Citations

Abstract Rationale Despite its effectiveness at improving outcomes in patients with COPD, pulmonary rehabilitation (PR) attrition rates are high, often exceeding 50%. Fully remote programs, like home-based PR (HBPR), face additional challenges engaging patients in real-world settings. With the growing adoption of wearables, sensor-derived behavioral measures, like sleep patterns, may identify patients most likely to benefit from remote PR. This study evaluated whether incorporating baseline sleep measures from a wrist-worn activity monitor in machine learning (ML) models improved the prediction of 12-week engagement with HBPR in patients with COPD. Methods Among participants (n = 124), the following sleep measures were collected for 1 week before HBPR: sleep duration (min), standard deviation of sleep duration, wake-after-sleep-onset (WASO) (min), standard deviation of WASO, number of awakenings, length of awakenings (min), activity counts, movement index, movement fragmentation index, and sleep fragmentation index. Sleep measures were processed (1) using a previously validated Tudor-Locke algorithm and (2) applying partial least squares-discriminant analysis (PLS-DA) to generate a single latent variable (Composite Sleep Health Score) for inclusion in models. Engagement was defined as completion of one or more recommended activities per week for the full program duration (12 weeks). To determine if the inclusion of objective sleep measures improved the prediction of HBPR engagement, nested model comparisons for parametric (logistic regression and SVM) and non-parametric (decision tree and naïve bayes) ML models were performed. A sensitivity analysis (SA) replaced the Composite Sleep Health Score with sleep regularity (duration and WASO) measures only. Results In models adjusted for age, sex, Charlson Comorbidity Index, current smoker status, modified Medical Research Council (mMRC), and forced expiratory volume in 1 second (FEV1), the inclusion of the Composite Sleep Health Score significantly improved the prediction of 12-week engagement only in SVM models (AUC 0.755; p = 0.007). Specificity (18.2%) and accuracy (68.5%) also improved by 27.3% and 7.3%, respectively. The Composite Sleep Health Score and mMRC scores ranked most frequently as the variables of highest importance. Identical trends were observed in the SA. Conclusion Although preliminary, these findings support additional investigation into the use of wearable-derived sleep measures to improve screening for HBPR eligibility, identifying patients who will clinically benefit from fully remote PR. Building on these primary findings, future researchers should carefully select variables for model inclusion and initiate new hypotheses elucidating the link between objective sleep measures and HBPR outcomes in COPD patients. This abstract is funded by: NIH R56 HL173214, NIH R01 HL140486, and the Robert D. and Patricia E. Kern Center for the Science of Health Delivery

Introduction Despite its demonstrated effectiveness at improving outcomes, pulmonary rehabilitation (PR) for chronic obstructive pulmonary disease (COPD) is underutilized. Sensor-generated data from wearable devices have the potential to mitigate this challenge by generating digital endpoints that provide insights into patient behaviors at home; however, there is no consensus on how to measure home-based PR (HBPR) outcomes with these tools. This review aims to describe (1) the most frequent digital endpoints used in HBPR studies and (2) the devices used to capture these endpoints, summarizing gaps in their applications to HBPR for COPD patients. Methods We completed a scoping review using the PRISMA checklist across databases (Web of Science, Scopus, and OVID) from January 1, 2005 to June 1, 2025. We included peer-reviewed articles on HBPR for COPD, excluding reviews, commentaries/editorials, poster abstracts, and conference proceedings. Eligible articles included cohort studies and clinical trials of adult patients (age ≥ 18 years) with COPD participating in HBPR that include one or more digital endpoints. Results Among eligible articles ( n = 218), 13 (6.0%) met inclusion criteria, the majority of which were published after 2020 (61.5%). Most studies enrolled fewer than 100 COPD patients (76.9%) for an average monitoring period of 12.5 weeks. Activity trackers were the most commonly used device (46.2%) to capture data. The most frequently used digital endpoints were step count (84.6%), time spent active (38.5%), and time spent sedentary (30.8%). Two study designs were used: randomized controlled trial (76.9%) and observational cohort. Study designs were heterogenous with more than one-third (38.5%) presenting a lack of statistically significant results. Discussion Although we identified analogous digital endpoints in some studies, dissimilar methods and study designs remain barriers to synthesizing results generated from HBPR programs for COPD. Wearable devices have the potential to build novel PR models, but more work is needed to translate real-world data into clinically meaningful measures. Future research should elucidate which participants would benefit most from and complete HBPR to build an evidence base for the validation of HBPR-relevant digital endpoints, particularly those derived from less common sources like cardiovascular and sleep measures.

Objectives: Electronic health records (EHR)-based triggers (eTriggers) have been used to study diagnostic errors in the emergency department (ED), often with suboptimal performance. Our objective was to investigate incremental value of multi-factor machine learning (ML) approaches to improve eTrigger performance. Methods: Patients presenting to an academic ED were categorized into trigger-positive and trigger-negative using standard trigger (T) definitions: (T1) ED return visits resulting in admission within 10 days; (T2) care escalation from the inpatient unit to the ICU within 24 hours; and (T3) deaths within 24 hours of admission. We trained and evaluated 6 supervised ML models. Results: A total of 124,053 consecutive encounters (5791 T-positive and 118,262 T-negative) were included. Among the T-positive, 4159 (72%) were associated with T1, 1415 (24%) with T2, and 217 (4%) with T3. The T-based positive predictive values (PPV) were 5.2% for T1, 8.2% for T2, and 6.5% for T3. ML models trained and evaluated on balanced training dataset and imbalanced test set had low classification performances (accuracy: 0.72–0.95; PPV: 0.00–0.16; F1-score: 0.00–0.23). Higher performances were observed in balanced test sets (accuracy: 0.80–0.97; PPV: 0.82–1.00; F1-score: 0.79–0.97). Comparing models trained on clinically annotated data with models trained on T-based labels identified other important factors. Conclusions: Utilizing machine learning to refine e-triggers slightly improves the identification of diagnostic errors, as evidenced by an increase in PPV values. We identified new potential factors contributing to ED diagnostic errors. These findings open new avenues to construct or modify more accurate e-triggers for diagnostic error identification.

Importance Diagnostic errors represent a major patient safety concern, with the potential to significantly impact patient outcomes. To address this, various trigger-based strategies have been developed to identify diagnostic errors, aiming to enhance clinical decision-making and improve patient safety. Objective To evaluate the performance of three pre-established triggers (T) in the emergency department (ED) setting and assess their effectiveness in detecting diagnostic errors. Design Consecutive cohort, retrospective observational design. Setting Academic ED with 80 000 annual visits. Participants Adults and children presenting to a single ED in the USA between 1 May 2018 and 1 January 2020. Intervention/outcomes Electronic health records (EHRs) were retrieved and categorised into trigger-positive and trigger-negative cases using the following criteria: T1—unscheduled returnvisits to the ED with admission within 7–10 days of theinitial visit; T2—care escalation from the inpatient unitto the intensive care unit (ICU) within 6, 12 or 24 hoursof ED admission; and T3—all deaths in the ED or within24 hours of ED admission, excluding palliative care. A random sample of trigger-positive cases was reviewed using the SaferDx tool to determine the presence or absence of a diagnostic error. Results A total of 5791 trigger-positive and 118262 trigger-negative cases were identified. Among trigger-positive cases, 4159 (72%) were associated with T1, 1415 (24%) with T2, and 217 (4%) with T3. A preliminary chart review of 462 trigger-positive and 251 trigger-negative cases showed most were error-negative (279 and 217, respectively). Detailed reviews found 32 diagnostic errors among 183 trigger-positive cases, yielding PPVs of 5.4% (T1), 8.9% (T2), and 6.9% (T3). No errors were found in 34 reviewed trigger-negative cases, resulting in a 100% NPV. Sepsis was the most common diagnosis among error-positive cases (n=11, 34.4%). Those with non-specific chief complaints like altered mental status or shortness of breath had higher diagnostic error risk. Conclusion and relevance While previously proposed EHR-based triggers can identify some diagnostic errors, they are insufficient for detecting all cases. To improve error detection performance, we recommend exploring data-driven strategies, such as machine learning techniques, to more effectively identify underlying contributing factors to diagnostic errors and enhance detection accuracy in the ED.

Abstract The accurate measurement of wrist range of motion (ROM) is crucial for assessing functional impairments and monitoring recovery in clinical and rehabilitation settings. Traditional methods, such as Goniometric measurement, while effective, require clinician involvement and provide limited dynamic data. Recent advancements in wearable and smartphone-based technologies offer promising alternatives, enabling remote self-monitoring and dynamic assessments. This paper presents a novel smartphone-based application for measuring wrist ROM, leveraging built-in accelerometers and gyroscopes to capture joint motion with high accuracy. The app provides real-time feedback through visualizations and haptic or auditory cues, enhancing usability and ensuring accurate measurements without clinician supervision. By focusing on accessibility, affordability, and ease of use, the proposed app aims to bridge the gap between clinical and home-based care, offering a scalable solution for remote rehabilitation and continuous care. This paper discusses the current state of ROM measurement technologies, the design of the app, and its potential to improve patient engagement and outcomes in both clinical and home settings.

Abstract Femtech, a sector dedicated to the development of technologies addressing women's health issues, holds significant potential to revolutionize healthcare for women. This study provides an overview of current Femtech solutions tailored to meet the unique needs of women. It analyzes the present landscape, highlighting key areas where artificial intelligence (AI) and generative technologies have made substantial impacts. Furthermore, the study explores strategies to foster future growth and broader adoption of Femtech within the healthcare sector with a glance at Artificial Intelligence. The ultimate objective is to advance the development of more effective, personalized, accessible, and equitable healthcare solutions for women. The study will review and categorize 56 women’s health-related applications and devices currently available on the US market, identifying existing gaps and areas for potential improvement.

Abstract Hypertrophic Cardiomyopathy (HCM) is an inheritable heart disease with the highest rate of sudden cardiac death (SCD) in young adults. Implantable Cardioverter Defibrillator (ICD) therapy is recommended for HCM patients at high risk for SCD. Reviewing the recent clinical literature revealed the potential to improve the selection of candidates for ICD implantation. The current study uses information extracted from echocardiography reports to evaluate HCM patients with ICD and aims to provide a comparative insight into patients who benefited the most from the device therapy, including shock and anti-tachycardia pacing (ATP). The proposed interpretable machine learning approach has used the XGboost algorithm. The model’s performance was considered satisfactory, as evidenced by an accuracy score of 81% and an area under the receiver operating characteristic curve (AUC) value of 69% and SHapley Additive exPlanations (SHAP) identified common properties of HCM patients in each category and provided high-level reasoning and foundation for a clinical decision support tool.

The analysis and interpretation of cardiac magnetic resonance (CMR) images are often time-consuming. The automated segmentation of cardiac structures can reduce the time required for image analysis. Spatial similarities between different CMR image types were leveraged to jointly segment multiple sequences using a segmentation model termed a multi-image type UNet (MI-UNet). This model was developed from 72 exams (46% female, mean age 63 ± 11 years) performed on patients with hypertrophic cardiomyopathy. The MI-UNet for steady-state free precession (SSFP) images achieved a superior Dice similarity coefficient (DSC) of 0.92 ± 0.06 compared to 0.87 ± 0.08 for a single-image type UNet (p < 0.001). The MI-UNet for late gadolinium enhancement (LGE) images also had a superior DSC of 0.86 ± 0.11 compared to 0.78 ± 0.11 for a single-image type UNet (p = 0.001). The difference across image types was most evident for the left ventricular myocardium in SSFP images and for both the left ventricular cavity and the left ventricular myocardium in LGE images. For the right ventricle, there were no differences in DCS when comparing the MI-UNet with single-image type UNets. The joint segmentation of multiple image types increases segmentation accuracy for CMR images of the left ventricle compared to single-image models. In clinical practice, the MI-UNet model may expedite the analysis and interpretation of CMR images of multiple types.