搜索结果：JMIR cardio

Telemedicine enables the provision of health services at a distance using information and communication technologies and includes different types of services: telemonitoring, remote control, virtual visit or televisit, telereferral, teleassistance, medical teleconsultation, health professionals' teleconsultation, and telerehabilitation. Continuous monitoring, early care, and greater therapeutic adherence could be benefits of telemedicine in the management of cardiovascular diseases. There are not many studies in the literature investigating the use of telemedicine in cardiology in Italy. The aim of this study is to illustrate the results of a survey on telemedicine services in cardiology conducted by the Department of Cardiovascular, Endocrine-Metabolic Diseases and Aging of the Italian National Institute of Health. The Telehealth Quality of Care Tool (TQoCT) from the World Health Organization (WHO) was used as the model. A survey was disseminated by the National Association of Doctors and Hospital Cardiologists (ANMCO) from June 2024 to October 2024 through a link provided to hospital and university cardiology operative units identified through the 8th Census of Cardiological Structures in Italy. The facilities were contacted by email or telephone. The survey was built using Microsoft Forms and composed of 52 questions divided into 6 sections. The analysis was carried out for the whole national territory and by geographical area. Of the 443 hospitals contacted, the response rate was 56.7% (251/443). Overall, 78.9% (198/251) of facilities reported telemedicine initiatives providing telemonitoring (128/198, 64.6%), telereferrals (104/198, 52.5%), medical teleconsultations (93/198, 47%), televisits (82/198, 41.4%), health professionals' teleconsultations (64/198, 32.3%), and telerehabilitation (10/198, 5.1%). The most frequently followed cardiovascular conditions were heart failure, ischemic heart disease, and cardiac arrhythmias, especially atrial fibrillation. Of the facilities, 51% (101/198) used deliberations, procedures, protocols, or informed consent for their activities, and 46% (91/198) of the reported services were paid. Lack of dedicated staff, complexity in organizational terms, and lack of technological equipment in the structure were the principal obstacles for health professionals; lack of familiarity with technology was the principal obstacle for patients. There are still organizational and clinical limitations to resolve to make telemedicine in cardiology an integral part of medical practice. The true challenge of telecardiology is likely the integration of available technology with precise, concrete, and simplified organizational models. As a tool, technology is fundamental only if it is accessible and adequate. However, it must be integrated with new paths built according to the needs of the territory, patients, and health personnel. Such a survey could provide help for the future design and use of telemedicine services in cardiology in Italy.

Large language models (LLMs) are increasingly used in health care, but their role in cardiology has not yet been systematically evaluated. This review aimed to assess the applications, performance, and limitations of LLMs across diverse cardiology tasks, including chronic and progressive conditions, acute events, education, and diagnostic testing. A systematic search was conducted in PubMed and Scopus for studies published up to April 14, 2024, using keywords related to LLMs and cardiology. Studies evaluating LLM outputs in cardiology-related tasks were included. Data were extracted across 5 predefined domains and the risk of bias was assessed using an adapted QUADAS-2 tool (developed by Whiting et al at the University of Bristol). The review protocol was registered in PROSPERO (CRD42024556397). A total of 33 studies contributed quantitative outcome data to a descriptive synthesis. Across chronic conditions, ChatGPT-3.5 (OpenAI) answered 91% (43/47) heart failure questions accurately, although readability often required college-level comprehension. In acute scenarios, Bing Chat omitted key myocardial infarction first aid steps in 25% (5/20) to 45% (9/20) of cases, while cardiac arrest information was rated highly (mean 4.3/5, SD 0.7) but written above recommended reading levels. In physician education tasks, ChatGPT-4 (OpenAI) demonstrated higher accuracy than ChatGPT-3.5, improving from 38% (33/88) to 66% (58/88). In patient education studies, ChatGPT-4 provided scientifically adequate explanations (5.0-6.0/7) comparable to hospital materials but at higher reading levels (11th vs 7th grade). In diagnostic testing, ChatGPT-4 interpreted 91% (36/40) electrocardiogram vignettes correctly, significantly better than emergency physicians (31/40, 77%; P< .001), but showed lower performance in echocardiography. LLMs show meaningful potential in cardiology, especially for education and electrocardiogram interpretation, but performance varies across clinical tasks. Limitations in emergency guidance and readability, as well as small in silico study designs, highlight the need for multimodal models and prospective validation.

Type D personality, characterized by high negative affectivity and social inhibition, has been linked to poorer mental health and heightened risk for adverse cardiovascular outcomes. Although previous studies have examined associations between type D personality, psychological distress, and cardiovascular disease (CVD), many have assessed these factors independently, relied on clinical samples, or overlooked the simultaneous assessment of psychological distress and CVD history. Consequently, less is known about how type D traits relate to emotional distress and CVD history within the general population. Understanding these relationships may support early identification of at-risk individuals and strengthen the integration of psychological screening into cardiovascular care. This study aimed to (1) examine associations between type D personality, emotional distress (depression, anxiety, and stress), and self-reported CVD history; (2) compare distress levels among participants with and without CVD history; and (3) determine whether type D personality predicts emotional distress independent of demographic factors and CVD history. A cross-sectional online survey was completed by 146 adults aged 30 to 85 years, recruited through convenience and snowball sampling on social media. Type D personality was assessed using the Type D Scale-14, and emotional distress was measured using the Depression Anxiety and Stress Scale-21 items. CVD history was captured through a single self-report question regarding prior diagnosis of a cardiovascular condition. Descriptive statistics characterized the sample. Two-tailed independent samples t tests compared distress between individuals with and without type D personality and between participants with and without CVD history. Pearson correlation coefficients examined associations among key variables. Hierarchical multiple regression assessed whether type D personality predicted emotional distress beyond age, gender, education, and CVD history. Of the 146 participants, 40 (27.4%) reported a history of CVD and 62 (42.5%) met criteria for type D personality. Individuals with type D personality exhibited significantly higher depression, anxiety, and stress levels than non-type D participants (all P<.001). Participants with CVD history also reported greater distress compared with those without CVD history. Hierarchical regression analyses showed that type D personality remained a strong independent predictor of emotional distress (&#x3b2;=.46; P<.001) after adjusting for demographics and CVD history. CVD history made an additional but smaller contribution to distress (&#x3b2;=.18; P=.008). These findings highlight the cumulative influence of personality traits and cardiovascular background on psychological well-being. Type D personality traits have been associated with higher levels of psychological distress and with a greater likelihood of self-reported CVD history in the general population. Type D personality remained a significant predictor of distress after accounting for demographic factors and cardiovascular history, underscoring its potential role in early psychological risk identification. Incorporating brief personality and mental health screening into cardiovascular assessment may support more comprehensive care.

Noncommunicable diseases are a global concern with high mortality. Among these, cardiovascular disease requires more attention due to recurrence with altered physical activity. "SRCardioCare" (Sri Ramachandra Cardio Care) is an integrated mobile software that was developed to engage patients with effective communication and e-media support. We intend to explore the development of mobile software and its perceived impacts among health care professionals. This study aimed to develop a more economical and feasible platform for cardiac rehabilitation following conservatively managed coronary arterial disease, heart failure, postoperative adult cardiac surgical revascularization, and other cardiac surgeries, and to develop software that facilitates effective communication among participants and health care professionals. The software application was developed based on the experts' interviews. The core components that are included in the software were assessed for their usefulness and applicability among people with cardiac disease using standardized questions. Physicians, nurse practitioners, and physiotherapists' opinions were obtained. The developed app features providing e-media content for patients and pre- or postphysical activity response, including vitals and feedback from patients at set regular intervals, which were updated to the software. Opinions obtained from practicing physicians (cardiologists), nurse practitioners, and physiotherapists were hopeful for the development and future implementation of "SRCardioCare" among patients. "SRCardioCare" is designed for the effective implementation of physical activity among patients after conservatively managed coronary arterial disease, heart failure, postoperative adult cardiac surgical revascularization, and other cardiac surgeries. An integrated communication medium and regular postphysical activity feedback of vitals may offer safety in implementing physical activity among the vulnerable in a remote setup. Remote rehabilitation is an essential and unexplored forum of practice in the field of rehabilitation, yet it requires wearable technology for remote monitoring and virtual reality and mixed reality for enhancing the adherence of the participants. To incorporate the telehealth forum effectively, especially in settings like India, the design must include an economically feasible and convenient model.

The HeartHealth program is a 6-month SMS text messaging-based support program offered to patients with a recent cardiovascular hospitalization or recent cardiovascular clinic visit in Western Sydney, Australia. Its customized content focuses on cardiovascular risk factors, lifestyle, treatments, and general heart health information. This study aimed to evaluate the implementation of the HeartHealth program. A mixed methods study was conducted assessing program reach, effectiveness, implementation, and maintenance using program data, participant feedback surveys, and staff focus group discussions. Consecutive adult patients who had attended cardiology clinics or had been discharged from cardiology hospitalization at Westmead Hospital, between April 2020 and April 2024, were included in the analysis. Content analysis was used to interpret the qualitative data. A total of 23,095 patients were invited, 8804 (38.1%) enrolled into the program, and 7964 out of 8804 (90.5%) completed the 6-month duration. Participants enrolled in the HeartHealth program had a mean age of 58.6 years, 60.3% (5302/8788) were male, and 62.4% (5382/8624) were recruited from an outpatient clinic setting. A total of 851,058 SMS text messages were sent, with 99.41% (846,009/851,058) delivered successfully. A total of 3533 out of 7964 (44.4% of program completers) participants completed the postintervention survey, and 4 HeartHealth staff members participated in a focus group discussion. Among the participants who completed the survey, 60.5% (2137/3533) reported that the program improved the healthiness of their diet, 53.6% (1894/3533) reported improved physical activity levels, and 56.1% (1982/3533) reported that it helped remind them to take their medications. Content analysis of participant feedback identified that the program was effective in prompting participants to change their diet, providing emotional support, reminding them of the importance of behavior change, improving their confidence in managing their health, and keeping participants focused. Key barriers included limited personalization, language options, and SMS text messaging scheduling flexibility. Recommended adaptations focused on enhancing personalization, greater engagement by local clinical teams, and expanding program dissemination. The program had a broad reach, translated to improved patient-reported health behaviors, and provided participants with needed support at low cost and low resource requirements. This analysis highlights the successful implementation and scalability of the HeartHealth program and provides key learnings for health systems that are looking to implement similar programs in the future.

Digital twin systems are emerging as promising tools in precision cardiology, enabling dynamic, patient-specific simulations to support diagnosis, risk assessment, and treatment planning. However, the current landscape of cardiovascular digital twin development, validation, and implementation remains fragmented, with substantial variability in modeling approaches, data use, and reporting practices. This systematic review aims to synthesize the current state of cardiovascular digital twin research by addressing 11 research questions spanning modeling technologies, data infrastructure, clinical applications, clinical impact, implementation barriers, and ethical considerations. We systematically searched 5 databases (PubMed, Scopus, Web of Science, IEEE Xplore, and Google Scholar) and screened 330 records. Forty-two original studies met the predefined eligibility criteria and were included. Data extraction was guided by 11 thematic research questions. Mechanistic and artificial intelligence (AI) or machine learning (ML) modeling strategies, data modalities, visualization formats, clinical use cases, reported impacts, limitations, and ethical or legal issues were coded and summarized. Risk of bias was evaluated using a custom checklist for modeling studies, the Prediction Model Risk of Bias Assessment Tool (PROBAST) for prediction models, and the Risk of Bias in Non-Randomized Studies - of Interventions (ROBINS-I) for observational studies. Most digital twins (29/42, 69%) relied on mechanistic models, while hybrid mechanistic-data-driven approaches and purely data-driven designs were less frequent (13/42, 31%). Only 18 studies explicitly described ML or AI, most often deep learning, Bayesian methods, or optimization algorithms. Personalization depended primarily on imaging (32/42, 76%) and electrocardiography or other electrical signals (18/42, 43%). Visualization was dominated (41/42, 98%) by static figures and anatomical snapshots. Clinically, digital twins were most commonly applied to therapy planning, risk prediction, and monitoring. Reported benefits focused on improved decision-making and therapy-related impacts, with occasional (8/42, 19%) reports of increased accuracy or faster diagnosis, but there was limited evidence for downstream improvements in patient outcomes. Key barriers included strong model assumptions and simplifications; high computational cost; data quality and availability constraints; limited external validation; and challenges in real-time performance, workflow integration, and usability. Explicit discussion of ethical, legal, or governance issues was rare (7/42, 17%). Cardiovascular digital twins show substantial potential to advance precision cardiology by linking personalized physiological models with clinical decision support, particularly for therapy planning and risk prediction in arrhythmia and heart failure. However, real-world implementation is constrained by methodological heterogeneity, restricted data and validation practices, limited openness of code and models, and sparse engagement with ethical and governance questions. Future research should prioritize standardized evaluation frameworks, robust clinical validation, interoperable and user-centered system design, and ethically grounded, patient-centered development to realize the full clinical value of digital twin systems.

Heart failure (HF) is a prevalent chronic condition for which optimal management depends not only on guideline-directed medical therapy but also on patients' understanding of their disease, recognition of warning signs, and sustained medication adherence, which remains challenging in routine care. Mobile health interventions may support therapeutic education and self-management; however, many available apps lack validated content and local relevance. Cardio-Meds is a mobile app developed at Geneva University Hospitals to support HF self-management through structured educational content, interactive quizzes, medication lists with reminders, and tools for monitoring weight and vital signs. This study aims to evaluate the impact of a 30-day Cardio-Meds intervention on HF knowledge and medication adherence in patients with HF with reduced or mildly reduced ejection fraction. We conducted a single-center, pilot randomized controlled trial in patients followed at the outpatient HF clinic or enrolled in cardiac rehabilitation at Geneva University Hospitals in 2024. Eligible participants had HF with a left ventricular ejection fraction less than 50%, were receiving HF-specific pharmacotherapy, speak French, and owned a smartphone. Participants were recruited by phone and randomized to Cardio-Meds use for 30 days, a self-guided intervention with a single standardized technical support call. Outcomes were self-assessed using standardized questionnaires: HF knowledge and self-management using the Dutch Heart Failure Knowledge Scale (DHFKS; score range 0-15); medication adherence using the Basel Assessment of Adherence to Immunosuppressive Medication Scale, covering initiation, implementation, and persistence; and usability in the intervention group using the System Usability Scale (score range 0-100). Between-group differences in DHFKS scores were analyzed using analysis of covariance adjusted for baseline values. A total of 49 participants were included (25 intervention, 24 control); 78% (n=38) were male, and the mean age was 62 (SD 11.4) years. In the intervention group, median app usage was 123 (IQR 74-273) minutes, with a median of 43 (IQR 19-85) logins. Mean baseline DHFKS scores were similar between groups (intervention 11.1, SD 2.4 vs control 10.5, SD 2.9). At 30 days, mean scores increased significantly in the intervention group (12.4, SD 2.4; mean change +1.3; P<.001) and remained stable in the control group (10.4, SD 3; mean change -0.1; P=.82), with a significant adjusted between-group difference of +1.3 points (P<.001). No significant between-group differences were observed for medication adherence. Usability was high, with a mean score of 84.3 (SD 15), and 64% (16/25) of intervention participants reported that they would continue using the app. In a stable ambulatory HF population, the Cardio-Meds intervention demonstrated short-term improvement in HF knowledge, while no effect was observed on medication adherence within the 30-day follow-up period. The app showed high usability and acceptability. Larger multicenter studies with longer follow-up are needed to assess clinical impact.

Accurate identification of clinical symptoms and signs (S&S) is essential for the early detection of high-burden cardiorespiratory conditions, including lung cancer, chronic obstructive pulmonary disease, and heart failure. Although symptom data play a central role in diagnostic reasoning and predictive modeling, most S&S information remains embedded in unstructured electronic health record notes, limiting their use in automated phenotyping, surveillance, and clinical decision support. Traditional natural language processing systems struggle with domain variability and contextual nuance in clinical text. Recent advances in large language models (LLMs) offer a promising alternative, yet challenges remain in hallucinations, overinference, and safe deployment. This study evaluated whether locally deployed open-source models could reliably extract cardiorespiratory S&S and map them to ICD-10-CM (International Classification of Diseases, Tenth Revision, Clinical Modification) codes using optimized prompting strategies. This study aims to assess the accuracy of open-source LLMs in extracting explicitly stated cardiorespiratory S&S from clinical notes and mapping them to ICD-10-CM codes (R00-R09) and to compare performance across 4 prompt-engineering strategies, including a multimodule LLM framework. A total of 593 clinical notes from the MTSamples database were manually reviewed, with 93 notes used for prompt development and comparison using Llama 3.3-70B, and 500 notes used as testing data for the final best prompt setting using both Llama 3.3-70B and gpt-oss-120B. Four prompting conditions were evaluated: (1) instruction-only, (2) ICD-10-CM definition-based prompts, (3) assumption-free prompts, and (4) a multimodule LLM framework with postprocessing. Performance was measured using precision, recall, and F1-score for both S&S extraction and ICD-10-CM code generation. Across all prompt strategies, model performance improved as more structure and constraints were added. Instruction-only prompting demonstrated high recall but poor precision. Incorporating ICD-10-CM definitions improved coding accuracy, and assumption-free prompting further balanced precision and recall. The multimodule approach with postprocessing achieved the highest performance during prompt development. On the independent test corpus, entity-level microaveraged evaluation showed that gpt-oss-120B outperformed Llama 3.3-70B in both tasks. For S&S extraction, Llama 3.3-70B achieved a precision of 0.63, a recall of 0.86, and an F1-score of 0.73, whereas gpt-oss-120B achieved a precision of 0.89, a recall of 0.87, and an F1-score of 0.88. For ICD-10-CM code mapping, Llama 3.3-70B achieved a precision of 0.59, a recall of 0.83, and an F1-score of 0.69, whereas gpt-oss-120B achieved a precision of 0.90, a recall of 0.84, and an F1-score of 0.87. Locally deployed LLMs, when paired with optimized prompting and multimodule orchestration, can accurately extract cardiorespiratory S&S and generate ICD-10-CM codes from unstructured clinical notes. This approach increases the level of data safety by enabling on-premises processing without external data transmission and demonstrates strong potential for scalable, domain-adaptive symptom extraction pipelines in biomedical informatics. Future work should expand datasets and evaluate generalizability across clinical domains.

Acute kidney injury critically impacts outcomes in cardiogenic shock secondary to acute myocardial infarction (CS-AMI). Acute kidney injury is one of the strongest independent predictors of in-hospital mortality in CS-AMI. Despite evidence that early renal replacement therapy (RRT) initiation improves survival, comprehensive prediction models for RRT in this population remain lacking. This study aimed to develop and internally validate a Least Absolute Shrinkage and Selection Operator (LASSO) regression-based prediction model and clinical nomogram for in-hospital RRT in patients with CS-AMI. This multicenter retrospective cohort study included 1431 patients with CS-AMI from the Gulf Cardiogenic Shock (Gulf-CS) registry across 13 centers in 6 Gulf countries (2020-2022). LASSO logistic regression was applied to a training set (1071/1431, 80%) to select baseline predictors of RRT; performance was evaluated on a held-out testing set (268/1431, 20%). Internal validation included 10-fold cross-validation and bootstrapping (1000 iterations). Cluster-robust SEs accounted for center effects. The model was compared to a parsimonious model (age+creatinine clearance), and a clinical nomogram was developed. Of 1431 patients, 190 (13.3%) required RRT. Patients requiring RRT were significantly older (mean 64.17, SD 12.14 y vs mean 59.75, SD 11.77 y; P<.001), with higher prevalences of diabetes mellitus (72.1% vs 61.9%; P=.008), peripheral arterial disease (11.6% vs 3.7%; P<.001), and prior cerebrovascular accident (11.1% vs 5.7%; P=.005). The RRT group had lower creatinine clearance (46 vs 72 mL/min; P<.001), higher baseline lactate (2.7 vs 2.1 mmol/L; P<.001), and more advanced Society for Cardiovascular Angiography and Interventions (SCAI) shock stages (stages D and E: 90.5% vs 64.9%; P<.001). LASSO selected 15 baseline predictors. The model achieved an area under the receiver operating characteristic curve (AUC) of 0.714 on the testing set, significantly outperforming the parsimonious model (AUC: 0.624; P<.001). Bootstrap-corrected AUC was 0.745 (95% CI 0.730-0.756). In-hospital mortality was markedly higher in the RRT group (75.8% vs 38.8%; P<.001), with longer hospital stay (10 vs 6 d; P<.001), more major bleeding (16.8% vs 7.3%; P<.001), and cerebrovascular accidents (11.1% vs 4.9%; P=.001). We have developed and internally validated a robust 15-variable nomogram (Gulf-CS-Nomogram) that accurately predicts the need for RRT in patients with CS-AMI using baseline data intended for use after coronary angiography. This tool may facilitate early nephrology consultation and timely RRT initiation to improve outcomes.

Both poor sleep health and hypertensive disorders of pregnancy (HDP) are independent risk factors for cardiovascular disease. Whether poor postpartum sleep contributes to the relationship between HDP and future cardiovascular disease is unknown. This pilot study evaluated the feasibility and acceptability of studying sleep health using a wearable device (Oura ring) among mothers of young children. We evaluated indices of sleep health both objectively with the Oura ring and subjectively via questionnaires and qualitative interviews among mothers with and without a history of HDP. We also aimed to compare cardiovascular health (CVH) among mothers with vs without a prior history of HDP. Women who were 3 to 7 years after childbirth completed baseline questionnaires (the Pittsburgh Sleep Quality Index [PSQI], Mediterranean Eating Pattern for Americans tool, and 7-item International Physical Activity Questionnaire) and wore the Oura ring continuously for 2 weeks to monitor sleep. Optimal sleep health was defined as a sleep duration of &#x2265;7 hours, a PSQI score of &#x2264;5, sleep timing with a sleep midpoint between 2 AM and 4 AM, a sleep efficiency of >85%, and a sleep onset variability of <60 minutes. CVH was assessed using the Life's Essential 8 score, with 8 factors assessed via questionnaires (diet, physical activity, and nicotine exposure) and objective measurements (BMI, blood pressure, blood lipids, blood glucose, and sleep duration). Semistructured interviews were conducted. In total, among 49 women, 28 (57%) with prior HDP and 21 (43%) with prior normotensive pregnancy were included, with an average of 4.9 (SD 1.2) years after delivery. Average sleep quality was suboptimal in both groups (mean PSQI score 7.0, SD 3.5 in the HDP group vs mean PSQI score 5.9, SD 2.4 in the control group; P=.22). Average sleep duration was suboptimal (6.7, SD 0.8 hours), with no difference between groups. Approximately half (n=23, 47%) had abnormal sleep timing, which was more common among those with a prior normotensive pregnancy. Sleep onset variability was high (mean 1.2, SD 0.5 hours), with no significant differences by HDP status. The mean CVH score fell within the moderate range (70.7, SD 12.8), with no differences between groups. The components of the CVH score that were lowest (ie, worst) among the entire cohort were diet (mean 37.3, SD 25.6) and BMI (mean 50.8, SD 35.4 kg/m2). Common barriers to sleep included parenting, work, and household responsibilities. The study met our criteria for the feasibility and acceptability of using the Oura ring to study sleep in this population. Among postpartum women, sleep health was suboptimal regardless of HDP history. Interventions to improve sleep and CVH should target all mothers during the first decade after childbirth.

The integration of artificial intelligence in health care presents a significant opportunity to revolutionize patient care. In the United States, an estimated 129 million people have at least 1 chronic illness, with 42% having 2 or more. Despite being largely preventable, the prevalence of chronic illness is expected to rise and impose significant economic burdens and financial toxicity on health care consumers. We leveraged an interdisciplinary team encompassing nursing, public health, and computer science to optimize health through prevention education for cardiovascular and metabolic comorbidities in persons living with HIV. In this tutorial, we describe the iterative, data-based development and evaluation of an intersectionality-informed large language model designed to support patient teaching in this population. First, we curated data by scraping publicly available, authoritative, evidence-based sources to capture a comprehensive dataset, supplemented by publicly available HIV forum content. Second, we benchmarked candidate large language models and generated a fine-tuning dataset using GPT-4 through multiturn question and answer conversations, using standardized metrics to assess baseline model performance. Third, we iteratively refined the selected model via low-rank adaptation and reinforcement learning, integrating quantitative metrics with qualitative expert evaluations. Pre-existing large language models (LLMs) demonstrated poor n-gram agreement, dissonance from model answers (accuracy 4.16, readability 4.63, and professionalism 4.58), and difficult readability (Kincaid 8.54 and Jargon 4.44). After prompt adjustments and fine-tuning, preliminary results demonstrate the potential of a customized Llama-based LLM to provide personalized, culturally salient patient education. We present a data-based, step-by-step tutorial for interdisciplinary development of CARDIO, a specialized LLM, for cardiovascular health education in HIV care. Through comprehensive data curation and scraping, systematic benchmarking, and a dual-stage fine-tuning pipeline, CARDIO's performance improved markedly (accuracy 5.0, readability 4.98, professionalism 4.98, Kincaid 7.17, and Jargon 2.92). Although patient pilot testing remains forthcoming, our results demonstrate that targeted data curation, rigorous benchmarking, and iterative fine-tuning have provided a robust evaluation of the model's potential. By building an LLM tailored to cardiovascular health promotion and patient education, this work lays the foundation for innovative artificial intelligence-driven strategies to manage comorbid conditions in people living with HIV.

Virtual reality (VR) has emerged as a promising, low-risk strategy to manage many forms of psychological stress and may be a modality to improve cardiovascular health. Recent scientific statements on the mind-heart-body connection call for better adherence to psychological screening and adoption of more holistic "behavioral cardiology" interventions that improve the overall health of patients with or at risk for cardiovascular disease (CVD). The aim of this study is to assess safety and preliminarily explore how a VR experience can aid in stress reduction among patients with or at risk for CVD. A convergent mixed methods approach was used for this single-arm prospective pilot study. In total, 20 patients were recruited from the University of California Los Angeles adult cardiology clinics and cardiac rehabilitation. Surveys and physiologic parameters were collected before, during, and after a 30-minute VR experience aimed at relaxation. The primary outcome was the State-Trait Anxiety Inventory-State (STAI-S) scale. They participated in a 90-minute visit, during which they completed surveys, including the STAI-S scale, before and after a 30-minute VR experience. Physiological parameters were also collected before, during, and after the experience. Visits concluded with semistructured interviews analyzed with inductive thematic analysis to add depth and nuance to our analysis. STAI-S scale scores after the VR experience were significantly decreased from baseline (median 31, IQR 28-38 vs median 24, IQR-29.25; P<.001). Verbal feedback revealed that participants experienced a relaxing sense of "distance from stress" moderated by unexpected, intense audiovisual components. Heart rate significantly decreased (mean 73, SD 8 vs mean 67, SD 6 beats per minute; P<.001), while blood pressure (mean systolic 128, SD 14 vs mean systolic 129, SD 18 mm Hg; P=.75 and mean diastolic 79, SD 9 vs mean diastolic 80, SD 10 mm Hg; P=.60) and galvanic skin response (mean 0.74, SD 0.89 vs mean 0.70, SD 0.57 microsiemens; P=.45) remained the same. Changes in heart rate variability parameters were consistent with increased vagal tone over time but were only statistically significant at certain time points. Survey results and interviews generally indicated safety, tolerability, and openness to using VR again. This sample of patients with CVD or risk of CVD had above-average stress, consistent with epidemiological data; the statistically and clinically significant decrease in subjective perception of stress partially converged with physiologic data. Overall, the VR intervention was a safe and feasible stress reduction method. Future research is needed to evaluate the effectiveness of this immersive therapy in reducing cardiovascular risk profiles.

Photoplethysmography-based smartwatches are increasingly used for continuous heart rate (HR) monitoring. Their accuracy has been demonstrated at rest or during low-intensity activity, but data are scarce for maximal-intensity exercise, when motion artifacts and rapid hemodynamic changes can degrade the photoplethysmography signal. Validating these devices under such demanding conditions is essential before they are applied to clinical exercise testing, athletic training, or remote health monitoring. This study aimed to evaluate the validity of the Samsung Galaxy Watch&#x202f;6 (GW6) in estimating HR throughout a graded, maximal ramp cardiopulmonary exercise test performed on a treadmill. A secondary aim was to explore whether measurement error varies across 5 predefined intensity zones (50%-60%, 60%-70%, 70%-80%, 80%-90%, and 90%-100% of the maximum HR determined individually for each participant). Overall, 55 healthy adults (30 men, 25 women; mean age 30.3, SD 8.2 years) completed a symptom-limited incremental treadmill protocol to volitional exhaustion. Simultaneous HR recordings were obtained from the GW6 (left arm) and a Polar H10 chest strap monitor, which served as the reference standards. For each intensity zone, the following agreement indices were computed: intraclass correlation coefficient (ICC), median absolute error, median absolute percentage error, and root mean squared error. Bland-Altman analysis was performed to quantify the mean bias and 95% limits of agreement between the GW6 and the Polar H10. Statistical significance was set at P<.05. Agreement between the GW6 and Polar H10 varied across exercise intensities. ICC indicated moderate to good agreement at low to moderate intensities (ICC=0.71 at 50%-60%; ICC=0.89 at 60%-70%; ICC=0.54 at 70%-80%; and ICC=0.64 at 80%-90% HRmax), and at 90%-100% of HRmax the agreement was good-to-excellent (ICC=0.90). Absolute error metrics showed stable or reduced errors with increasing intensity, with median absolute error consistently around 1-3 bpm and median absolute percentage error declining from 2.90% at 50%-60% HRmax to 0.60%-0.75% at &#x2265;70% HRmax. Root mean squared error ranged from 4.62 to 4.88 bpm across intensity zones. Bland-Altman analysis showed that the GW6 consistently underestimated HR compared with the Polar H10, with an overall mean bias of -2.67 bpm and wide limits of agreement (-16.90 to 11.57 bpm). This negative bias was present across all HR zones. The agreement was adequate for group-level comparisons but displayed substantial individual variability. The GW6 provides a good degree of validity for HR monitoring during a maximal treadmill cardiopulmonary exercise test in healthy young adults. Although measurement error increases modestly at near-maximal workloads, absolute errors remain well within clinically acceptable thresholds. These findings support the potential use of GW6 as a convenient, noninvasive alternative for HR tracking in laboratory-based exercise testing.

Wearable devices offer a promising solution for remotely monitoring heart rate (HR) during home-based cardiac rehabilitation. However, evidence regarding their accuracy across varying exercise intensities and patient profiles remains limited, particularly in populations with cardiovascular disease (CVD) such as those with heart failure (HF). The objective of this study was to evaluate the accuracy of HR measurements obtained using the Fitbit Inspire 3 during cardiopulmonary exercise testing (CPX) in patients with CVD, including those with HF. In this single-center, prospective pilot study, we enrolled 30 patients with CVD undergoing CPX. HR was simultaneously recorded using electrocardiography and the Fitbit Inspire 3 at 1-minute intervals across various CPX phases: rest, exercise below and above the anaerobic threshold (AT), and recovery. The correlation between the two methods was assessed using the Pearson correlation coefficient. Measurement error was quantified by mean absolute error and mean absolute percentage error (MAPE), with a MAPE of &#x2264;10% defined as the threshold for acceptable agreement. All data points were 630 points per minute. The Fitbit Inspire 3 device demonstrated a strong overall correlation with electrocardiography-derived HR (r=0.90; IQR 0.88-0.91) and an acceptable MAPE of 5.40% (SD 8.33%). The total error was 14.9% (94/630), with overestimation and underestimation of 37 (5.8%) points and 57 (9%) points, respectively. The rate of HR underestimation reached 19 (16%) points during exercise above the AT, compared to 1 (3%) point at rest. When stratified by HF stage (B vs C), underestimation was more pronounced in patients with HF (14/275, 5% points vs 40/355, 11.2% points). The Fitbit Inspire 3 provides acceptable validity for HR monitoring during CPX in patients with CVD. However, clinicians should interpret HR data with caution during high-intensity exercise, especially in patients with HF.

Mobile health (mHealth) interventions are increasing in popularity for the management of heart failure and coronary artery disease. The use of these interventions is dependent on rates of smartphone ownership. It is estimated that approximately 90% of the Australian adult population owns a smartphone; however, international studies suggest that smartphone ownership is significantly lower in patient populations, ranging from 34% to 91%. Smartphone ownership in patients with cardiovascular disease has not previously been examined. This study aimed to examine and compare pre-COVID-19 and post-COVID-19 pandemic smartphone ownership rates of inpatients admitted with coronary artery disease or heart failure. Data from prescreening logs of 2 multicenter randomized controlled trials, TeleClinical Care (TCC)-Pilot and TCC-Cardiac, were reviewed. TCC-Pilot recruited patients between February 2019 and March 2020. This formed the pre-COVID-19 cohort, with 377 patients screened who lived in Sydney, had a qualifying hospital admission, and had information regarding their phone ownership status. TCC-Cardiac recruited patients from July 2021 to February 2023, with 718 patients meeting the criteria and forming the post-COVID-19 cohort. Supplemental patient demographic and medical history data were collected from the electronic medical record. In the pre-COVID-19 cohort (N=377), 194 (51.5%) patients owned smartphones, 79 (21%) owned phones that were incompatible with the mHealth intervention, and the remaining 104 (27.6%) did not own a mobile phone. Smartphone owners were predominantly male (P<.001) and more often had private health insurance (P=.002). In the post-COVID-19 cohort (N=718), 366 (51%) patients owned smartphones, 106 (14.8%) owned incompatible phones, and the remaining 246 (34.3%) did not own any mobile phone. In both cohorts, younger patients were more likely to own smartphones (P<.001). Multiple comorbidities were associated with not owning a phone. Smartphone ownership accounted for just over 50% of the patients in this population. It was less common among older adults, patients with comorbidities, and those with markers of lower socioeconomic status. This needs to be considered when delivering mHealth interventions.

Hypertension is associated with a high rate of disability and mortality and leads to a substantial socioeconomic burden. Moxibustion is an external treatment in traditional Chinese medicine that has been used to treat mild to moderate hypertension in individuals with phlegm-dampness constitution and has demonstrated acupoint specificity. However, a standard large-scale randomized controlled trial is still needed to verify its effectiveness. This study is proposed to examine the clinical effectiveness and potential cardioprotective benefits of moxibustion performed at home as a treatment for individuals with phlegm-dampness hypertension. The objective of this trial is to evaluate the cardio-cerebral protective clinical efficacy of moxibustion for phlegm-dampness type hypertension and to explore its acupoint specific effects. This study is a multicenter, randomized controlled trial. A total of 120 patients with mild to moderate hypertension and phlegm-dampness constitution will be recruited and randomly assigned in a 1:1 ratio to the treatment group (acupoint: Zusanli, ST36) or the control group (acupoint: Xuanzhong, GB39). All patients will receive 12 weeks of treatment and a 12-week follow-up period. The primary outcome measure is the change in morning systolic blood pressure from baseline to week 12. The secondary outcome measures include blood pressure-related indicators (morning diastolic blood pressure, average systolic blood pressure, average diastolic blood pressure, nighttime systolic blood pressure, nighttime diastolic blood pressure, and blood pressure circadian rhythm) and short-term blood pressure variability coefficient, all of which will be measured by 24-hour ambulatory blood pressure monitoring. Additionally, cardiac-related indicators measured by 24-hour Holter monitoring, metabolic disorder-related indicators, liver and kidney function indicators, transformed scores of the traditional Chinese medicine phlegm-dampness constitution scale, and the Montreal Cognitive Assessment will also be evaluated. This study was registered on July 5, 2024, with the Chinese Clinical Trial Registry. Data collection began in June 2023 and ended in February 2025. Currently, data from this trial are in the collection phase, and no data analysis has been performed. As of January 2025, we have collected data from 118 patients. The results of this trial are expected to be submitted for publication in May 2026. This multicenter, randomized, controlled clinical trial will provide evidence on the clinical effectiveness and potential cardioprotective benefits of moxibustion performed at home as a treatment for individuals with phlegm-dampness type of hypertension. Chinese Clinical Trial Registry ChiCTR2400086582; https://www.chictr.org.cn/showproj.html?proj=211688. DERR1-10.2196/79158.

Smart textiles (ie, electronic textiles) offer a promising solution to ease continuous electrocardiogram (ECG) monitoring, but their real-world clinical application has been limited. This review comprehensively examines the current state of research on textile-based ECG monitoring systems, synthesizing current evidence with respect to performance (ie, signal quality, function under static and dynamic conditions), user experience, and current challenges. A systematic literature search across the PubMed, MEDLINE, and Embase databases from 2000 to 2025 identified 34 research papers eligible for inclusion. Textile-based ECG electrodes demonstrated good signal quality and comfort, particularly under static conditions. Nonetheless, integration into clinical practice requires addressing critical issues, which include greater efforts at validating these technologies in clinical settings and populations, as well as ensuring data security, cost&#x2011;effectiveness, user&#x2011;friendliness, and data interoperability. Considering the prominence of feasibility research, the successful clinical integration of textile-based ECG monitoring systems requires comprehensive efforts at establishing a clinical evaluation research base (via clinical trials) and developing regulatory policies.

Heart failure (HF) is a complex clinical syndrome with a high morbidity and mortality rate. Despite advancements in treatment, the recurrence of HF remains a significant challenge, often leading to deteriorating health conditions and increased pressure on the health care system. Early detection of recurrence is pivotal in mitigating and managing the adverse outcomes associated with HF. The primary objective of this study is to collect data to facilitate the identification of digital biomarkers that may indicate deterioration of the heart and, ultimately, develop algorithms that can predict HF. This prospective cohort study is conducted in Copenhagen, Denmark, and will recruit individuals diagnosed with decompensated HF. Participants will be followed for a period of 1 year, during which they will undergo a quarterly assessment period every 3 months. Each quarterly assessment period spans 7 days and involves continuous monitoring using an ambulatory electrocardiogram sensor. Throughout each quarterly assessment period, participants will also complete daily assessments and questionnaires. All data will be collected using a dedicated mobile app installed on the participants' personal smartphones and securely stored in a cloud-based system. This study is part of the Cardio-Share Model for Cross-Sectoral Ambulatory Treatment of Congestive Heart Disease Based on Personal Health Technology project. Technical and regulatory preparation started in 2023. Recruitment for this study started in January 2025 and is expected to be completed by the end of 2026. The dataset will be anonymized and published for further research. This study aims to provide a comprehensive longitudinal open-source dataset of HF recorded in real-world ambulatory conditions that enhances our understanding of HF signs and symptoms. This dataset will provide an important source for detailed analysis and understanding of HF based on ambulatory and contextual physiological data. Such insight has the potential to enhance the clinical management of individuals with HF and enable them to handle their condition at home. DERR1-10.2196/79651.

Emergency department (ED) crowding is often attributed to a slow hospitalization process, leading to reduced quality of care. Predicting early disposition in patients presenting with cardiac issues is challenging: most are ultimately discharged, yet those with a cardiac etiology frequently require hospital admission. Existing scores rely on single-time-point data and often underperform when patient risk evolves during the visit. This study aimed to develop and validate a real-time deep-learning model that fuses serial 12-lead electrocardiogram (ECG) waveforms with sequential vitals and routinely available clinical data to predict hospital admission early in ED encounters. We conducted a retrospective cohort study using the Medical Information Mart for Intensive Care (MIMIC) IV, MIMIC-IV Emergency Department module, and MIMIC-IV electrocardiogram module databases. Adults presenting with chest pain, dyspnea, syncope, or presyncope and at least 1 ECG within their ED stay were included. Two evaluation cohorts were defined: all stays with &#x2265;1 ECG (n=30,421) and a subset with &#x2265;2 ECGs during the encounter (n=11,273). To predict hospital admission, we first established 2 baseline models: a tabular model (random forest [RF]) trained on structured clinical variables, including demographics, triage acuity, past medical history, medications, and laboratory results, and an ECG-only model that learned directly from raw 12-lead waveforms. We then developed a multimodal deep-learning model that combined ECGs with sequential vital signs as well as the same static tabular features. All models were restricted to data available during the stay up to the time of the last ECG. Performance was assessed with stratified 5-fold cross-validation using identical splits across models. The multimodal model achieved an area under receiver operating characteristic (AUROC) of 0.911 when trained on all eligible stays. The model predicted disposition after the final ECG was taken, which was a median of 0.3 (IQR 0.2-5.3) hours after triage and 4.6 (IQR 2.7-7.3) hours before ED departure. Baseline models performed worse: the ECG-only model had an AUROC of 0.852, and the tabular RF had an AUROC of 0.886. In the subset requiring at least 2 ECGs within the stay, ECG-only reached an AUROC of 0.859, and RF, with the longer interval to chart tabular data, reached a higher AUROC of 0.911. The multimodal model had an AUROC of 0.924 and outperformed baselines in each cohort (paired DeLong P<.001). Serial ECGs, when integrated with evolving vitals and routine clinical features, enable accurate, early prediction of ED disposition in patients presenting with cardiac issues. This open-source, reproducible framework highlights the potential of multimodal deep learning to streamline ED flow, prioritize higher risk cases, and detect evolving, time-critical pathology.