Acoustic Voice and Speech Biomarkers of Treatment Status during Hospitalization for Acute Decompensated Heart Failure.

This study investigates acoustic voice and speech features as biomarkers for acute decompensated heart failure (ADHF), a serious escalation of heart failure symptoms including breathlessness and fatigue. ADHF-related systemic fluid accumulation in the lungs and laryngeal tissues is hypothesized to affect phonation and respiration for speech. A set of daily spoken recordings from 52 patients undergoing inpatient ADHF treatment was analyzed to identify voice and speech biomarkers for ADHF and to examine the trajectory of biomarkers during treatment. Results indicated that speakers produce more stable phonation, a more creaky voice, faster speech rates, and longer phrases after ADHF treatment compared to their pre-treatment voices. This project builds on work to develop a method of monitoring ADHF using speech biomarkers and presents a more detailed understanding of relevant voice and speech features.

Deep-Learning based Sleep Apnea Detection using SpO2 and Pulse Rate.

This work presents automated apnea event de-tection using blood oxygen saturation (SpO2) and pulse rate (PR), conveniently recorded with a pulse oximeter. A large, diverse cohort of patients (n=8068, age≥40 years) from the sleep heart health study dataset with annotated sleep events have been employed in this study. A deep-learning model is trained to detect apnea in successive 30 s epochs and performances are assessed on two independent sub-cohorts of test data. The proposed algorithm showcases the highest test performance of 90.4 % area under the receiver operating characteristic curve and 58.9% area under the precision-recall curve for epoch-based apnea detection. Additionally, the model consistently performs well across various apnea subtypes, with the highest sensitivity of 93.4 % for obstructive apnea detection followed by 90.5 % for central apnea and 89.1 % for desaturation associated hypopnea. Overall, the proposed algorithm provides a robust and sensitive approach for sleep apnea event detection using a noninvasive pulse oximeter sensor. Clinical Relevance - The study establishes high sensitivity for automated epoch-based apnea detection across a diverse study cohort with various comorbidities using simply a pulse oximeter. This highly cost-effective approach could also enable convenient sleep and health monitoring over long-term.

Pulse Rate Guided Oxygen Saturation Monitoring Using a Wearable Armband Sensor.

Continuous clinical grade measurement of SpO2 in out-of-hospital settings remains a challenge despite the widespread use of photoplethysmography (PPG) based wearable devices for health and wellness applications. This article presents two SpO2 algorithms: PRR (pulse rate derived ratio-of-ratios) and GPDR (green-assisted peak detection ratio-of-ratios), that utilize unique pulse rate frequency estimations to isolate the pulsatile (AC) component of red and infrared PPG signals and derive SpO2 measurements. The performance of the proposed SpO2 algorithms are evaluated using an upper-arm wearable device derived green, red, and infrared PPG signals, recorded in both controlled laboratory settings involving healthy subjects (n=36) and an uncontrolled clinic application involving COVID-19 patients (n=52). GPDR exhibits the lowest root mean square error (RMSE) of 1.6±0.6% for a respiratory exercise test, 3.6 ±1.0% for a standard hypoxia test, and 2.2±1.3% for an uncontrolled clinic use-case. In contrast, PRR provides relatively higher error but with greater coverage overall. Mean error across all combined datasets were 0.2±2.8% and 0.3±2.4% for PRR and GPDR respectively. Both SpO2 algorithms achieve great performance of low error with high coverage on both uncontrolled clinic and controlled laboratory conditions.

Towards Remote Continuous Monitoring of Cytokine Release Syndrome.

Cytokine release syndrome (CRS) is a noninfec-tious systemic inflammatory response syndrome condition and a principle severe adverse event common in oncology patients treated with immunotherapies. Accurate monitoring and timely prediction of CRS severity remain a challenge. This study presents an XGBoost-based machine learning algorithm for forecasting CRS severity (no CRS, mild- and severe-CRS classes) in the 24 hours following the time of prediction utilizing the common vital signs and Glasgow coma scale (GCS) questionnaire inputs. The CRS algorithm was developed and evaluated on a cohort of patients (n=1,139) surgically treated for neoplasm with no ICD9 codes for infection or sepsis during a collective 9,892 patient-days of monitoring in ICU settings. Different models were trained with unique feature sets to mimic practical monitoring environments where different types of data availability will exist. The CRS models that incorporated all time series features up to the prediction time showcased a micro-average area under curve (AUC) statistic for the receiver operating characteristic curve (ROC) of 0.94 for the 3 classes of CRS grades. Models developed on a second cohort requiring data within the 24 hours preceding prediction time showcased a relatively lower 0.88 micro-average AUROC as these models did not benefit from implicit information in the data availability. Systematic removal of blood pressure and/or GCS inputs revealed significant decreases (p<0.05) in model performances that confirm the importance of such features for CRS prediction. Accurate CRS prediction and timely intervention can reverse CRS adverse events and maximize the benefit of immunotherapies in oncology patients.

Smartphone camera based assessment of adiposity: a validation study.

Body composition is a key component of health in both individuals and populations, and excess adiposity is associated with an increased risk of developing chronic diseases. Body mass index (BMI) and other clinical or commercially available tools for quantifying body fat (BF) such as DXA, MRI, CT, and photonic scanners (3DPS) are often inaccurate, cost prohibitive, or cumbersome to use. The aim of the current study was to evaluate the performance of a novel automated computer vision method, visual body composition (VBC), that uses two-dimensional photographs captured via a conventional smartphone camera to estimate percentage total body fat (%BF). The VBC algorithm is based on a state-of-the-art convolutional neural network (CNN). The hypothesis is that VBC yields better accuracy than other consumer-grade fat measurements devices. 134 healthy adults ranging in age (21-76 years), sex (61.2% women), race (60.4% White; 23.9% Black), and body mass index (BMI, 18.5-51.6 kg/m2) were evaluated at two clinical sites (N = 64 at MGH, N = 70 at PBRC). Each participant had %BF measured with VBC, three consumer and two professional bioimpedance analysis (BIA) systems. The PBRC participants also had air displacement plethysmography (ADP) measured. %BF measured by dual-energy x-ray absorptiometry (DXA) was set as the reference against which all other %BF measurements were compared. To test our scientific hypothesis we run multiple, pair-wise Wilcoxon signed rank tests where we compare each competing measurement tool (VBC, BIA, …) with respect to the same ground-truth (DXA). Relative to DXA, VBC had the lowest mean absolute error and standard deviation (2.16 ± 1.54%) compared to all of the other evaluated methods (p < 0.05 for all comparisons). %BF measured by VBC also had good concordance with DXA (Lin's concordance correlation coefficient, CCC: all 0.96; women 0.93; men 0.94), whereas BMI had very poor concordance (CCC: all 0.45; women 0.40; men 0.74). Bland-Altman analysis of VBC revealed the tightest limits of agreement (LOA) and absence of significant bias relative to DXA (bias -0.42%, R2 = 0.03; p = 0.062; LOA -5.5% to +4.7%), whereas all other evaluated methods had significant (p < 0.01) bias and wider limits of agreement. Bias in Bland-Altman analyses is defined as the discordance between the y = 0 axis and the regressed line computed from the data in the plot. In this first validation study of a novel, accessible, and easy-to-use system, VBC body fat estimates were accurate and without significant bias compared to DXA as the reference; VBC performance exceeded those of all other BIA and ADP methods evaluated. The wide availability of smartphones suggests that the VBC method for evaluating %BF could play an important role in quantifying adiposity levels in a wide range of settings.Trial registration: ClinicalTrials.gov Identifier: NCT04854421.

Engaging frontline employees using innovation contests: Lessons from Massachusetts General Hospital.

In this article, we describe how innovation contests-a vehicle to crowdsource ideas and problem-solving efforts-propelled frontline employees to exert discretionary efforts in organizational problem-solving at Massachusetts General Hospital. As designers and administrators of four innovation contests in three disease centers, we share firsthand knowledge of how the contests enabled clinicians and administrative staff, whose primary job is delivering high-quality patient care, to become involved in ideation, selection, and implementation of their own ideas. We describe the processes that we designed and implemented, ideas that these processes generated, and findings from interviewing employees about their experiences afterwards. Our findings suggest that the benefits of implementing innovation contests were multifaceted. To employees, the contests provided a platform to voice suggestions and participate in any aspect of the innovation process that they found interesting. To managers, the contests revealed real, empirical issues affecting operation and patient care based on frontline employees' knowledge. To the organization as a whole, the contests promoted collaborative problem-solving among likeminded, innovative employees.

Cost-effectiveness of a Digital Health Intervention for Acute Myocardial Infarction Recovery.

BACKGROUND: Acute myocardial infarction (AMI) is a common cause of hospital admissions, readmissions, and mortality worldwide. Digital health interventions (DHIs) that promote self-management, adherence to guideline-directed therapy, and cardiovascular risk reduction may improve health outcomes in this population. The "Corrie" DHI consists of a smartphone application, smartwatch, and wireless blood pressure monitor to support medication tracking, education, vital signs monitoring, and care coordination. We aimed to assess the cost-effectiveness of this DHI plus standard of care in reducing 30-day readmissions among AMI patients in comparison to standard of care alone. METHODS: A Markov model was used to explore cost-effectiveness from the hospital perspective. The time horizon of the analysis was 1 year, with 30-day cycles, using inflation-adjusted cost data with no discount rate. Currencies were quantified in US dollars, and effectiveness was measured in quality-adjusted life-years (QALYs). The results were interpreted as an incremental cost-effectiveness ratio at a threshold of $100,000 per QALY. Univariate sensitivity and multivariate probabilistic sensitivity analyses tested model uncertainty. RESULTS: The DHI reduced costs and increased QALYs on average, dominating standard of care in 99.7% of simulations in the probabilistic analysis. Based on the assumption that the DHI costs $2750 per patient, use of the DHI leads to a cost-savings of $7274 per patient compared with standard of care alone. CONCLUSIONS: Our results demonstrate that this DHI is cost-saving through the reduction of risk for all-cause readmission following AMI. DHIs that promote improved adherence with guideline-based health care can reduce hospital readmissions and associated costs.

Digital Health Intervention in Acute Myocardial Infarction.

BACKGROUND: Thirty-day readmissions among patients with acute myocardial infarction (AMI) contribute to the US health care burden of preventable complications and costs. Digital health interventions (DHIs) may improve patient health care self-management and outcomes. We aimed to determine if patients with AMI using a DHI have lower 30-day unplanned all-cause readmissions than a historical control. METHODS: This nonrandomized controlled trial with a historical control, conducted at 4 US hospitals from 2015 to 2019, included 1064 patients with AMI (DHI n=200, control n=864). The DHI integrated a smartphone application, smartwatch, and blood pressure monitor to support guideline-directed care during hospitalization and through 30-days post-discharge via (1) medication reminders, (2) vital sign and activity tracking, (3) education, and (4) outpatient care coordination. The Patient Activation Measure assessed patient knowledge, skills, and confidence for health care self-management. All-cause 30-day readmissions were measured through administrative databases. Propensity score-adjusted Cox proportional hazard models estimated hazard ratios of readmission for the DHI group relative to the control group. RESULTS: Following propensity score adjustment, baseline characteristics were well-balanced between the DHI versus control patients (standardized differences <0.07), including a mean age of 59.3 versus 60.1 years, 30% versus 29% Women, 70% versus 70% White, 54% versus 54% with private insurance, 61% versus 60% patients with a non ST-elevation myocardial infarction, and 15% versus 15% with high comorbidity burden. DHI patients were predominantly in the highest levels of patient activation for health care self-management (mean score 71.7±16.6 at 30 days). The DHI group had fewer all-cause 30-day readmissions than the control group (6.5% versus 16.8%, respectively). Adjusting for hospital site and a propensity score inclusive of age, sex, race, AMI type, comorbidities, and 6 additional confounding factors, the DHI group had a 52% lower risk for all-cause 30-day readmissions (hazard ratio, 0.48 [95% CI, 0.26-0.88]). Similar results were obtained in a sensitivity analysis employing propensity matching. CONCLUSIONS: Our results suggest that in patients with AMI, the DHI may be associated with high patient activation for health care self-management and lower risk of all-cause unplanned 30-day readmissions. Registration: URL: https://www.clinicaltrials.gov; Unique identifier: NCT03760796.

Observational study on wearable biosensors and machine learning-based remote monitoring of COVID-19 patients.

Patients infected with SARS-CoV-2 may deteriorate rapidly and therefore continuous monitoring is necessary. We conducted an observational study involving patients with mild COVID-19 to explore the potentials of wearable biosensors and machine learning-based analysis of physiology parameters to detect clinical deterioration. Thirty-four patients (median age: 32 years; male: 52.9%) with mild COVID-19 from Queen Mary Hospital were recruited. The mean National Early Warning Score 2 (NEWS2) were 0.59 ± 0.7. 1231 manual measurement of physiology parameters were performed during hospital stay (median 15 days). Physiology parameters obtained from wearable biosensors correlated well with manual measurement including pulse rate (r = 0.96, p < 0.0001) and oxygen saturation (r = 0.87, p < 0.0001). A machine learning-derived index reflecting overall health status, Biovitals Index (BI), was generated by autonomous analysis of physiology parameters, symptoms, and other medical data. Daily BI was linearly associated with respiratory tract viral load (p < 0.0001) and NEWS2 (r = 0.75, p < 0.001). BI was superior to NEWS2 in predicting clinical worsening events (sensitivity 94.1% and specificity 88.9%) and prolonged hospitalization (sensitivity 66.7% and specificity 72.7%). Wearable biosensors coupled with machine learning-derived health index allowed automated detection of clinical deterioration.

Digital Health and the Care of the Patient With Arrhythmia: What Every Electrophysiologist Needs to Know.

The field of cardiac electrophysiology has been on the cutting edge of advanced digital technologies for many years. More recently, medical device development through traditional clinical trials has been supplemented by direct to consumer products with advancement of wearables and health care apps. The rapid growth of innovation along with the mega-data generated has created challenges and opportunities. This review summarizes the regulatory landscape, applications to clinical practice, opportunities for virtual clinical trials, the use of artificial intelligence to streamline and interpret data, and integration into the electronic medical records and medical practice. Preparation of the new generation of physicians, guidance and promotion by professional societies, and advancement of research in the interpretation and application of big data and the impact of digital technologies on health outcomes will help to advance the adoption and the future of digital health care.

Corrie Health Digital Platform for Self-Management in Secondary Prevention After Acute Myocardial Infarction.

BACKGROUND: Unplanned readmissions after hospitalization for acute myocardial infarction are among the leading causes of preventable morbidity, mortality, and healthcare costs. Digital health interventions could be an effective tool in promoting self-management, adherence to guideline-directed therapy, and cardiovascular risk reduction. A digital health intervention developed at Johns Hopkins-the Corrie Health Digital Platform (Corrie)-includes the first cardiology Apple CareKit smartphone application, which is paired with an Apple Watch and iHealth Bluetooth-enabled blood pressure cuff. Corrie targets: (1) self-management of cardiac medications, (2) self-tracking of vital signs, (3) education about cardiovascular disease through articles and animated videos, and (4) care coordination that includes outpatient follow-up appointments. METHODS AND RESULTS: The 3 phases of the MiCORE study (Myocardial infarction, Combined-device, Recovery Enhancement) include (1) the development of Corrie, (2) a pilot study to assess the usability and feasibility of Corrie, and (3) a prospective research study to primarily compare time to first readmission within 30 days postdischarge among patients with Corrie to patients in the historical standard of care comparison group. In Phase 2, the feasibility of deploying Corrie in an acute care setting was established among a sample of 60 patients with acute myocardial infarction. Phase 3 is ongoing and patients from 4 hospitals are being enrolled as early as possible during their hospital stay if they are 18 years or older, admitted with acute myocardial infarction (ST-segment-elevation myocardial infarction or type I non-ST-segment-elevation myocardial infarction), and own a smartphone. Patients are either being enrolled with their own personal devices or they are provided an iPhone and/or Apple Watch for the duration of the study. Phase 3 started in October 2017 and we aim to recruit 140 participants. CONCLUSIONS: This article will provide an in-depth understanding of the feasibility associated with implementing a digital health intervention in an acute care setting and the potential of Corrie as a self-management tool for acute myocardial infarction recovery.

Biovitals™: A Personalized Multivariate Physiology Analytics Using Continuous Mobile Biosensors.

More than 50% of the whole world lives with chronic diseases leading to a global economic burden of 47 trillion dollars. Healthcare organizations are moving towards managing patients outside hospital, thereby improving patient safety and quality of life. Current at-home ambulatory remote monitoring analytics based on population level thresholds of individuals physiology have shown poor outcomes and high degree of false alarm burden. The personalized multivariate physiology analytics leverages readily-available low-cost wearable biosensors to detect subtle physiology changes precursor of patient's health deterioration. In this paper we present a novel personalized multivariate physiology analytics for remote patient monitoring in an ambulatory setting. We also present our verification and validation results using perturbation testing along with clinical trial results.

Spectro-Temporal Feature Based Multi-Channel Convolutional Neural Network for ECG Beat Classification.

Automatic classification of abnormal beats in ECG signals is crucial for monitoring cardiac conditions and the performance of the classification will improve the success rate of the treatment. However, under certain circumstances, traditional classifiers cannot be adapted well to the variation of ECG morphologies or variation of different patients due to fixed hand-crafted features selection. Additionally, existing deep learning related solutions reach their limitation because they fail to use the beat-to-beat information together with single-beat morphologies. This paper applies a novel solution which converts one-dimensional ECG signal into spectro-temporal images and use multiple dense convolutional neural network to capture both beat-to-beat and single-beat information for analysis. The results of simulation on the MIT-BIH arrhythmias database demonstrate the effectiveness of the proposed methodology by showing an outstanding detection performance compared to other existing methods.

A New Test for the Detection of Direct Oral Anticoagulants (Rivaroxaban and Apixaban) in the Emergency Room Setting.

Determining whether a patient has taken a direct oral anticoagulant (DOAC) is critical during the periprocedural and preoperative period in the emergency department. However, the inaccessibility of complete medical records, along with the generally inconsistent sensitivity of conventional coagulation tests to these drugs, complicates clinical decision making and puts patients at risk of uncontrollable bleeding. In this study, we evaluate the utility of inhibitor-II-X (i-II-X), a novel, microfluidics-based diagnostic assay for the detection and identification of Factor Xa inhibitors (FXa-Is) in an acute care setting. DESIGN: First-in-human, 91-patient, single-center retrospective pilot study. SETTING: Emergency room. PATIENTS: Adult patients admitted into the emergency department, which received any clinician-ordered coagulation test requiring a 3.2% buffered sodium citrate blood collection tube. INTERVENTIONS: None. MEASUREMENTS AND MAIN RESULTS: Plasma samples from patients admitted to the emergency department were screened for the use of FXa-Is, including apixaban and rivaroxaban, within the past 24 hours using our new i-II-X microfluidic test. i-II-X results were then compared with results from conventional coagulation tests, including prothrombin time (PT) and international normalized ratio (INR), which were ordered by treating clinicians, and an anti-Xa assay for rivaroxaban. The i-II-X test detected DOACs in samples collected from the emergency department with 95.20% sensitivity and 100.00% specificity. Unlike PT and INR, i-II-X reliably identified patients who had prolonged clotting times secondary to the presence of a FXa-I. CONCLUSIONS: The i-II-X test overcomes the limitations of currently available coagulation tests and could be a useful tool by which to routinely screen patients for DOACs in emergency and critical care settings. Our new diagnostic approach is particularly relevant in clinical situations where medical records may be unavailable, or where precautions need to be taken prior to invasive interventions, such as specific reversal agent administration.

Design Rationale and Performance Evaluation of the Wavelet Health Wristband: Benchtop Validation of a Wrist-Worn Physiological Signal Recorder.

BACKGROUND: Wearable and connected health devices along with the recent advances in mobile and cloud computing provide a continuous, convenient-to-patient, and scalable way to collect personal health data remotely. The Wavelet Health platform and the Wavelet wristband have been developed to capture multiple physiological signals and to derive biometrics from these signals, including resting heart rate (HR), heart rate variability (HRV), and respiration rate (RR). OBJECTIVE: This study aimed to evaluate the accuracy of the biometric estimates and signal quality of the wristband. METHODS: Measurements collected from 35 subjects using the Wavelet wristband were compared with simultaneously recorded electrocardiogram and spirometry measurements. RESULTS: The HR, HRV SD of normal-to-normal intervals, HRV root mean square of successive differences, and RR estimates matched within 0.7 beats per minute (SD 0.9), 7 milliseconds (SD 10), 11 milliseconds (SD 12), and 1 breaths per minute (SD 1) mean absolute deviation of the reference measurements, respectively. The quality of the raw plethysmography signal collected by the wristband, as determined by the harmonic-to-noise ratio, was comparable with that obtained from measurements from a finger-clip plethysmography device. CONCLUSIONS: The accuracy of the biometric estimates and high signal quality indicate that the wristband photoplethysmography device is suitable for performing pulse wave analysis and measuring vital signs.

Mobile health apps preferences and practice among ambulatory cardiovascular patients.

AIM: To determine if patients in cardiology practices would be interested in or willing to use mobile health technologies. METHODS: Patients seen at an ambulatory cardiology clinic for any indication were included. A paper survey was administered during pre-intake that assessed frequency of use, familiarity with and interest in mobile health applications. Data were analyzed using an exploratory logistic regression analysis to determine demographic predictors for technology utilization. RESULTS: A total of 306 patients were included (a plurality, 39.3%, in age group 50-69; 62.7% male). Those from median household incomes between US$30,000 and US$74,999 and those 18-29 years old were more likely to have used a health app (0.53 and 1.21, respectively). Those between 18 and 29 years were less interested in virtual visits with their healthcare provider (-0.92) and those over age 70 were less comfortable using their phone apps (-0.80). CONCLUSION: Age and income are important predictors of mobile health app adoption.

Noninvasive Assessment of Jugular Venous Pressure via Force-Coupled Single Crystal Ultrasound.

OBJECTIVE: we have developed a handheld device for noninvasive quantitative assessment of jugular venous pressure (JVP). METHODS: we used a single crystal ultrasound coupled to a force-sensing load cell to measure JVP based on the force necessary to collapse the internal jugular vein (IJV) walls. We used a gelatin-based model system of the IJV to test the ability of single crystal ultrasound to identify the IJV and verified the cross-sectional position and diameter of the vessels with conventional imaging ultrasound. We also tested our prototype device on healthy human volunteers. RESULTS: experiments on model system demonstrated that vessel diameters determined with single crystal ultrasound were in close agreement with the diameters derived from conventional 2-D ultrasound. Proof-of-concept human experiments demonstrate that single crystal ultrasound can detect the IJV in basal and collapsed states, as compared to gold-standard sonography (insert stats). Assessment of JVP in human volunteers was physiologically consistent with and sensitive to postural changes (supine JVP 6.6 ± 2.4 mmHg; standing JVP 4.2 ± 1.9 mmHg (p < 0.0001). CONCLUSION: noninvasive assessment of JVP could prove valuable in informing rapid clinical decision-making across various pathologies and conditions leading to derangements in intravascular volume status.

Using Digital Health Technology to Better Generate Evidence and Deliver Evidence-Based Care.

As we enter the information age of health care, digital health technologies offer significant opportunities to optimize both clinical care delivery and clinical research. Despite their potential, the use of such information technologies in clinical care and research faces major data quality, privacy, and regulatory concerns. In hopes of addressing both the promise and challenges facing digital health technologies in the transformation of health care, we convened a think tank meeting with academic, industry, and regulatory representatives in December 2016 in Washington, DC. In this paper, we summarize the proceedings of the think tank meeting and aim to delineate a framework for appropriately using digital health technologies in healthcare delivery and research.

Training for our digital future: a human-centered design approach to graduate medical education for aspiring clinician-innovators.

In the current era of value-based healthcare with increasing emphasis on delivering higher quality care at lower costs, US healthcare innovation as a metric is at a premium. However, an implementation gap exists between technology-enabled innovations and patient-centered care secondary to a lack of formal training rooted in implementation science, healthcare operations, and clinical informatics for healthcare providers. We illustrate the application of human-centered design principles with focus on medical trainees as the end-user in a unique approach to developing clinician-innovators best suited to bridge the implementation gap.