DIGITAL HEALTH最新文献

Objective: To evaluate the design, implementation, and early impact of the Tasty Spoon™ - a hybrid digital-analogue, electrostimulation device intended to restore taste perception in people with dementia - and to identify the organisational and market conditions required for its routine use.

Methods: A ProcessOriented Holistic (PrOH) Modelling Methodology was applied across four phases:1. Userneeds assessment through three focus groups (n = 28), semistructured interviews with individuals living with dementia (n = 10), caregivers (n = 5) and healthcare professionals (n = 15).2. Iterative codesign and lab prototyping, informed by thematic analysis and smallscale electrogustometry studies (n = 15; people with dementia = 10, control = 5).3. Feasibility testing the prototype in care-home dining routines to explore practicality, user acceptance, and caregiver workload, documented through field notes, post use interviews and caregiver workload diaries.4. Regulatory and commercial pathway mapping (UKCA/CE precompliance review, 3i stakeholder analysis). Quantitative data were analysed descriptively; qualitative insights were integrated into the PrOH workflow to expose implementation pinchpoints.

Results: PrOH analysis identified three design features that underpinned acceptability - familiar spoon form, automatic activation on contact, and dishwashersafe construction - while highlighting outstanding challenges in cost control, training, and individual differences in taste sensitivity. Participants consistently reported that the Tasty Spoon™ made food 'taste stronger' and restored variety to meals they had previously found bland. Our research also highlighted the importance of co-developing ethical procedures in collaboration with people with dementia.

Conclusion: Early, smallscale evidence suggests that a sensoryfocused assistive device can complement existing cognitive and mobility technologies in dementia care by enhancing mealtime enjoyment and easing caregiver burden. Larger, rigorously controlled studies are needed to quantify nutritional and clinical outcomes and to refine personalised stimulation settings before widescale deployment.

The disparity in the data from intensive care units, where stroke victims and heart attack patients make up a minority, makes this effort extremely difficult. A well-known difficulty in data mining is handling unbalanced data. The main contribution of this work is a method that accurately identifies and categorises minority-class data, even in highly imbalanced datasets with small class sizes. This work predicts stroke from the balanced and compressed data from MIMIC III dataset. The Convolutional Neural Network-Gated Recurrent Unit with Imbalanced Data Handling (CNN-GRU-IDH) is proposed. Additionally, it reduces the amount of data transferred by compressing healthcare data using the Lempel Ziv Markov Chain Algorithm (LZMA). Class imbalance problems are addressed with the Synthetic Minority Over-sampling Technique (SMOTE). Notably, this study adds a novel element by employing the Improved Multi-Objective Wolf Pack Algorithm (IMOWPA) to choose the appropriate K nearest neighbour value for SMOTE. The suggested model surpasses existing models when used on the dataset, obtaining a remarkable accuracy rate of 87.66% and 85.63% of F1 score for 70% of training and 30% of testing data. The CNN-GRU-IDH approach, which tries to forecast the incidence of strokes, is used as the major data classification technique. This study makes a substantial advancement to improving patient-specific early stroke prediction, which might save lives and lower death rates.

Aims: This study combines bibliometric and structured analyses to comprehensively examine the development, methodological characteristics, and application trends of multimodal artificial intelligence (AI) in Alzheimer's disease (AD) diagnosis.

Materials and methods: Literature from January 1, 2017 to December 31, 2024, was retrieved from the Web of Science Core Collection. Retrospective bibliometric and visual analyses were conducted using VOSviewer, CiteSpace, and the Bibliometrix R package.

Results: A total of 234 papers were identified, showing a continuous increase in publication volume, with the United States and China as dominant contributors. The analysis focused on data modalities, fusion architectures, and clinical applications. Data trends highlight the fusion of imaging data with genetics, biomarkers, and clinical data. Methodologically, five fusion approaches were categorized, with intermediate fusion being the most widely used strategy for its ability to balance heterogeneous data integration. In application, multimodal AI demonstrated clear advantages in early diagnosis, disease classification, and progression prediction.

Conclusion: Research on multimodal AI for AD has gained global attention and remains a key direction for diagnostic innovation. By synthesizing bibliometric insights with structured analyses of modalities and fusion strategies, this study offers a systematic understanding of current progress and provides valuable guidance for future methodological and translational research.

Background: Despite being a prevalent peripheral vestibular disorder in China, Meniere's disease (MD) suffers from low awareness, frequent misdiagnosis, and unsatisfactory treatment rates. As TikTok has become a prominent source of health information, no study has systematically evaluated the quality of its MD-related content. We therefore assessed the accuracy and reliability of MD videos on Chinese TikTok.

Methods: Top 100 videos for "Meniere's disease/syndrome" (TikTok, 1 May 2025) were analyzed. Quality was assessed using Video Information and Quality Index (VIQI), Global Quality Score (GQS), modified DISCERN (mDISCERN), and Patient Education Materials Assessment Tool for Audio-Visual Content (PEMAT-A/V). Descriptive statistics, correlation analyses, and predictive modeling were applied to 83 valid videos.

Results: Among 83 videos, 91.6% (n = 76) were physician-uploaded (primarily otolaryngologists/neurologists). Monologue, Q&A, and medical scenario formats showed superior quality. Symptoms dominated content (47%). Neurologists generated significantly higher normalized engagement per second than otolaryngologists (all adj. p < 0.05, r ＞ 0.35). Physicians outperformed news agencies in GQS scores (adj. p < 0.05, r = 0.291). Otolaryngologists scored higher than both neurologists and Traditional Chinese Medicine practitioners in PEMAT-A/V Understandability (all adj. p < 0.05, r > 0.37). Attending physicians exceeded chief physicians on all quality metrics (all adj. p < 0.05, r ＞ 0.35), an advantage potentially linked to their younger age, greater digital literacy, and more frequent social media use. Engagement metrics (likes, comments, favorites, shares) correlated strongly (r > 0.8). Predictive models for PEMAT-U/A were significant (p < 0.001), lacking multicollinearity/autocorrelation.

Conclusion: Physician-created MD content ensures credibility but requires quality improvement. PEMAT-U/A models guide enhancements, though broader application needs validation. Key health informatics priorities include certified creator engagement, algorithm optimization, and innovative content design.

Background: Early diagnosis of pulmonary nodules is crucial for improving the survival rate of lung cancer patients. However, significant variability in nodule size, shape, and anatomical location presents ongoing challenges for automated detection systems, often resulting in high false-positive rates.

Objective: This study aims to develop a dual-stage pulmonary nodule detection framework based on cross-layer attention fusion, with the goal of improving sensitivity while reducing false positives in chest CT scans.

Methods: We propose a two-stage detection pipeline. In the candidate detection stage, we design an Attention-guided Spatial and Channel Residual Module that integrates multi-scale residual connections with cross-dimensional attention to enhance discriminative features while preserving spatial detail. For false positive reduction, we introduce a Multi-scale Progressive Perception Network, which processes candidates across three anatomical resolutions through parallel branches and integrates top-down semantic fusion with localized attention. The model is evaluated on the LUNA16 dataset.

Results: Experimental results demonstrate that the proposed method achieves a sensitivity of 90.0% at 0.55 false positives per scan on the LUNA16 dataset. Compared to state-of-the-art approaches, our framework provides a favorable balance between sensitivity and precision.

Conclusions: The proposed dual-stage detection framework effectively enhances the performance of pulmonary nodule detection by incorporating cross-layer attention mechanisms and multi-scale feature integration. These findings suggest its potential for clinical deployment in computer-aided lung cancer screening.

Objective: Estimating the diverse symptoms of patients with advanced cancer is helpful for young physicians and medical teams in planning appropriate palliative care. We evaluated the use of medication, comorbidities, laboratory test results, and vital signs in hospitalized patients to predict death within 14 days.

Methods: We retrospectively selected hospitalized patients with advanced cancer who were admitted to the hospice ward. We are using extreme gradient boosting (XGBoost) and a combination of random forest (RF) and XGBoost (RF-XGBoost) models to analyze sixteen comorbidities, eighteen types of medications, twenty-six laboratory tests, and six vital signs. Finally, SHapley Additive exPlanations (SHAP) analysis was employed to interpret the contribution of each feature to survival prediction.

Results: Among the 2276 patients, 73% survived less than 14 days. The Area under the curve (AUC) of the XGBoost and RF-XGBoost models was 0.82 and 0.81 (P < 0.001), respectively. Among the top 10 most important feature values of both machine learning models after SHAP analysis, seven were related to medication use, whereas three were related to laboratory tests. The top three ranked feature values were stool softeners, antiemetics and sedatives. Patients who received these medications generally had a strong positive correlation with survival beyond 14 days.

Conclusions: Our results suggest that the types of medications used by patients, especially stool softeners, antiemetics, and sedatives, are valuable in predicting survival beyond 14 days for hospitalized patients with advanced cancer. This result may assist young physicians and medical teams in developing appropriate palliative care plans for patients and their families.

Background: Mental health challenges are common among older adults with mild cognitive impairment. Despite growing use of digital health interventions to improve cognitive function, their effects on mental health remain unexplored.

Objective: To assess the overall and subgroup-specific effectiveness of digital health interventions on mental health in older adults with mild cognitive impairment.

Methods: A systematic review and meta-analysis of randomized controlled trials was conducted following PRISMA guidelines, searching seven databases from inception to March 2024. Evidence quality was assessed using the GRADE framework and risk of bias with the Cochrane Collaboration's tool. Interrater agreement for screening and data extraction was assessed using the Kappa coefficient. Subgroup analyses assessed differences based on intervention characteristics such as type, setting, and duration, while meta-regression and sensitivity analysis identified other sources of heterogeneity and tested robustness.

Results: Eleven studies involving 610 participants met the criteria. Digital health interventions significantly reduced depressive symptoms (Standardized Mean Difference [SMD] -0.55, 95% CI -0.92 to -0.19) and anxiety symptoms (SMD -0.47, -0.76 to -0.18), but showed no significant effects on positive (SMD 0.74, -0.46 to 1.94) or negative affect (SMD -0.23, -0.60 to 0.14). Subgroup analyses indicated that hospital or nursing home settings with non-portable modality were optimal. Interventions over 6 weeks, with sessions exceeding 30 min up to 2 per week, were more effective for depressive symptoms. Among intervention types, only robot interventions reduced depressive symptoms. Fully digital interventions showed greater effectiveness than hybrid formats and yielded more favorable outcomes compared to controls. Overall, digital health interventions showed a significant benefit over usual care, while effects compared to waitlist controls were larger but not statistically significant.

Conclusions: This review indicates that digital health interventions hold promise for enhancing mental health in older adults with mild cognitive impairment. Future research should integrate digital therapeutic technologies to optimize interventions.

Objective: To address the critical data scarcity and privacy constraints that limit video-based motor behaviour assessment in clinical settings through a synthetic data generation framework, enabling robust human detection with high fidelity across challenging scenarios.

Methods: We employed synthetic data generation tailored to specific environments, implementing a novel synthetic retargeting approach based on procedural image synthesis. This method addresses the critical obstacles of limited training data in clinical settings due to privacy concerns, constrained views, occlusions, and uncontrolled environmental characteristics.

Results: Our synthetic retargeting approach yielded substantial and statistically significant performance improvements in human detection under real-world clinical data regimes. Evaluated across two clinical scenarios, the method improved existing models' performance (human detection score) by up to 19.4% in the more challenging scenario and up to 9.8% in the less challenging scenario (both with p < 0.001), demonstrating both high fidelity and robustness against challenging environments.

Conclusion: Synthetic retargeting provides an efficient and effective solution for adapting pre-trained human detection models to specific clinical deployment scenarios by generating scenario-tailored synthetic data, circumventing the privacy and logistical constraints that limit real data collection in healthcare settings. This approach enables robust video-based motor behaviour quantification with significant implications for both clinical management and research.

Objectives: This systematic review aims to evaluate studies that implemented and evaluated machine learning models in emergency department settings, focusing on their clinical and operational impact.

Methods: A comprehensive search was conducted across multiple databases from inception to January 2024. Studies were eligible if they assessed the implementation of machine learning models in emergency departments, with a particular focus on clinical and operational impact.

Results: A total of 84 studies met the inclusion criteria. Gradient boosting and neural networks were the most frequently used models. Mortality prediction models achieved AUC values ranging from 0.618 to 0.978, with key predictors including age, sex, race, vital signs, and comorbidities. Disposition prediction models showed AUC values of 0.675-0.96, often incorporating age, sex, vital signs, triage data, and past medical history. Length of stay prediction studies identified demographic data, triage level, chief complaints, and comorbidities as significant predictors, with gradient boosting models yielding the highest predictive accuracy. Machine learning-based treatment decision models showed promise in sepsis detection and cardiovascular triage. Wait time prediction models using gradient boosting decreased patient wait times by 18%-26%. Emergency department cost prediction studies were limited, with logistic regression models achieving AUCs of 0.71-0.76 for identifying high-cost patients.

Conclusion: Machine learning is widely used in emergency department research, but issues with generalizability and workflow integration limit its clinical use. Future work should improve data quality, representation, and ongoing model validation to enhance real-world utility.

Background: Digital health interventions are transforming hypertension management, yet the evolution and focus of this research domain remain underexplored. This study provides a comprehensive bibliometric analysis of global research trends and hotspots in digital health for hypertension management.

Methods: Relevant publications were retrieved from the Web of Science Core Collection. Bibliometric and visualization analyses were conducted using VOSviewer, CiteSpace, and R-Bibliometrix, covering the period from 1992 to 2025.

Results: A total of 1368 English-language articles, authored by 8918 researchers from 5268 institutions in over 100 countries/regions, were identified. These articles appeared in 435 journals, with publication output showing rapid growth since 2011 and peaking in 2022. The United States led in both productivity and international collaboration, with Duke University, the University of California System, and Harvard University as the top institutions. Bosworth HB emerged as the most prolific and influential author, while JMIR mHealth and uHealth and the Journal of Medical Internet Research were the leading journals. Keyword co-occurrence analysis revealed five major research clusters: (1) Digital interventions and patient management; (2) Population health and lifestyle factors; (3) Clinical practice, guidelines, and measurement; (4) Disease burden, outcomes, and epidemiology; and (5) Health equity, access, and technology utilization. Evidence suggests that digital health interventions improved patient self-management, medication adherence, and blood pressure control, highlighting their potential for better clinical outcomes. Recent burst keywords such as "burden," "telehealth," "meta-analysis," and "United States" indicate shifting research priorities toward implementation, health equity, and real-world impact.

Conclusion: This study identified rapid growth and diversification in digital health research for hypertension, with the United States, leading academic institutions, and journals such as JMIR mHealth and uHealth at the forefront. Five major research clusters were revealed, spanning digital interventions, clinical practices, lifestyle factors, disease burden, and health equity. Recent trends show increased focus on telehealth, implementation challenges, and equity of access. Future research should further integrate digital health solutions into routine hypertension care, address disparities, and systematically evaluate their real-world impact.