Dysphagia occurs secondary to a variety of underlying etiologies and can contribute to increased risk of adverse events such as aspiration pneumonia and premature mortality. Dysphagia is primarily diagnosed and characterized by instrumental swallowing exams such as videofluoroscopic swallowing studies. videofluoroscopic swallowing studies involve the inspection of a series of radiographic images for signs of swallowing dysfunction. Though effective, videofluoroscopic swallowing studies are only available in certain clinical settings and are not always desirable or feasible for certain patients. Because of the limitations of current instrumental swallow exams, research studies have explored the use of acceleration signals collected from neck sensors and demonstrated their potential in providing comparable radiation-free diagnostic value as videofluoroscopic swallowing studies. In this study, we used a hybrid deep convolutional recurrent neural network that can perform multi-level feature extraction (localized and across time) to annotate swallow segments automatically via multi-channel swallowing acceleration signals. In total, we used signals and videofluoroscopic swallowing study images of 3144 swallows from 248 patients with suspected dysphagia. Compared to other deep network variants, our network was superior at detecting swallow segments with an average area under the receiver operating characteristic curve value of 0.82 (95% confidence interval: 0.807-0.841), and was in agreement with up to 90% of the gold standard-labeled segments.

Introduction: Women are less likely than men to be promptly diagnosed with acute coronary syndrome (ACS) and have worse post-ACS outcomes. These diagnostic failures are partially due to ACS findings on surface ECG manifesting differently in women, which may result in unnecessary delays in treatment. To narrow health disparities, we aim to investigate the sex-specific signatures of ACS as they appear on ECGs. Methods: This was a prospective observational cohort study of chest-pain patients evaluated for suspected ACS at 3 UPMC-affiliated tertiary care hospitals. After featurization, all ECG data were separately fed into 7 machine learning classifiers to predict ACS. We examined the results by sex. We also investigated two other methods: (1) building two independent models based respectively on the female and male subgroups and (2) building a model based on the initial total sample supplemented by the patients’ sex. We used Shapley values to explain the decision-making criteria of the models. We report the results for random forest, the best performing classifier. Results: Our sample consisted of 4132 patients (Age 59 ± 16; 47% female; 15% ACS). Machine learning models continue to disproportionately underperform in females across all classifiers evaluated. The sensitivity, specificity and false negative rate in the global model blinded to sex were 82.89%, 76.22% and 17.11% for men, and 67.39%, 74.16% and 32.61% for women (p<0.0001). This statistically significant gap could not be alleviated by building sex-specific models or feeding sex to the input of the model. Indeed, the rate of false negatives in sex-specific models and global models with sex as input were 14.67% and 18.42% for men, and 34.04% and 32.61% for women (p<0.0001). The explainability analysis of the sex-specific models revealed that STT configuration in lateral leads is most informative in women, whereas STT configuration in all leads and particularly in anterolateral leads most informative in men. Conclusions: Machine learning models display crucial sex differences in the ACS signatures on the ECG that consistently put women in a detrimental situation. The alternative methods investigated here are not adequate solutions for this disparity. Thus, further investigations should be conducted.

Abstract We compared the impact of performing dual-task walking on gait quality and prefrontal cortical activation assessed by functional near-infrared spectroscopy (fNIRS). We hypothesized a greater increase in fNIRS averaged over the left prefrontal cortex during dual-task walking would be associated with a greater decrease in gait quality (increased step-time variability; decreased gait speed, cadence, smoothness, and adaptability). In older adults (n=60, 75±5.8 years, 57% female), we quantified the change in fNIRS and gait metrics from single-task walking (even surface) to walking with attentional (reciting every-other letter of the alphabet) and physical (uneven surface) dual-task challenges using four 15m repetitions of each task. Gait metrics were computed from a tri-axial accelerometer at the lower-back. Changes in fNIRS from single to dual-task walking were not associated with changes in gait quality for both attentional and physical challenges (Spearman correlations, all p>0.08). Variability in response across individuals may contribute to our findings.

Dysphagia or swallowing dysfunction is any impairment in the swallowing function that may cause difficulty or discomfort in initiating or transferring the bolus from the oral cavity into the stomach. Dysphagia can cause the bolus to reroute into the airway, known as aspiration, which can lead to more adverse outcomes such as pneumonia or even death. Videofluoroscopic swallowing study (VFSS) is the gold standard procedure for dysphagia diagnosis. In VFSS, trained clinicians calculate swallowing kinematics and inspect pathophysiological processes in a frame-by-frame manner. Though effective, VFSS evaluation is time-consuming, prone to subjectivity in judgment, and human error. In this study, we present a cascaded pipeline that employs various deep learning algorithms to automate VFSS analysis to identify swallowing abnormalities. The pipeline initially segments the VFSS video into static and dynamic frames which include all the relevant features of swallowing for the subsequent VFSS analysis tasks. These tasks include pharyngeal swallow segmentation, hyoid bone tracking, bolus segmentation, and aspiration detection. The pipeline starts with a shallow neural network (NN) that differentiates between static and dynamic VFSS frames with a 98% accuracy using spatio-temporal features from TV-L1 optical flow. Then, a Single Shot Multi-box Detector (SSD) model localizes the hyoid bone body with a mean average precision (mAP) of 40% at an intersection over union (IOU) of 0.5 in a fast and beyond average performance even when the hyoid bone is occluded by the mandible. So far, the developed automated pipeline has shown comparable performance to the manual analysis performed by trained clinicians.

Abstract Background falls and fall-related injuries are common in older adults, have negative effects on functional independence and quality of life and are associated with increased morbidity, mortality and health related costs. Current guidelines are inconsistent, with no up-to-date, globally applicable ones present. Objectives to create a set of evidence- and expert consensus-based falls prevention and management recommendations applicable to older adults for use by healthcare and other professionals that consider: (i) a person-centred approach that includes the perspectives of older adults with lived experience, caregivers and other stakeholders; (ii) gaps in previous guidelines; (iii) recent developments in e-health and (iv) implementation across locations with limited access to resources such as low- and middle-income countries. Methods a steering committee and a worldwide multidisciplinary group of experts and stakeholders, including older adults, were assembled. Geriatrics and gerontological societies were represented. Using a modified Delphi process, recommendations from 11 topic-specific working groups (WGs), 10 ad-hoc WGs and a WG dealing with the perspectives of older adults were reviewed and refined. The final recommendations were determined by voting. Recommendations all older adults should be advised on falls prevention and physical activity. Opportunistic case finding for falls risk is recommended for community-dwelling older adults. Those considered at high risk should be offered a comprehensive multifactorial falls risk assessment with a view to co-design and implement personalised multidomain interventions. Other recommendations cover details of assessment and intervention components and combinations, and recommendations for specific settings and populations. Conclusions the core set of recommendations provided will require flexible implementation strategies that consider both local context and resources.

Although machine learning has permeated many disciplines, the convergence of causal methods and machine learning remains sparse in the existing literature. Our aim was to formulate a marginal structural model in which we envisioned hypothetical (i.e. counterfactual) dynamic treatment regimes using a combination of drug therapies to manage diabetes: metformin, sulfonylurea and SGLT-2. We were interested in estimating “diabetes care provision” in next calendar year using a composite measure of chronic disease prevention and screening elements. We demonstrated the application of dynamic treatment regimes using the National Diabetes Action Canada Repository in which we applied a collection of mainstream statistical learning algorithms. We generated an ensemble of statistical learning algorithms using the SuperLearner based on the following base learners: (i) least absolute shrinkage and selection operator, (ii) ridge regression, (iii) elastic net, (iv) random forest, (v) gradient boosting machines, (vi) neural network. Each statistical learning algorithm was fitted using the pseudo-population with respect to the marginalization of the time-dependent confounding process. The covariate balance was assessed using the longitudinal (i.e. cumulative-time product) stabilized weights with calibrated restrictions. Our results indicated that the treatment drop-in cohorts (with respect to metformin, sulfonylurea and SGLT-2) may improve diabetes care provision in relation to treatment naïve cohort. As a clinical utility, we hope that this article will facilitate discussions around the prevention of adverse chronic outcomes associated with diabetes through the improvement of diabetes care provisions in primary care.

Upper esophageal sphincter opening (UESO), and laryngeal vestibule closure (LVC) are two essential kinematic events whose timings are crucial for adequate bolus clearance and airway protection during swallowing. Their temporal characteristics can be quantified through time‐consuming analysis of videofluoroscopic swallow studies (VFSS).

As different scientific disciplines begin to converge on machine learning for causal inference, we demonstrate the application of machine learning algorithms in the context of longitudinal causal estimation using electronic health records. Our aim is to formulate a marginal structural model for estimating diabetes care provisions in which we envisioned hypothetical (i.e. counterfactual) dynamic treatment regimes using a combination of drug therapies to manage diabetes: metformin, sulfonylurea and SGLT-2i. The binary outcome of diabetes care provisions was defined using a composite measure of chronic disease prevention and screening elements [27] including (i) primary care visit, (ii) blood pressure, (iii) weight, (iv) hemoglobin A1c, (v) lipid, (vi) ACR, (vii) eGFR and (viii) statin medication. We used several statistical learning algorithms to describe causal relationships between the prescription of three common classes of diabetes medications and quality of diabetes care using the electronic health records contained in National Diabetes Repository. In particular, we generated an ensemble of statistical learning algorithms using the SuperLearner framework based on the following base learners: (i) least absolute shrinkage and selection operator, (ii) ridge regression, (iii) elastic net, (iv) random forest, (v) gradient boosting machines, and (vi) neural network. Each statistical learning algorithm was fitted using the pseudo-population generated from the marginalization of the time-dependent confounding process. Covariate balance was assessed using the longitudinal (i.e. cumulative-time product) stabilized weights with calibrated restrictions. Our results indicated that the treatment drop-in cohorts (with respect to metformin, sulfonylurea and SGLT-2i) may have improved diabetes care provisions in relation to treatment naïve (i.e. no treatment) cohort. As a clinical utility, we hope that this article will facilitate discussions around the prevention of adverse chronic outcomes associated with type II diabetes through the improvement of diabetes care provisions in primary care.

Artificial intelligence and machine learning techniques have progressed dramatically and become powerful tools required to solve complicated tasks, such as computer vision, speech recognition, and natural language processing. Since these techniques have provided promising and evident results in these fields, they emerged as valuable methods for applications in human physiology and healthcare. General physiological recordings are time-related expressions of bodily processes associated with health or morbidity. Sequence classification, anomaly detection, decision making, and future status prediction drive the learning algorithms to focus on the temporal pattern and model the nonstationary dynamics of the human body. These practical requirements give birth to the use of recurrent neural networks (RNNs), which offer a tractable solution in dealing with physiological time series and provide a way to understand complex time variations and dependencies. The primary objective of this article is to provide an overview of current applications of RNNs in the area of human physiology for automated prediction and diagnosis within different fields. Finally, we highlight some pathways of future RNN developments for human physiology.