DIGITAL HEALTH最新文献

{"title":"Facial recognition and analysis: A machine learning-based pathway to corporate mental health management.","authors":"Zicheng Zhang, Tianshu Zhang, Jie Yang","doi":"10.1177/20552076251335542","DOIUrl":"https://doi.org/10.1177/20552076251335542","url":null,"abstract":"<p><strong>Background: </strong>In modern workplaces, emotional well-being significantly impacts productivity, interpersonal relationships, and organizational stability. This study introduced an innovative facial-based emotion recognition system aimed at the real-time monitoring and management of employee emotional states.</p><p><strong>Methods: </strong>Utilizing the RetinaFace model for facial detection, the Dlib algorithm for feature extraction, and VGG16 for micro-expression classification, the system constructed a 10-dimensional emotion feature vector. Emotional anomalies were identified using the K-Nearest Neighbors algorithm and assessed with a 3σ-based risk evaluation method.</p><p><strong>Results: </strong>The system achieved high accuracy in emotion classification, as demonstrated by an empirical analysis, where VGG16 outperformed MobileNet and ResNet50 in key metrics such as accuracy, precision, and recall. Data augmentation techniques were employed to enhance the performance of the micro-expression classification model.</p><p><strong>Conclusion: </strong>These techniques improved the across diverse emotional expressions, resulting in more accurate and robust emotion recognition. When deployed in a corporate environment, the system successfully monitored employees' emotional trends, identified potential risks, and provided actionable insights into early intervention. This study contributes to advancing corporate mental health management and lays the foundation for scalable emotion-based support systems in organizational settings.</p>","PeriodicalId":51333,"journal":{"name":"DIGITAL HEALTH","volume":"11 ","pages":"20552076251335542"},"PeriodicalIF":2.9,"publicationDate":"2025-04-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12035250/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144040485","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Evaluating the effectiveness of a collaborative care digital mental health intervention on obsessive-compulsive symptoms in adolescents: A retrospective study.","authors":"Darian Lawrence-Sidebottom, Kelsey McAlister, Monika Roots, Jennifer Huberty","doi":"10.1177/20552076251331885","DOIUrl":"https://doi.org/10.1177/20552076251331885","url":null,"abstract":"<p><strong>Objective: </strong>Obsessive-compulsive (OC) symptoms, characterized by distressing and repetitive thoughts and behaviors, frequently onset during adolescence for individuals with obsessive-compulsive disorder or anxiety disorders. Digital mental health interventions (DMHIs) offer a promising platform to deliver mental health treatment, which may address OC symptoms. The purpose of this retrospective study was to determine the effects of a DMHI, Bend Health, on various domains of OC symptoms, including contamination, responsibility (for harm), unwanted thoughts, and symmetry, in adolescents.</p><p><strong>Methods: </strong>OC symptoms were assessed at baseline (before beginning care) and monthly in adolescents engaged in different care programs involving coaching and/or therapy with the DMHI. Retrospective analyses were used to identify characteristics associated with OC symptoms (N = 2151) and to characterize treatment responsiveness of adolescents with elevated OC symptoms (n = 553).</p><p><strong>Results: </strong>Adolescents with elevated OC symptoms (32.2%; n = 693 of 2151) were more likely than those with non-elevated OC symptoms to be female (p < .001), to have comorbid symptoms (e.g. anxiety and depression; p < .001), and participate in therapy (p < .001). Further, their caregivers had higher rates of sleep problems and burnout (p < .05). OC symptoms improved for 87.7% (n = 485 of 532) of adolescents during care with the DMHI, and 46.6% (n = 249 of 534) reported clinically substantive improvement. Scores decreased significantly over months in care (t₁₁₈₇ = -8.06, p < .001). Improvements were also identified for OC symptom dimensions (contamination, responsibility (for harm), unwanted thoughts, and symmetry).</p><p><strong>Conclusions: </strong>Our results deliver compelling preliminary evidence that participation in coaching and therapy with a DMHI may mitigate a variety of OC symptoms for adolescents. Improvements were observed across different OC symptom types, demonstrating the broad applicability of the DMHI to address various presentations and complexities of OC symptoms.</p>","PeriodicalId":51333,"journal":{"name":"DIGITAL HEALTH","volume":"11 ","pages":"20552076251331885"},"PeriodicalIF":2.9,"publicationDate":"2025-04-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12034962/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144032272","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Digital health interventions for spinal surgery patients: A systematic scoping review.","authors":"Annemieke Y van der Horst, Saskia M Kelders, Ernst T Bohlmeijer, Karlein M G Schreurs, Jan S Jukema","doi":"10.1177/20552076251328549","DOIUrl":"https://doi.org/10.1177/20552076251328549","url":null,"abstract":"<p><strong>Introduction: </strong>The potential of digital health interventions to optimize healthcare is promising also in the context of spinal surgery. However, a systematic review assessing the quality of digital health interventions for spinal surgery patients and the potential effects on these patients is lacking.</p><p><strong>Method: </strong>The objective of the current scoping review was to provide a systematic overview of digital health interventions for spinal surgery patients described in scientific literature. The focus was on describing the current digital health interventions, assessing the quality of these descriptions, reviewing the reported effects and assessing the methodological quality of the included studies.</p><p><strong>Results: </strong>A total of 14 full-text articles, describing 11 digital health interventions were included in the final analysis. These digital health interventions ranged from a website and app to a mobile phone messaging system and mobile phone interface. Most digital health interventions aim to improve adherence to rehabilitation guidelines and physical health. The included studies were generally of moderate to high quality and showed significant effects on physical health. Vital aspects of digital interventions such as \"working mechanism theory\" and \"prompts and reminders\" were often absent in the description of interventions.</p><p><strong>Conclusion: </strong>The study of digital interventions for spinal surgery patient is emerging and promising. However, there is a scarcity of studies using a rigorous design. A more systematic and comprehensive framework for developing and describing digital interventions for spinal surgery patients is highly recommended.</p>","PeriodicalId":51333,"journal":{"name":"DIGITAL HEALTH","volume":"11 ","pages":"20552076251328549"},"PeriodicalIF":2.9,"publicationDate":"2025-04-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12035155/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144065253","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Development of a digital treatment analyzer for the management of prostate cancer patients, with the help of real world data and use of predictive modelling.","authors":"Lev Korolkov, Heather A Robinson, Konstantinos Mouratis","doi":"10.1177/20552076251326021","DOIUrl":"https://doi.org/10.1177/20552076251326021","url":null,"abstract":"<p><p>Prostate cancer is the second most diagnosed cancer in the world. Treatment guidelines involve a multitude of therapies, however adherence to them is not fully established, while lack of personalized treatment strategies fails to put the patient as an individual clinical profile at the center of their treatment. We aim to present the concept of a digital treatment analyzer (TA) for the management of prostate cancer (PC) patients, leveraging real-world data (RWD) and predictive modeling to enhance personalized disease management strategies and adherence to PC guidelines, ultimately aiming to optimize therapeutic efficacy and improve outcomes. The TA comprises digital tools integrated into one user-intuitive interface, facilitating the development of patient-specific clinical profiles, classification of patients into matched historical RWD cohorts, presentation of relevant clinical guidelines, visual representation of treatment and outcomes, and mortality risk prediction based on a validated machine learning models. The Medical Information Mart for Intensive Care (MIMIC) IV dataset was utilized, including structured and unstructured data from the patient journey. The developed TA represents a promising approach to enhance personalized disease management strategies and adherence to PC guidelines. By integrating contemporary clinical guidelines, RWD and AI-driven insights, our digital TA aims to optimize therapeutic efficacy and improve patient outcomes. The presented concept demonstrates the potential for using a digital approach that integrates RWD into a treatment journey, to provide healthcare stakeholders with a holistic approach to PC management involving all available modern tools to achieve optimal outcomes.</p>","PeriodicalId":51333,"journal":{"name":"DIGITAL HEALTH","volume":"11 ","pages":"20552076251326021"},"PeriodicalIF":2.9,"publicationDate":"2025-04-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12034955/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144052777","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A serialization method for digitizing the image-based medical laboratory report.","authors":"Xiaoyang Ren, Dongwei Dou, Xianying He, Fangfang Cui, Jie Zhao","doi":"10.1177/20552076251334431","DOIUrl":"https://doi.org/10.1177/20552076251334431","url":null,"abstract":"<p><strong>Background: </strong>When applying for teleconsultations, medical laboratory reports are usually photographed with a mobile phone, and the photographic results are uploaded as teleconsultation application materials. It is very meaningful to extract the content of the image medical laboratory report and store the content digitally. There are already applications of OCR technology for medical text file recognition, but no researchers have recognized the format of the medical laboratory report and obtained the report content as a serialized process to digitize the image report. This article proposes a serialization method to digitize the medical laboratory report image.</p><p><strong>Materials and methods: </strong>This article first collects 330 image-based medical laboratory reports, annotates the format of the medical laboratory reports, and forms a training dataset for the layout analysis model. Then, using the pre-trained model, the dataset is trained to obtain a layout analysis model that can correctly recognize the format of the medical laboratory report. Then, the layout of the input image-based medical laboratory report is analyzed, and the layout analysis results are used to call the text detection and text recognition models to obtain the digital content of the image report. Finally, adjusting the layout of the digital content and storing the digital content as a docx file.</p><p><strong>Results: </strong>After training the layout analysis model, integrating layout analysis, text detection, and text recognition, we have obtained a serialization method that digitizes the content of the image medical laboratory report, restores the report format, shields sensitive and irrelevant content, and digitizes the report content of interest.</p><p><strong>Conclusions: </strong>By digitizing the image medical laboratory report through the serialization method, we can correctly display the content of the medical laboratory report for teleconsultation, while removing irrelevant content in the report, such as user names, examination equipment numbers, etc.</p>","PeriodicalId":51333,"journal":{"name":"DIGITAL HEALTH","volume":"11 ","pages":"20552076251334431"},"PeriodicalIF":2.9,"publicationDate":"2025-04-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12035204/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144031740","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Effect of patient satisfaction on the utilization of mHealth services by patients with chronic disease.","authors":"Jiao Lu, Heng Zhao, Zhilin Ji, Yanan Dong","doi":"10.1177/20552076251333983","DOIUrl":"https://doi.org/10.1177/20552076251333983","url":null,"abstract":"<p><strong>Background: </strong>Mobile health (mHealth) is considered an effective way to manage chronic disease patients' health. However, patients often do not use or continue using mHealth services due to concerns about service quality. Patient satisfaction may influence their utilization, and this effect may be moderated by platform regulations and differences in physicians' qualifications and regions.</p><p><strong>Objective: </strong>This study aims to enhance mHealth service utilization by examining how patient satisfaction affects both use and continuous use behaviors. It also explores the role of platform regulations (recommendation, commenting, and reward systems) and assesses differences due to physician proficiency and regional variations.</p><p><strong>Methods: </strong>Data were collected from Haodf, a leading mHealth platform in China, from October 2021 to March 2022 at 3-month intervals. Generalized additive models were used to verify nonlinear relationships between patient satisfaction and their utilization behaviors. The role of platform regulations and heterogeneity analysis regarding physician region was also examined.</p><p><strong>Results: </strong>Efficacy satisfaction positively influenced mHealth service utilization (<i>p</i> < 0.01), while efficacy and attitude satisfaction positively influenced continuous use (<i>p</i> < 0.01). Platform regulations altered these relationships (<i>p</i> < 0.01). Comprehensive recommendations had a greater effect than patient recommendations (<i>p</i> < 0.01). High-quality evaluations played a significant role (<i>p</i> < 0.01), while nonmaterial rewards had no significant effect (<i>p</i> ≥ 0.05). These relationships varied by physician region (<i>p</i> < 0.01).</p><p><strong>Conclusions: </strong>Efficacy satisfaction drives mHealth service utilization, while efficacy and attitude satisfaction drive continuous use. Recommendation and commenting systems exhibit a positive synergistic effect, while the reward system plays a supplementary role. Professional recommendations and high-quality evaluations enhance the evaluation system's reliability, promoting service utilization. Patients in big cities are more sensitive to physicians' proficiency and therapeutic effects. A multidimensional, transparent evaluation and feedback system should be integrated into the platform to improve service quality and increase patient utilization and loyalty.</p>","PeriodicalId":51333,"journal":{"name":"DIGITAL HEALTH","volume":"11 ","pages":"20552076251333983"},"PeriodicalIF":2.9,"publicationDate":"2025-04-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12035501/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144025955","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Enhancing stroke-associated pneumonia prediction in ischemic stroke: An interpretable Bayesian network approach.","authors":"Xingyu Liu, Jiali Mo, Zuting Liu, Yanqiu Ge, Tian Luo, Jie Kuang","doi":"10.1177/20552076251333568","DOIUrl":"https://doi.org/10.1177/20552076251333568","url":null,"abstract":"<p><strong>Background: </strong>Stroke-associated pneumonia (SAP) is a major cause of mortality following ischemic stroke (IS). However, existing predictive models for SAP often lack transparency and interpretability, limiting their clinical utility. This study aims to develop an interpretable Bayesian network (BN) model for predicting SAP in IS patients, focusing on enhancing both predictive accuracy and clinical interpretability.</p><p><strong>Methods: </strong>This retrospective study included patients diagnosed with IS and admitted to the Second Affiliated Hospital of Nanchang University between January and December 2019. Clinical data collected within 48 h of admission and SAP occurrences within 7 days were analyzed. Dimensionality reduction was performed using Least Absolute Shrinkage and Selection Operator regression, while data imbalances were addressed using synthetic minority oversampling technique. A BN model was trained using a hill-climbing algorithm and compared to logistic regression, decision trees, deep neural networks, and existing risk-scoring systems. Decision curve analysis was used to assess clinical usefulness.</p><p><strong>Results: </strong>Of the 1252 patients, 165 (13.18%) patients had SAP within 7 days of admission. The BN model identified age, risk of pressure injury (PI), National Institutes of Health Stroke Scale (NIHSS) score, and C-reactive protein (CRP) as significant prognostic factors. The BN model achieved an area under the curve of 0.85(95% CI: 0.78-0.92) on the test set, outperforming other models and demonstrating a greater net benefit in clinical decision-making.</p><p><strong>Conclusions: </strong>Age, risk of PI, NIHSS score, and CRP are significant predictors of SAP in IS patients. The interpretable BN model demonstrates superior predictive performance and interpretability, suggesting its potential as an effective and interpretable tool for clinical decision support in SAP risk assessment.</p>","PeriodicalId":51333,"journal":{"name":"DIGITAL HEALTH","volume":"11 ","pages":"20552076251333568"},"PeriodicalIF":2.9,"publicationDate":"2025-04-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12035493/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144042896","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"The mechanism of word-of-mouth learning on chronic disease patients' physician choice in online health communities: Latent Dirichlet allocation analyses and cross-sectional study.","authors":"Linlin Han, Narongsak Tek Thongpapanl, Ou Li","doi":"10.1177/20552076251332685","DOIUrl":"https://doi.org/10.1177/20552076251332685","url":null,"abstract":"<p><strong>Background: </strong>Word-of-mouth learning (WOML) plays a substantial role in patients' physician choice behavior. However, there is still a research gap in analyzing the mechanism of WOML on chronic disease patients' physician choice in online health communities (OHCs) considering individual differences.</p><p><strong>Objective: </strong>This study aims to develop a physician choice mechanism research model to reveal the influence of WOML on chronic disease patients' physician choice decision process from external interaction to internal cognition and emotion in OHCs based on social learning theory (SLT). The moderating effects of reasons for consultation and patients' demographic characteristics on the model's relationships were also explored.</p><p><strong>Methods: </strong>Guided by SLT, this study identified the external interaction factors and internal cognitive and emotional factors by analyzing 72,123 patients' online reviews based on a Latent Dirichlet Allocation model and developed the physician choice mechanism research model. The model was validated using structural equation modeling based on an online questionnaire survey of 526 valid Chinese patients with chronic disease. The moderating effect of reasons for medical consultation and demographic characteristics was examined using multi-group analysis.</p><p><strong>Results: </strong>Status capital (SC), decisional capital (DC), and price value (PV)) were the main external interaction factors to initiating chronic disease patients' internal cognition and emotion (perceived convenience (PC), perceived health benefits (PH), and patients' physician choice intention (CI)). PH and PC significantly mediated the relationship between SC, DC, PV, and CI. Reasons for medical consultation, district, and sex significantly moderated the relationships in the model.</p><p><strong>Conclusions: </strong>Considering individual differences, the results of this study advance a comprehensive understanding of how chronic disease patients interact with the environment through WOML to make physician choice decisions. OHCs can recommend suitable physician information to chronic disease patients considering individual differences to match patients' demands and improve service quality.</p>","PeriodicalId":51333,"journal":{"name":"DIGITAL HEALTH","volume":"11 ","pages":"20552076251332685"},"PeriodicalIF":2.9,"publicationDate":"2025-04-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12035117/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144057931","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Deep-learning-based detection of large vessel occlusion: A comparison of CT and diffusion-weighted imaging.","authors":"JaeYoung Kang, JunYoung Park, YoungJae Kim, BumJoon Kim, SangHee Ha, KwangGi Kim","doi":"10.1177/20552076251334040","DOIUrl":"https://doi.org/10.1177/20552076251334040","url":null,"abstract":"<p><strong>Background: </strong>Rapid and accurate identification of large vessel occlusion (LVO) is crucial for determining eligibility for endovascular treatment. We aimed to validate whether computed tomography combined with clinical information (CT&CI) or diffusion-weighted imaging (DWI) offers better predictive accuracy for anterior circulation LVO.</p><p><strong>Methods: </strong>Computed tomography combined with clinical information and DWI data from patients diagnosed with acute ischemic stroke were collected. Three deep-learning models, convolutional neural network, EfficientNet-B2, and DenseNet121, were used to compare CT&CI and DWI for detecting anterior circulation LVO.</p><p><strong>Results: </strong>A total of 456 patients, 228 patients with LVO [68.91 ± 12.84 years, 63.60% male; initial National Institutes of Health Stroke Scale (NIHSS) score: median 11 (7-14)] and without LVO [67.06 ± 12.29 years, 64.04% male; initial NIHSS score: median 2 (1-4)] were enrolled. Diffusion-weighted imaging achieved better results than CT&CI did in each performance metric. In DenseNet121, the area under the curves (AUCs) were found to be 0.833 and 0.756, respectively, while in EfficientNet-B2, the AUCs were 0.815 and 0.647, respectively.</p><p><strong>Conclusions: </strong>In detecting the presence of anterior circulation LVO, DWI showed better results in each performance metric than CT&CI did, and the best-performing deep-learning model was DenseNet121.</p>","PeriodicalId":51333,"journal":{"name":"DIGITAL HEALTH","volume":"11 ","pages":"20552076251334040"},"PeriodicalIF":2.9,"publicationDate":"2025-04-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12035260/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144028095","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Inertial sensor-based heel strike and energy expenditure prediction using a hybrid machine learning approach.","authors":"Kethohalli R Vidyarani, Viswanath Talasila, Raafay Umar, Venkatesan Prem, Ravi Prasad K Jagannath, Gurusiddappa R Prashanth","doi":"10.1177/20552076251333375","DOIUrl":"https://doi.org/10.1177/20552076251333375","url":null,"abstract":"<p><strong>Objective: </strong>Gait analysis plays a critical role in healthcare, biomechanics, and sports science, particularly for estimating energy expenditure (EE). This study introduces a hybrid machine learning approach integrating convolutional neural networks (CNNs), long-short-term memory (LSTM) networks, and transfer learning (TL) to estimate volume of oxygen (VO₂) and detect heel strikes (HS) using data from a single 9-axis inertial measurement unit (IMU).</p><p><strong>Methods: </strong>A clinical-grade VO₂ machine provided reference data for model training. The hybrid model was designed to combine spatial and temporal feature extraction capabilities from CNNs and LSTM networks while leveraging pre-trained weights through TL. The study compared the performance of the hybrid model with an LSTM-only approach to quantify improvements in VO₂ prediction.</p><p><strong>Results: </strong>The hybrid model significantly reduced the VO₂ prediction error from 20% to 3% compared to using LSTM-only approach. Additionally, the model demonstrated high accuracy for HS detection, achieving 93.53% accuracy as indicated by training and validation results. The lightweight IMU-based system proved effective for VO₂ estimation, offering a practical alternative to traditional VO₂ measurement systems, which are often complex, bulky, and uncomfortable for subjects.</p><p><strong>Conclusions: </strong>This study highlights the potential of a hybrid machine learning approach using IMU-based systems for accurate VO₂ estimation and HS detection. While the results are promising, the model's performance is constrained by 10 healthy subject datasets. Future work will require validation with more diverse datasets to enhance generalizability and robustness.</p>","PeriodicalId":51333,"journal":{"name":"DIGITAL HEALTH","volume":"11 ","pages":"20552076251333375"},"PeriodicalIF":2.9,"publicationDate":"2025-04-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12035216/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144032227","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}