Physiological measurement最新文献

{"title":"Updated consensus terminology and definitions in chest electrical impedance tomography.","authors":"I Frerichs, A Adler, M B P Amato, G Bellani, S H Böhm, F Braun, L Brochard, L Camporota, J Gamelin, H He, R Holzhacker, A H Jonkman, M Kallio, S Leonhardt, T Mauri, T Muders, T Riedel, C Rüegger, A Shono, S Spadaro, F Suarez-Sipmann, E Teschner, D G Tingay, A H van Kaam, H Wrigge, T Yoshida, Z Zhao","doi":"10.1088/1361-6579/ae8b55","DOIUrl":"10.1088/1361-6579/ae8b55","url":null,"abstract":"<p><p>Functional chest imaging using electrical impedance tomography (EIT) has experienced an impressive technological development since its invention in the early eighties of the last century. The number of experimental and clinical studies using this technology is continuously rising, and the increasing availability of devices approved for clinical use accelerates and diversifies its applications in patients. EIT is predominantly used in intensive care units but its utilisation in operating theatres, delivery rooms, pulmonary function laboratories and even remote outpatient settings is growing. Chest EIT is mainly applied to determine the regional distribution of pulmonary ventilation, aeration changes, and respiratory system mechanics both during mechanical ventilation and spontaneous breathing, but an increase in the use of chest EIT for imaging lung perfusion and cardiac action has recently been noted. The ongoing innovation of both EIT hardware and software, the new application fields, and the rising number of users of this technology require consensus on EIT terminology and definitions. This secures a common framework for conducting EIT studies, patient examinations and guarantees unified analysis of EIT data, documentation, reporting and comparability of findings. Our article provides a comprehensive consensus document on EIT terminology and definitions generated by EIT experts of the international TRanslational EIT development stuDy group in cooperation with the producers of EIT technology. It not only updates and extends the first consensus EIT terminology published in 2017, but also offers a new taxonomy of EIT measures, systematically based on the quantification of ventilation-related, heartbeat-related, and contrast-enhanced EIT signals. Thanks to its clear structure with tabulated recommended EIT terms, abbreviations, comprehensible explanations, notes, extensive literature sources and parameter calculations, EIT researchers, clinical users and manufacturers may use this document as a reference source of information relevant for chest EIT.</p>","PeriodicalId":20047,"journal":{"name":"Physiological measurement","volume":" ","pages":""},"PeriodicalIF":2.5,"publicationDate":"2026-08-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148456416","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"From PhysioNet to foundation models-a history and potential futures.","authors":"Gari D Clifford","doi":"10.1088/1361-6579/ae92e7","DOIUrl":"10.1088/1361-6579/ae92e7","url":null,"abstract":"<p><p>Over the last 35 years, the sharing of medical data and models for research has evolved from sneakernet to the internet-from mailing magnetic tapes and compact discs of a handful of well-curated recordings, to the high-speed download of relatively comprehensive hospital databases. More recently, the fervor around the potential for modern machine learning and 'AI' to catapult us into the next industrial revolution has led to a seemingly insatiable desire to pump almost any source of data into large models. Although this has great potential, it also presents a whole set of new challenges. In this article I examine these trends over the last 30 years, drawing on examples from the world of physiology, and in particular, cardiology, since it is one of the oldest data-intensive fields, and is undergoing a renaissance in the context of machine learning. From the early days of computerized cardiology, the Research Resource for Complex Physiologic Signals (<i>PhysioNet</i>) has been at the cutting edge of this field. This article, therefore, includes much of the Resource's history and the contributions of many key people involved, drawn from almost three decades of firsthand experience of co-developing elements of the Resource with its founders. I address what I perceive to be the most promising future directions for the PhysioNet Resource, and more generally, the growing issues and opportunities around dissemination and use of massive physiological databases, associated open access code, and public competitions, together with potential solutions to the key issues that face our field. Topics range from how we should approach foundation models in the context of the rapidly growing AI carbon footprint, to the potential of<i>Tiny-ML</i>and edge computing. I also cover issues around prizes and incentives, funding models, and scientific repeatability, as well as how we might address these issues by leveraging the PhysioNet Challenges, consistent with the philosophy of open-access from the early days of the PhysioNet Resource. Since this article is almost 38 pages long, I have added a single-page ten-point summary at the end of the article.</p>","PeriodicalId":20047,"journal":{"name":"Physiological measurement","volume":"47 8","pages":""},"PeriodicalIF":2.5,"publicationDate":"2026-08-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13488288/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148796687","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Analysis of the frequency-dependent sensitivity distribution for lower forearm impedance plethysmography.","authors":"Niko Strotmann, Christian Wiede, Roman Kusche, Karsten Seidl","doi":"10.1088/1361-6579/ae9035","DOIUrl":"10.1088/1361-6579/ae9035","url":null,"abstract":"<p><p><i>Objective</i>. The assessment of hemodynamic parameters plays an important role for the diagnosis and detection of cardiovascular diseases. One of the promising techniques for the non-invasive and continuous monitoring of arterial hemodynamics is the impedance plethysmography (IPG), which aims to measure the pulsatile volume change of superficial arteries. Crucial for the success of an IPG measurement is the selection of a setup that effectively targets the artery.<i>Approach</i>. This work analyzes the influence of the injection frequency alteration on the sensitivity distribution of an IPG measurement and the sensitivity towards the radial artery. The analysis is performed both in simulation and using study data obtained from 41 participants. The simulation consists of a finite element model representing the lower forearm and a fixed IPG setup, in which the sensitivity distribution is analyzed for different excitation frequencies between 3 kHz and 100 kHz. To allow for comparison, the IPG study utilizes the same five excitation frequencies between 3 kHz and 100 kHz.<i>Main results</i>. The results of the finite element simulation show that increasing excitation frequency shifts the region of large sensitivity away from the artery and especially towards deeper muscle tissue layers. The analysis of the study data supports this hypothesis by demonstrating that lower-frequency excitation yields larger relative pulsatile changes than larger excitation frequencies, thus indicating a larger sensitivity towards the pulsatile artery.<i>Significance</i>. The results of this paper show, that injection frequency directly affects the sensitivity regions. Analysis of the finite element method and study results indicate that the largest sensitivity towards the radial artery occurs at the smallest injection frequency. Thus, comparison of IPG measurements performed at different excitation frequencies should be interpreted carefully, comparability between individual measurements may be limited.</p>","PeriodicalId":20047,"journal":{"name":"Physiological measurement","volume":" ","pages":""},"PeriodicalIF":2.5,"publicationDate":"2026-08-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148584970","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Ventilation heterogeneity under mechanical ventilation via mask during anesthetic induction: a randomized trial comparing two ventilation strategies.","authors":"Li Yang, Lingling Gao, Congxia Pan, Lili Pan, Zhanqi Zhao, Jun Zhang","doi":"10.1088/1361-6579/ae8d7c","DOIUrl":"10.1088/1361-6579/ae8d7c","url":null,"abstract":"<p><p><i>Objective.</i>Low tidal volume, limited driving pressure and positive end expiratory pressure (PEEP) are well-accepted lung-protected ventilation strategies. The study aimed to compare driving pressure targeted strategy (PTS) and low tidal volume + PEEP strategy (VPS) in term of ventilation heterogeneity under mechanical ventilation via mask during anesthetic induction.<i>Approach</i>. Patients undergoing anesthesia induction were randomly assign to either PTS or VPS groups. An inspiration pressure of 15 cm H₂O and zero PEEP was used in PTS group. A tidal volume of 8 ml kg^-1predicted body weight and a PEEP of 5 cm H₂O was used in VPS group. The regional lung ventilation was monitored by electrical impedance tomography (EIT). Expiratory tidal volume (<i>TV</i>e), driving pressure, hemodynamics and PaO₂/FiO₂, and EIT-derived ventilation homogeneity (global inhomogeneity index,<i>GI</i>), the ratio of left/right lung ventilation distribution were calculated.<i>Main results.</i>Finally, 30 patients with PTS and 25 with VPS were analyzed. Compared with spontaneous ventilation, mechanical ventilation via mask caused significant changes in the<i>GI</i>(0.40 ± 0.04 vs 0.48 ± 0.08,<i>P</i>< 0.05). When compared with PTS, VPS was associated with improved<i>GI</i>(0.52 ± 0.09 vs 0.42 ± 0.03, PTS vs VPS,<i>P</i>< 0.001, power value 0.955), lower<i>TV</i>e (625.3 ± 192.7 vs 429.1 ± 48.8 ml,<i>P</i>< 0.001), lower driving pressure (15 ± 0 vs 10.5 ± 1.5 cm H₂O,<i>P</i>< 0.001), and slightly higher but not statistically different PaO₂/FiO₂(370.7 ± 90.1 vs 391.1 ± 48.9 mm Hg),<i>P</i>= 0.380).<i>Significance</i>. Mask mechanical ventilation with VPS may improve the lung ventilation inhomogeneity during anesthetic induction when compared with PTS.</p>","PeriodicalId":20047,"journal":{"name":"Physiological measurement","volume":" ","pages":""},"PeriodicalIF":2.5,"publicationDate":"2026-08-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148536597","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"How can we best predict energy expenditure in preschoolers? A comparison of machine learning models, METs computation and physical activity classification.","authors":"Hannah J Coyle-Asbil, Katarina Osojnicki, Christoph Buck, Jamie F Burr, David W L Ma, Marvin N Wright, Lori Ann Vallis","doi":"10.1088/1361-6579/ae9038","DOIUrl":"10.1088/1361-6579/ae9038","url":null,"abstract":"<p><p><i>Objective</i>. The first objective of this study was to refine previously designed machine learning models that predict energy expenditure (EE) of preschool children by modifying the method used to calculate metabolic equivalents (METs). The secondary objective was to compare estimates of time spent in different physical activity intensities across the newly developed models, previously published METs models, existing METs-based models from the literature, and models calibrated using direct observation.<i>Approach</i>. The model training dataset included 35 Canadian children (aged 3.0-5.99 years) equipped with GT9X accelerometers on their right hip. A portable metabolic unit was used to measure EE during a semi-structured protocol consisting of activities ranging from low- to high-intensity. The resulting models were applied to a sample of Canadian preschool children (<i>n</i>= 118; aged 3.0-5.99 years) to estimate time spent in sedentary (SED), light (LPA), moderate-to-vigorous (MVPA), and total physical activity (TPA). A repeated measures ANOVA was used to compare time estimates across models and according to three different configurations of METs activity thresholds.<i>Main results</i>. Results indicated that the newly developed models from Objective 1 produced significantly different estimates of time spent in SED, LPA, MVPA, and TPA compared to both previously published models and other existing METs-based models, highlighting the impact of different approaches to calculating METs.<i>Significance</i>. Model selection and METs calculation methods markedly influenced activity intensity estimates, underscoring the need for consistent methodology. Classification models yielded the most plausible free-living estimates.</p>","PeriodicalId":20047,"journal":{"name":"Physiological measurement","volume":" ","pages":""},"PeriodicalIF":2.5,"publicationDate":"2026-08-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148584695","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Efficient ECG representation learning via intervention-based attribution alignment.","authors":"Hongpo Zhang, Jiaang Li, Yuheng Li","doi":"10.1088/1361-6579/ae9450","DOIUrl":"10.1088/1361-6579/ae9450","url":null,"abstract":"<p><p><i>Objective.</i>The widespread adoption of wearable ECG devices has driven an explosive growth of long-term, multi-lead ECG data. However, clinical analysis and model development remain constrained by the excessive cost of data annotation, a limitation particularly pronounced in conventional deep learning methods that rely on labeled information.<i>Approach.</i>To address the problem of label scarcity in wearable ECG measurement scenarios, this paper proposes a self-supervised learning method for ECG based on intervention-based attribution alignment, aiming to improve the model's analytical performance on sparsely labeled ECG signals using unlabeled data. Our method integrates an intervention-inspired operation into the contrastive learning framework: it extracts compact representations from unlabeled data on one hand, and employs feature intervention to learn view-invariant mechanisms on the other, thereby reducing dependency on labeled data for target tasks.<i>Main results.</i>Experimental results on four public ECG datasets demonstrate that the proposed method achieves performance on par with state-of-the-art approaches in typical tasks such as arrhythmia classification, while utilizing only a fraction of the annotated samples.<i>Significance.</i>The findings indicate that our method effectively accommodates the multi-scenario monitoring characteristics of wearable devices, offering a practical solution to alleviate ECG data annotation burdens and advance the efficient analysis of long-term ECG signals.</p>","PeriodicalId":20047,"journal":{"name":"Physiological measurement","volume":" ","pages":""},"PeriodicalIF":2.5,"publicationDate":"2026-08-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148669942","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Comparison of activPAL and ActiGraph measured moderate to vigorous physical activity in people with COPD.","authors":"Jung Yoen Son, Katelyn Webster-Dekker, Deanna J Marriot, She'Lon Tucker, Janet L Larson","doi":"10.1088/1361-6579/ae87f1","DOIUrl":"10.1088/1361-6579/ae87f1","url":null,"abstract":"<p><p><i><b>Objective.</b></i>ActiGraph accelerometers are commonly used to assess moderate-to-vigorous physical activity (MVPA), but there are concerns about missing data due to compliance issues with hip-worn devices. ActivPAL devices are worn 24 h d^-1on the anterior thigh, minimizing missing data, but activPAL software does not directly estimate MVPA. This study aimed to evaluate the agreement between MVPA derived from activPAL (using five methods) and ActiGraph data (using two methods) among individuals with chronic obstructive pulmonary disease (COPD).<i><b>Approach</b>.</i>In this secondary analysis, 148 inactive adults aged50 years with COPD wore an activPAL3 on the thigh and an ActiGraph GT9X on the hip concurrently for 7 days. MVPA was estimated from activPAL using five methods: 1) metabolic equivalents, 2)75 steps/min, 3)100 steps/min, 4) vector magnitude (VM) counts, and 5) vertical-axis counts, and from ActiGraph using Freedson VM3 (2011) and Evenson VM (2023) cut-points, 60-s epochs, and normal filters. We assessed agreement between activPAL and ActiGraph MVPA using concordance correlation coefficients (CCCs), two one-sided t-tests (TOST), and Bland-Altman plots.<i><b>Main Results</b>.</i>Compared with ActiGraph MVPA using Freedson VM3, CCCs ranged from 0.51 to 0.53 across activPAL methods, except for100 steps/min (CCC = 0.42); four of five activPAL methods were equivalent at a ±8 min d^-1equivalence margin. Compared with ActiGraph MVPA using Evenson, CCCs ranged from 0.20 to 0.25 across activPAL methods, and no activPAL method was equivalent.<i><b>Significance</b>.</i>ActivPAL MVPA showed moderate group-level agreement and equivalence when compared to ActiGraph MVPA using Freedson VM3. Findings suggest activPAL may estimate group-level MVPA in this population, though further research is needed to validate this approach for individual-level MVPA.</p>","PeriodicalId":20047,"journal":{"name":"Physiological measurement","volume":" ","pages":""},"PeriodicalIF":2.5,"publicationDate":"2026-08-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148412626","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Effects of beta-adrenergic blockade on cardiorespiratory coupling markers in a clinically referred cardiopulmonary exercise test cohort with preserved left ventricular ejection fraction.","authors":"Beatrice Cairo, Raphael Martins de Abreu, Vlasta Bari, Ilaria Burzo, Beatrice De Maria, Francesco Bandera, Massimo Piepoli, Alberto Porta","doi":"10.1088/1361-6579/ae92cb","DOIUrl":"10.1088/1361-6579/ae92cb","url":null,"abstract":"<p><p><i>Objective</i>. Cardiorespiratory coupling (CRC) subsumes diverse mechanisms of reciprocal influence between cardiac and respiratory rhythms. Its most studied manifestation, the respiratory sinus arrhythmia (RSA), is thought to be under full vagal control. However, some studies on healthy subjects observed an enhancement of RSA following beta-adrenergic blockade. In the present study, we investigated the effects of habitual beta-blocker therapy on CRC using non-directional and directional measures.<i>Approach.</i>A retrospective analysis was performed on heart period (HP) variability and respiration (R) in a cohort of subjects with preserved left ventricular ejection fraction scheduled for cardiopulmonary exercise test subdivided into those under beta-blocker therapy (BB: 67 ± 9 yrs, 34 males, 21 females) and treatment-naïve participants (noBB: 63 ± 10 yrs, 22 males, 35 females). CRC was assessed through the RSA, estimated as the high frequency (HF) power of HP variability, a non-directional linear measure such as HP-R squared coherence (<i>K</i>²), and a directional model-free nonlinear cross-predictability (CP) approach opening the HP-R closed loop.<i>Main results.</i>Spectral analysis showed a significant increase of the HF power of HP variability in the BB group.<i>K</i>²did not show significant between-group differences, while the CP was higher in the BB cohort regardless of the time direction of the HP-R interactions.<i>Significance.</i>These findings support the hypothesis that CRC is not a pure vagal phenomenon but is also modulated by sympathetic activity. Furthermore, they highlight the importance of directional analysis to clarify the physiological pathways underlying CRC and their modulation by beta-adrenergic blockade.</p>","PeriodicalId":20047,"journal":{"name":"Physiological measurement","volume":" ","pages":""},"PeriodicalIF":2.5,"publicationDate":"2026-08-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148631289","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Automated analysis of wearable ECG: machine learning methods, preprocessing pipelines, and benchmark datasets.","authors":"Bartosz Puszkarski","doi":"10.1088/1361-6579/ae8b71","DOIUrl":"10.1088/1361-6579/ae8b71","url":null,"abstract":"<p><p>Wearable and mobile electrocardiography (ECG) has rapidly expanded access to rhythm monitoring outside of clinical settings, but the single- or few-lead nature of these recordings and their susceptibility to motion artefacts complicate clinical-grade interpretation. This review synthesises recent machine learning approaches for wearable ECG across three problem domains-signal-quality assessment (SQA), single-lead interpretation, and multi-lead interpretation-and covers preprocessing pipelines, quality-handling strategies, and 18 benchmark datasets. Across studies, no single architectural paradigm consistently outperforms others when evaluated fairly: classical feature-based methods remain competitive for rhythm tasks under resource constraints, while deep discriminative models show stronger cross-database generalisation when training data are sufficient. The most consistent predictor of reported performance is dataset characteristics-particularly scale, demographic composition, and whether signals were acquired from wearable devices in free-living conditions-rather than architectural choice. Three recurring evaluation deficiencies are identified: subject-level data leakage inflating classification metrics, SQA performance reported without acceptance rates, and reconstructed ECG leads validated solely by waveform similarity, without downstream diagnostic concordance. A structured comparative summary of all the methods reviewed is provided. The analysis clarifies that future progress depends primarily on evaluation rigour, representative datasets, and reproducible preprocessing rather than on architectural novelty alone.</p>","PeriodicalId":20047,"journal":{"name":"Physiological measurement","volume":" ","pages":""},"PeriodicalIF":2.5,"publicationDate":"2026-08-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148456483","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Predicting gait and timed up and go time using foot-tapping test and accelerometer.","authors":"Shamim Noroozi, Jacquelyn Puckey, Sabine Weyand, Jennifer M Jakobi","doi":"10.1088/1361-6579/ae8ca9","DOIUrl":"10.1088/1361-6579/ae8ca9","url":null,"abstract":"<p><p><i>Objective.</i>Gait and the timed up and go (TUG) time are the two tests that are widely used to assess functional mobility; however, their administration requires space and trained personnel, limiting feasibility for routine assessment. A brief, seated foot-tapping test may offer a practical alternative. This study examined whether accelerometer-derived features from unilateral foot-tapping can predict gait and TUG times across the adult lifespan.<i>Approach.</i>Seventy-eight adults aged 19-95 years (58 older adults: 80.25 ± 8.2 yrs; 20 young-to-midlife: 32.2 ± 14.1 yrs) performed 30-second unilateral foot-tapping trials with the dominant and nondominant foot. Time, frequency, and time-frequency domain features were extracted from triaxial accelerometer signals. Least absolute shrinkage and selection operator (LASSO) and Random Forest regression models were trained within a nested 5-fold cross-validation to predict gait and TUG.<i>Main Results.</i>Significant correlations (<i>p</i>< 0.05) were observed between foot-tapping features and both outcomes, with coefficients ranging from -0.5 to -0.35 for gait time and -0.6 to 0.55 for TUG time. Most of the corrected features were derived from time-frequency domain (∼65%) and time domain (∼35%), with the strongest associations observed for tapping rate, inter-tap variability, and variability in instantaneous tapping energy across trials. LASSO achieved lower average root mean square error (RMSE) than Random Forest for dominant-foot gait time (0.88 ± 0.23 s vs. 0.93 ± 0.21 s) and TUG time (2.76 ± 0.60 s vs. 2.88 ± 0.65 s), and for nondominant-foot gait time (0.89 ± 0.20 s vs 0.90 ± 0.20 s), and TUG time (2.93 ± 0.54 s vs. 2.96 ± 0.55 s). Prediction of both outcomes using the dominant foot with LASSO resulted in a lower RMSE than the nondominant foot.<i>Significance.</i>These findings suggest that accelerometer-based unilateral foot tapping may serve as a fast, space efficient, and low-burden complement to conventional mobility assessments. Further validation in larger clinical populations and comparison with additional regression methods are warranted.</p>","PeriodicalId":20047,"journal":{"name":"Physiological measurement","volume":" ","pages":""},"PeriodicalIF":2.5,"publicationDate":"2026-08-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148472018","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}