch2\\u003eAuthor Contribution\\u003c/h2\\u003e\\n\\u003cp\\u003eConceptualization: GYC, UC, IKO, SO; conducting outcome assessments, GYC, IKO, PS; analyzing and interpreting data, UC, GYC; writing and original draft preparation, GYC, UC; writing, review, and editing, GYC, UC, IKO, SO, PS; approving the final version to be published, GYC, UC, IKO, PS, SO. Funding: No funding was received for this work.Acknowledgment: We thank Necla Kara for granting permission to use her image in this manuscript. Furthermore, this work was honored with the Best Poster Award at the PhysAgeNet \\u0026amp; EGRAPA Conference 2025, granted by the EU COST Action CA20104 - Network on Evidence-Based Physical Activity in Old Age (PhsAgeNet), supported by COST (European Cooperation in Science and Technology).Conflict of Interest: The authors declare that they have no conflict of interest related to the publication of this manuscript.\\u003c/p\\u003e\\n\\u003ch2\\u003eAcknowledgement\\u003c/h2\\u003e\\n\\u003cp\\u003eWe thank Necla Kara for granting permission to use her image in this manuscript. Furthermore, this work was honored with the Best Poster Award at the PhysAgeNet \\u0026amp; EGRAPA Conference 2025, granted by the EU COST Action CA20104 - Network on Evidence-Based Physical Activity in Old Age (PhsAgeNet), supported by COST (European Cooperation in Science and Technology).\\u003c/p\\u003e\"},{\"header\":\"References\",\"content\":\"\\u003col\\u003e\\u003cli\\u003e\\u003cspan\\u003eProgress report on the United Nations. Decade of Healthy Ageing, 2021\\u0026ndash;3.\\u003c/span\\u003e\\u003c/li\\u003e\\u003cli\\u003e\\u003cspan\\u003eGianfredi V, Nucci D, Pennisi F, Maggi S, Veronese N, Soysal P. Aging, longevity, and healthy aging: the public health approach. Aging Clin Exp Res. 2025;37(1):125. \\u003cspan class=\\\"ExternalRef\\\"\\u003e\\u003cspan class=\\\"RefSource\\\"\\u003e10.1007/s40520-025-03021-8\\u003c/span\\u003e\\u003cspan address=\\\"10.1007/s40520-025-03021-8\\\" targettype=\\\"DOI\\\" class=\\\"RefTarget\\\"\\u003e\\u003c/span\\u003e\\u003c/span\\u003e.\\u003c/span\\u003e\\u003c/li\\u003e\\u003cli\\u003e\\u003cspan\\u003eBautmans I, Knoop V, Amuthavalli Thiyagarajan J, Maier AB, Beard JR, Freiberger E, et al. WHO working definition of vitality capacity for healthy longevity monitoring. Lancet Healthy Longev. 2022;3(11):e789\\u0026ndash;96. \\u003cspan class=\\\"ExternalRef\\\"\\u003e\\u003cspan class=\\\"RefSource\\\"\\u003e10.1016/S2666-7568(22)00200-8\\u003c/span\\u003e\\u003cspan address=\\\"10.1016/S2666-7568(22)00200-8\\\" targettype=\\\"DOI\\\" class=\\\"RefTarget\\\"\\u003e\\u003c/span\\u003e\\u003c/span\\u003e.\\u003c/span\\u003e\\u003c/li\\u003e\\u003cli\\u003e\\u003cspan\\u003eSoltani A, Abolhassani N, Marques-Vidal P, Aminian K, Vollenweider P, Paraschiv-Ionescu A. Real-world gait speed estimation, frailty and handgrip strength: a cohort-based study. Sci Rep. 2021;11(1):18966. \\u003cspan class=\\\"ExternalRef\\\"\\u003e\\u003cspan class=\\\"RefSource\\\"\\u003e10.1038/s41598-021-98359-0\\u003c/span\\u003e\\u003cspan address=\\\"10.1038/s41598-021-98359-0\\\" targettype=\\\"DOI\\\" class=\\\"RefTarget\\\"\\u003e\\u003c/span\\u003e\\u003c/span\\u003e.\\u003c/span\\u003e\\u003c/li\\u003e\\u003cli\\u003e\\u003cspan\\u003eZhou H, Park C, Shahbazi M, York MK, Kunik ME, Naik AD, et al. Digital Biomarkers of Cognitive Frailty: The Value of Detailed Gait Assessment Beyond Gait Speed. Gerontology. 2022;68(2):224\\u0026ndash;33. \\u003cspan class=\\\"ExternalRef\\\"\\u003e\\u003cspan class=\\\"RefSource\\\"\\u003e10.1159/000515939\\u003c/span\\u003e\\u003cspan address=\\\"10.1159/000515939\\\" targettype=\\\"DOI\\\" class=\\\"RefTarget\\\"\\u003e\\u003c/span\\u003e\\u003c/span\\u003e.\\u003c/span\\u003e\\u003c/li\\u003e\\u003cli\\u003e\\u003cspan\\u003eLindemann U. Spatiotemporal gait analysis of older persons in clinical practice and research. Z Gerontol Geriatr. 2020;53(2):171\\u0026ndash;8. \\u003cspan class=\\\"ExternalRef\\\"\\u003e\\u003cspan class=\\\"RefSource\\\"\\u003e10.1007/s00391-019-01520-8\\u003c/span\\u003e\\u003cspan address=\\\"10.1007/s00391-019-01520-8\\\" targettype=\\\"DOI\\\" class=\\\"RefTarget\\\"\\u003e\\u003c/span\\u003e\\u003c/span\\u003e.\\u003c/span\\u003e\\u003c/li\\u003e\\u003cli\\u003e\\u003cspan\\u003eBoyer KA, Hayes KL, Umberger BR, Adamczyk PG, Bean JF, Brach JS, et al. Age-related changes in gait biomechanics and their impact on the metabolic cost of walking: Report from a National Institute on Aging workshop. Exp Gerontol. 2023;173:112102. \\u003cspan class=\\\"ExternalRef\\\"\\u003e\\u003cspan class=\\\"RefSource\\\"\\u003e10.1016/j.exger.2023.112102\\u003c/span\\u003e\\u003cspan address=\\\"10.1016/j.exger.2023.112102\\\" targettype=\\\"DOI\\\" class=\\\"RefTarget\\\"\\u003e\\u003c/span\\u003e\\u003c/span\\u003e.\\u003c/span\\u003e\\u003c/li\\u003e\\u003cli\\u003e\\u003cspan\\u003eSethi D, Bharti S, Prakash C. A comprehensive survey on gait analysis: History, parameters, approaches, pose estimation, and future work. Artif Intell Med. 2022;129:102314. \\u003cspan class=\\\"ExternalRef\\\"\\u003e\\u003cspan class=\\\"RefSource\\\"\\u003e10.1016/j.artmed.2022.102314\\u003c/span\\u003e\\u003cspan address=\\\"10.1016/j.artmed.2022.102314\\\" targettype=\\\"DOI\\\" class=\\\"RefTarget\\\"\\u003e\\u003c/span\\u003e\\u003c/span\\u003e.\\u003c/span\\u003e\\u003c/li\\u003e\\u003cli\\u003e\\u003cspan\\u003eBeauchet O, Allali G, Sekhon H, Verghese J, Guilain S, Steinmetz JP, et al. Guidelines for Assessment of Gait and Reference Values for Spatiotemporal Gait Parameters in Older Adults: The Biomathics and Canadian Gait Consortiums Initiative. Front Hum Neurosci. 2017;11. \\u003cspan class=\\\"ExternalRef\\\"\\u003e\\u003cspan class=\\\"RefSource\\\"\\u003e10.3389/fnhum.2017.00353\\u003c/span\\u003e\\u003cspan address=\\\"10.3389/fnhum.2017.00353\\\" targettype=\\\"DOI\\\" class=\\\"RefTarget\\\"\\u003e\\u003c/span\\u003e\\u003c/span\\u003e.\\u003c/span\\u003e\\u003c/li\\u003e\\u003cli\\u003e\\u003cspan\\u003ePrasanth H, Caban M, Keller U, Courtine G, Ijspeert A, Vallery H, et al. Wearable Sensor-Based Real-Time Gait Detection: A Systematic Review. Sensors. 2021;21(8):2727. \\u003cspan class=\\\"ExternalRef\\\"\\u003e\\u003cspan class=\\\"RefSource\\\"\\u003e10.3390/s21082727\\u003c/span\\u003e\\u003cspan address=\\\"10.3390/s21082727\\\" targettype=\\\"DOI\\\" class=\\\"RefTarget\\\"\\u003e\\u003c/span\\u003e\\u003c/span\\u003e.\\u003c/span\\u003e\\u003c/li\\u003e\\u003cli\\u003e\\u003cspan\\u003ePrisco G, Pirozzi MA, Santone A, Esposito F, Cesarelli M, Amato F, et al. Validity of Wearable Inertial Sensors for Gait Analysis: A Systematic Review. Diagnostics. 2024;15(1):36. \\u003cspan class=\\\"ExternalRef\\\"\\u003e\\u003cspan class=\\\"RefSource\\\"\\u003e10.3390/diagnostics15010036\\u003c/span\\u003e\\u003cspan address=\\\"10.3390/diagnostics15010036\\\" targettype=\\\"DOI\\\" class=\\\"RefTarget\\\"\\u003e\\u003c/span\\u003e\\u003c/span\\u003e.\\u003c/span\\u003e\\u003c/li\\u003e\\u003cli\\u003e\\u003cspan\\u003eV\\u0026iacute;tečkov\\u0026aacute; S, Hor\\u0026aacute;kov\\u0026aacute; H, Pol\\u0026aacute;kov\\u0026aacute; K, Krupička R, Růžička E, Brožov\\u0026aacute; H. Agreement between the GAITRite \\u0026reg; System and the Wearable Sensor BTS G-Walk \\u0026reg; for measurement of gait parameters in healthy adults and Parkinson\\u0026rsquo;s disease patients. PeerJ. 2020;8:e8835. \\u003cspan class=\\\"ExternalRef\\\"\\u003e\\u003cspan class=\\\"RefSource\\\"\\u003e10.7717/peerj.8835\\u003c/span\\u003e\\u003cspan address=\\\"10.7717/peerj.8835\\\" targettype=\\\"DOI\\\" class=\\\"RefTarget\\\"\\u003e\\u003c/span\\u003e\\u003c/span\\u003e.\\u003c/span\\u003e\\u003c/li\\u003e\\u003cli\\u003e\\u003cspan\\u003eDe Ridder R, Lebleu J, Willems T, De Blaiser C, Detrembleur C, Roosen P. Concurrent Validity of a Commercial Wireless Trunk Triaxial Accelerometer System for Gait Analysis. J Sport Rehabil. 2019;28(6). \\u003cspan class=\\\"ExternalRef\\\"\\u003e\\u003cspan class=\\\"RefSource\\\"\\u003e10.1123/jsr.2018-0295\\u003c/span\\u003e\\u003cspan address=\\\"10.1123/jsr.2018-0295\\\" targettype=\\\"DOI\\\" class=\\\"RefTarget\\\"\\u003e\\u003c/span\\u003e\\u003c/span\\u003e.\\u003c/span\\u003e\\u003c/li\\u003e\\u003cli\\u003e\\u003cspan\\u003eBautmans I, Jansen B, Van Keymolen B, Mets T. Reliability and clinical correlates of 3D-accelerometry based gait analysis outcomes according to age and fall-risk. Gait Posture. 2011;33(3):366\\u0026ndash;72. \\u003cspan class=\\\"ExternalRef\\\"\\u003e\\u003cspan class=\\\"RefSource\\\"\\u003e10.1016/j.gaitpost.2010.12.003\\u003c/span\\u003e\\u003cspan address=\\\"10.1016/j.gaitpost.2010.12.003\\\" targettype=\\\"DOI\\\" class=\\\"RefTarget\\\"\\u003e\\u003c/span\\u003e\\u003c/span\\u003e.\\u003c/span\\u003e\\u003c/li\\u003e\\u003cli\\u003e\\u003cspan\\u003eLiang S, Zhang Y, Diao Y, Li G, Zhao G. The reliability and validity of gait analysis system using 3D markerless pose estimation algorithms. Front Bioeng Biotechnol. 2022;10. \\u003cspan class=\\\"ExternalRef\\\"\\u003e\\u003cspan class=\\\"RefSource\\\"\\u003e10.3389/fbioe.2022.857975\\u003c/span\\u003e\\u003cspan address=\\\"10.3389/fbioe.2022.857975\\\" targettype=\\\"DOI\\\" class=\\\"RefTarget\\\"\\u003e\\u003c/span\\u003e\\u003c/span\\u003e.\\u003c/span\\u003e\\u003c/li\\u003e\\u003cli\\u003e\\u003cspan\\u003eCarvalho A, Vanrenterghem J, Cabral S, Assun\\u0026ccedil;\\u0026atilde;o A, Fernandes R, Veloso AP, et al. Markerless three-dimensional gait analysis in healthy older adults: test\\u0026ndash;retest reliability and measurement error. J Biomech. 2024;174:112280. \\u003cspan class=\\\"ExternalRef\\\"\\u003e\\u003cspan class=\\\"RefSource\\\"\\u003e10.1016/j.jbiomech.2024\\u003c/span\\u003e\\u003cspan address=\\\"10.1016/j.jbiomech.2024\\\" targettype=\\\"DOI\\\" class=\\\"RefTarget\\\"\\u003e\\u003c/span\\u003e\\u003c/span\\u003e.\\u003c/span\\u003e\\u003c/li\\u003e\\u003cli\\u003e\\u003cspan\\u003eRipic Z, Signorile JF, Best TM, Jacobs KA, Nienhuis M, Whitelaw C, et al. Validity of artificial intelligence-based markerless motion capture system for clinical gait analysis: Spatiotemporal results in healthy adults and adults with Parkinson\\u0026rsquo;s disease. J Biomech. 2023;155:111645. \\u003cspan class=\\\"ExternalRef\\\"\\u003e\\u003cspan class=\\\"RefSource\\\"\\u003e10.1016/j.jbiomech.2023.111645\\u003c/span\\u003e\\u003cspan address=\\\"10.1016/j.jbiomech.2023.111645\\\" targettype=\\\"DOI\\\" class=\\\"RefTarget\\\"\\u003e\\u003c/span\\u003e\\u003c/span\\u003e.\\u003c/span\\u003e\\u003c/li\\u003e\\u003cli\\u003e\\u003cspan\\u003eMehdizadeh S, Nabavi H, Sabo A, Arora T, Iaboni A, Taati B. Concurrent validity of human pose tracking in video for measuring gait parameters in older adults: a preliminary analysis with multiple trackers, viewing angles, and walking directions. J Neuroeng Rehabil. 2021;18(1):139. \\u003cspan class=\\\"ExternalRef\\\"\\u003e\\u003cspan class=\\\"RefSource\\\"\\u003e10.1186/s12984-021-00933-0\\u003c/span\\u003e\\u003cspan address=\\\"10.1186/s12984-021-00933-0\\\" targettype=\\\"DOI\\\" class=\\\"RefTarget\\\"\\u003e\\u003c/span\\u003e\\u003c/span\\u003e.\\u003c/span\\u003e\\u003c/li\\u003e\\u003cli\\u003e\\u003cspan\\u003eStenum J, Rossi C, Roemmich RT. Two-dimensional video-based analysis of human gait using pose estimation. PLoS Comput Biol. 2021;17(4):e1008935. \\u003cspan class=\\\"ExternalRef\\\"\\u003e\\u003cspan class=\\\"RefSource\\\"\\u003e10.1371/journal.pcbi.1008935\\u003c/span\\u003e\\u003cspan address=\\\"10.1371/journal.pcbi.1008935\\\" targettype=\\\"DOI\\\" class=\\\"RefTarget\\\"\\u003e\\u003c/span\\u003e\\u003c/span\\u003e.\\u003c/span\\u003e\\u003c/li\\u003e\\u003cli\\u003e\\u003cspan\\u003eLin TY, Maire M, Belongie S, Bourdev L, Girshick R, Hays J et al. Microsoft COCO: Common Objects in Context. 2015 Feb 21.\\u003c/span\\u003e\\u003c/li\\u003e\\u003cli\\u003e\\u003cspan\\u003eT\\u0026icirc;rziu E, Vasilevschi AM, Alexandru A, Tudora E. Enhanced Fall Detection Using YOLOv7-W6-Pose for Real-Time Elderly Monitoring. Future Internet. 2024;16(12):472. \\u003cspan class=\\\"ExternalRef\\\"\\u003e\\u003cspan class=\\\"RefSource\\\"\\u003e10.3390/fi16120472\\u003c/span\\u003e\\u003cspan address=\\\"10.3390/fi16120472\\\" targettype=\\\"DOI\\\" class=\\\"RefTarget\\\"\\u003e\\u003c/span\\u003e\\u003c/span\\u003e.\\u003c/span\\u003e\\u003c/li\\u003e\\u003cli\\u003e\\u003cspan\\u003eRana MS. Leveraging Markerless Computer Vision for Comprehensive Walking Automated Gait Analysis in Rehabilitation. 2024.\\u003c/span\\u003e\\u003c/li\\u003e\\u003cli\\u003e\\u003cspan\\u003eG\\u0026uuml;ngen C, Ertan T, Eker E, Yaşar R, Engin F. [Reliability and validity of the standardized Mini Mental State Examination in the diagnosis of mild dementia in Turkish population]. Turk Psikiyatri Derg. 2002;13(4). PMID: 12794644.\\u003c/span\\u003e\\u003c/li\\u003e\\u003cli\\u003e\\u003cspan\\u003ePodsiadlo D, Richardson S. The Timed Up \\u0026amp; Go: A Test of Basic Functional Mobility for Frail Elderly Persons. J Am Geriatr Soc. 1991;39(2):142\\u0026ndash;8. \\u003cspan class=\\\"ExternalRef\\\"\\u003e\\u003cspan class=\\\"RefSource\\\"\\u003e10.1111/j.1532-5415.1991.tb01616.x\\u003c/span\\u003e\\u003cspan address=\\\"10.1111/j.1532-5415.1991.tb01616.x\\\" targettype=\\\"DOI\\\" class=\\\"RefTarget\\\"\\u003e\\u003c/span\\u003e\\u003c/span\\u003e.\\u003c/span\\u003e\\u003c/li\\u003e\\u003cli\\u003e\\u003cspan\\u003eKhanam R, Hussain M. YOLOv11: An Overview of the Key Architectural Enhancements. 2024 Oct 23.\\u003c/span\\u003e\\u003c/li\\u003e\\u003cli\\u003e\\u003cspan\\u003ePortney LG. Foundations of clinical research: applications to evidence-based practice. FA Davis; 2020.\\u003c/span\\u003e\\u003c/li\\u003e\\u003cli\\u003e\\u003cspan\\u003eOlsen S, Rashid U, Barbado D, Suresh P, Alder G, Khan Niazi I, et al. The validity of smartphone-based spatiotemporal gait measurements during walking with and without head turns: Comparison with the GAITRite\\u0026reg; system. J Biomech. 2024;162:111899. \\u003cspan class=\\\"ExternalRef\\\"\\u003e\\u003cspan class=\\\"RefSource\\\"\\u003e10.1016/j.jbiomech.2023.111899\\u003c/span\\u003e\\u003cspan address=\\\"10.1016/j.jbiomech.2023.111899\\\" targettype=\\\"DOI\\\" class=\\\"RefTarget\\\"\\u003e\\u003c/span\\u003e\\u003c/span\\u003e.\\u003c/span\\u003e\\u003c/li\\u003e\\u003cli\\u003e\\u003cspan\\u003eGodfrey A, Del Din S, Barry G, Mathers JC, Rochester L. Instrumenting gait with an accelerometer: A system and algorithm examination. Med Eng Phys. 2015;37(4):400\\u0026ndash;7. \\u003cspan class=\\\"ExternalRef\\\"\\u003e\\u003cspan class=\\\"RefSource\\\"\\u003e10.1016/j.medengphy.2015.02.003\\u003c/span\\u003e\\u003cspan address=\\\"10.1016/j.medengphy.2015.02.003\\\" targettype=\\\"DOI\\\" class=\\\"RefTarget\\\"\\u003e\\u003c/span\\u003e\\u003c/span\\u003e.\\u003c/span\\u003e\\u003c/li\\u003e\\u003cli\\u003e\\u003cspan\\u003eKoo TK, Li MY. A Guideline of Selecting and Reporting Intraclass Correlation Coefficients for Reliability Research. J Chiropr Med. 2016;15(2):155\\u0026ndash;63. \\u003cspan class=\\\"ExternalRef\\\"\\u003e\\u003cspan class=\\\"RefSource\\\"\\u003e10.1016/j.jcm.2016.02.012\\u003c/span\\u003e\\u003cspan address=\\\"10.1016/j.jcm.2016.02.012\\\" targettype=\\\"DOI\\\" class=\\\"RefTarget\\\"\\u003e\\u003c/span\\u003e\\u003c/span\\u003e.\\u003c/span\\u003e\\u003c/li\\u003e\\u003cli\\u003e\\u003cspan\\u003eLatajka A, Stefańska M, Woźniewski M, Malicka I. Walking Speed and Risk of Falling Patients Operated for Selected Malignant Tumors. Healthcare. 2023;11(23):3069. \\u003cspan class=\\\"ExternalRef\\\"\\u003e\\u003cspan class=\\\"RefSource\\\"\\u003e10.3390/healthcare11233069\\u003c/span\\u003e\\u003cspan address=\\\"10.3390/healthcare11233069\\\" targettype=\\\"DOI\\\" class=\\\"RefTarget\\\"\\u003e\\u003c/span\\u003e\\u003c/span\\u003e.\\u003c/span\\u003e\\u003c/li\\u003e\\u003cli\\u003e\\u003cspan\\u003eKang CJ, Park SH, Son SM. Improving Gait Ability and Cognition Function through Action Observation Training in Elderly with Dementia. NeuroRehabilitation: An International. Interdisciplinary J. 2025;56(2):234\\u0026ndash;42. \\u003cspan class=\\\"ExternalRef\\\"\\u003e\\u003cspan class=\\\"RefSource\\\"\\u003e10.1177/10538135241296772\\u003c/span\\u003e\\u003cspan address=\\\"10.1177/10538135241296772\\\" targettype=\\\"DOI\\\" class=\\\"RefTarget\\\"\\u003e\\u003c/span\\u003e\\u003c/span\\u003e.\\u003c/span\\u003e\\u003c/li\\u003e\\u003cli\\u003e\\u003cspan\\u003eGianzina E, Yiannakopoulos CK, Kalinterakis G, Delis S, Chronopoulos E. Evaluation of the Timed Up and Go Test in Patients with Knee Osteoarthritis Using Inertial Sensors. Int J Translational Med. 2024;5(1):2. \\u003cspan class=\\\"ExternalRef\\\"\\u003e\\u003cspan class=\\\"RefSource\\\"\\u003e10.3390/ijtm5010002\\u003c/span\\u003e\\u003cspan address=\\\"10.3390/ijtm5010002\\\" targettype=\\\"DOI\\\" class=\\\"RefTarget\\\"\\u003e\\u003c/span\\u003e\\u003c/span\\u003e.\\u003c/span\\u003e\\u003c/li\\u003e\\u003cli\\u003e\\u003cspan\\u003eRipic Z, Signorile JF, Kuenze C, Eltoukhy M. Concurrent validity of artificial intelligence-based markerless motion capture for over-ground gait analysis: A study of spatiotemporal parameters. J Biomech. 2022;143:111278. \\u003cspan class=\\\"ExternalRef\\\"\\u003e\\u003cspan class=\\\"RefSource\\\"\\u003e10.1016/j.jbiomech.2022.111278\\u003c/span\\u003e\\u003cspan address=\\\"10.1016/j.jbiomech.2022.111278\\\" targettype=\\\"DOI\\\" class=\\\"RefTarget\\\"\\u003e\\u003c/span\\u003e\\u003c/span\\u003e.\\u003c/span\\u003e\\u003c/li\\u003e\\u003cli\\u003e\\u003cspan\\u003eCookson R, Doran T, Asaria M, Gupta I, Mujica FP. The inverse care law re-examined: a global perspective. Lancet. 2021;397(10276):828\\u0026ndash;38. \\u003cspan class=\\\"ExternalRef\\\"\\u003e\\u003cspan class=\\\"RefSource\\\"\\u003e10.1016/S0140-6736(21)00243-9\\u003c/span\\u003e\\u003cspan address=\\\"10.1016/S0140-6736(21)00243-9\\\" targettype=\\\"DOI\\\" class=\\\"RefTarget\\\"\\u003e\\u003c/span\\u003e\\u003c/span\\u003e.\\u003c/span\\u003e\\u003c/li\\u003e\\u003c/ol\\u003e\"}],\"fulltextSource\":\"\",\"fullText\":\"\",\"funders\":[],\"hasAdminPriorityOnWorkflow\":false,\"hasManuscriptDocX\":true,\"hasOptedInToPreprint\":true,\"hasPassedJournalQc\":\"\",\"hasAnyPriority\":false,\"hideJournal\":true,\"highlight\":\"\",\"institution\":\"\",\"isAcceptedByJournal\":false,\"isAuthorSuppliedPdf\":false,\"isDeskRejected\":\"\",\"isHiddenFromSearch\":false,\"isInQc\":false,\"isInWorkflow\":false,\"isPdf\":false,\"isPdfUpToDate\":true,\"isWithdrawnOrRetracted\":false,\"journal\":{\"display\":true,\"email\":\"info@researchsquare.com\",\"identity\":\"researchsquare\",\"isNatureJournal\":false,\"hasQc\":true,\"allowDirectSubmit\":true,\"externalIdentity\":\"\",\"sideBox\":\"\",\"snPcode\":\"\",\"submissionUrl\":\"/submission\",\"title\":\"Research Square\",\"twitterHandle\":\"researchsquare\",\"acdcEnabled\":true,\"dfaEnabled\":false,\"editorialSystem\":\"\",\"reportingPortfolio\":\"\",\"inReviewEnabled\":false,\"inReviewRevisionsEnabled\":true},\"keywords\":\"spatio-temporal gait parameters, human pose estimation, older adults, validity, reliability\",\"lastPublishedDoi\":\"10.21203/rs.3.rs-7987537/v1\",\"lastPublishedDoiUrl\":\"https://doi.org/10.21203/rs.3.rs-7987537/v1\",\"license\":{\"name\":\"CC BY 4.0\",\"url\":\"https://creativecommons.org/licenses/by/4.0/\"},\"manuscriptAbstract\":\"\\u003cp\\u003e\\u003cstrong\\u003eBackground: \\u003c/strong\\u003eThe spatio-temporal gait characteristics reflect physical function, fall risk, and rehabilitation processes in older adults. Recent advancements in artificial intelligence have led to the development of more practically accessible gait analysis systems.\\u003cstrong\\u003e \\u003c/strong\\u003eThe purpose of this study was to examine clinical validity and test-retest reliability of the YOLOv11 algorithm-based system for the assessment of spatio-temporal gait parameters in community-dwelling older adults.\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eMethods: \\u003c/strong\\u003eThirty-one community-dwelling older adults (mean Age: 66.94 years, mean BMI: 26.7) were recruited for the study. Participants were instrumented with a 4-meter gait test using the G-walk system, and video recording simultaneously. The collected videos were processed using the YOLOv11 algorithm to extract the spatio-temporal gait parameters. The consistency of the two methods was analysed using the Pearson correlation coefficient (\\u003cem\\u003er\\u003c/em\\u003e), and agreement was assessed using Bland-Altman Plots. The test-retest reliability was assessed using intra-class correlation coefficients (ICCs).\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eResults: \\u003c/strong\\u003ePrimary findings revealed that the YOLOv11 algorithm-based system exhibited strong to excellent consistency with the G-walk system for gait parameters such as cadence, gait speed, and stride length, gait cycle duration, and gait phases distribution (\\u003cem\\u003er\\u003c/em\\u003e=0.60-0.95; p\\u0026lt;0.001). Moderate agreement was observed between the two systems for all spatio-temporal variables, except for the first double support, which showed poor agreement. \\u0026nbsp;Additionally, reliability was excellent for cadence (ICC=0.92) and gait speed (ICC=0.91), good for gait cycle duration (ICC=0.89) and stride length (ICC=0.90), and moderate for gait phase distributions (ICC=0.59-0.61).\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eConclusions: \\u003c/strong\\u003eThe findings of this study suggest that the YOLOv11 algorithm-based system can be a potential alternative for gait analysis in older adults, providing a more accessible and affordable option for clinicians and researchers. Future research is warranted due to the cautious interpretation of stance, swing, and single/double support phases.\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eTrial registration: \\u003c/strong\\u003eThe study protocol was registered with ClinicalTrials.gov (identifier number: NCT07119944).\\u003c/p\\u003e\",\"manuscriptTitle\":\"Clinical validity and reliability of the pose estimation-based system to determine spatio-temporal gait parameters in older adults: A pilot study\",\"msid\":\"\",\"msnumber\":\"\",\"nonDraftVersions\":[{\"code\":1,\"date\":\"2025-11-19 12:05:29\",\"doi\":\"10.21203/rs.3.rs-7987537/v1\",\"editorialEvents\":[{\"type\":\"communityComments\",\"content\":0}],\"status\":\"published\",\"journal\":{\"display\":true,\"email\":\"info@researchsquare.com\",\"identity\":\"researchsquare\",\"isNatureJournal\":false,\"hasQc\":true,\"allowDirectSubmit\":true,\"externalIdentity\":\"\",\"sideBox\":\"\",\"snPcode\":\"\",\"submissionUrl\":\"/submission\",\"title\":\"Research Square\",\"twitterHandle\":\"researchsquare\",\"acdcEnabled\":true,\"dfaEnabled\":false,\"editorialSystem\":\"\",\"reportingPortfolio\":\"\",\"inReviewEnabled\":false,\"inReviewRevisionsEnabled\":true}}],\"origin\":\"\",\"ownerIdentity\":\"f8e42bf4-2768-47f6-8aac-204aadbe3f5e\",\"owner\":[],\"postedDate\":\"November 19th, 2025\",\"published\":true,\"recentEditorialEvents\":[],\"rejectedJournal\":[],\"revision\":\"\",\"amendment\":\"\",\"status\":\"posted\",\"subjectAreas\":[],\"tags\":[],\"updatedAt\":\"2026-02-13T16:40:13+00:00\",\"versionOfRecord\":[],\"versionCreatedAt\":\"2025-11-19 12:05:29\",\"video\":\"\",\"vorDoi\":\"\",\"vorDoiUrl\":\"\",\"workflowStages\":[]},\"version\":\"v1\",\"identity\":\"rs-7987537\",\"journalConfig\":\"researchsquare\"},\"__N_SSP\":true},\"page\":\"/article/[identity]/[[...version]]\",\"query\":{\"redirect\":\"/article/rs-7987537\",\"identity\":\"rs-7987537\",\"version\":[\"v1\"]},\"buildId\":\"8U1c8b4HqxoKbykW_rLl7\",\"isFallback\":false,\"isExperimentalCompile\":false,\"dynamicIds\":[84888],\"gssp\":true,\"scriptLoader\":[]}","source_license":"CC-BY-4.0","license_restricted":false}