Deep cross-modal feature learning applied to predict acutely decompensated heart failure using in-home collected electrocardiography and transthoracic bioimpedance.

BACKGROUND: Deep learning has been successfully applied to ECG data to aid in the accurate and more rapid diagnosis of acutely decompensated heart failure (ADHF). Previous applications focused primarily on classifying known ECG patterns in well-controlled clinical settings. However, this approach does not fully capitalize on the potential of deep learning, which directly learns important features without relying on a priori knowledge. In addition, deep learning applications to ECG data obtained from wearable devices have not been well studied, especially in the field of ADHF prediction. METHODS: We used ECG and transthoracic bioimpedance data from the SENTINEL-HF study, which enrolled patients (≥21 years) who were hospitalized with a primary diagnosis of heart failure or with ADHF symptoms. To build an ECG-based prediction model of ADHF, we developed a deep cross-modal feature learning pipeline, termed ECGX-Net, that utilizes raw ECG time series and transthoracic bioimpedance data from wearable devices. To extract rich features from ECG time series data, we first adopted a transfer learning approach in which ECG time series were transformed into 2D images, followed by feature extraction using ImageNet-pretrained DenseNet121/VGG19 models. After data filtering, we applied cross-modal feature learning in which a regressor was trained with ECG and transthoracic bioimpedance. Then, we concatenated the DenseNet121/VGG19 features with the regression features and used them to train a support vector machine (SVM) without bioimpedance information. RESULTS: The high-precision classifier using ECGX-Net predicted ADHF with a precision of 94 %, a recall of 79 %, and an F1-score of 0.85. The high-recall classifier with only DenseNet121 had a precision of 80 %, a recall of 98 %, and an F1-score of 0.88. We found that ECGX-Net was effective for high-precision classification, while DenseNet121 was effective for high-recall classification. CONCLUSION: We show the potential for predicting ADHF from single-channel ECG recordings obtained from outpatients, enabling timely warning signs of heart failure. Our cross-modal feature learning pipeline is expected to improve ECG-based heart failure prediction by handling the unique requirements of medical scenarios and resource limitations.

An Emergency Medicine-focused Summary of the HFSA/SAEM/ISHLT Clinical Consensus Document on the Emergency Management of Patients With Ventricular Assist Devices.

Mechanical circulatory support is increasingly used as a long-term treatment option for patients with end-stage heart failure. Patients with implanted ventricular assist devices are at high risk for a range of diverse medical urgencies and emergencies. Given the increasing prevalence of mechanical circulatory support devices, this expert clinical consensus document seeks to help inform emergency medicine and prehospital providers regarding the approach to acute medical and surgical conditions encountered in these complex patients.

HFSA/SAEM/ISHLT Clinical Expert Consensus Document on the Emergency Management of Patients with Ventricular Assist Devices.

Mechanical circulatory support is now widely accepted as a viable long-term treatment option for patients with end-stage heart failure (HF). As the range of indications for the implantation of ventricular assist devices grows, so does the number of patients living in the community with durable support. Because of their underlying disease and comorbidities, in addition to the presence of mechanical support, these patients are at a high risk for medical urgencies and emergencies (Table 1). Thus, it is the responsibility of clinicians to understand the basics of their emergency care. This consensus document represents a collaborative effort by the Heart Failure Society of America, the Society for Academic Emergency Medicine, and the International Society for Heart and Lung Transplantation (ISHLT) to educate practicing clinicians about the emergency management of patients with ventricular assist devices. The target audience includes HF specialists and emergency medicine physicians, as well as general cardiologists and community-based providers.

HFSA/SAEM/ISHLT clinical expert consensus document on the emergency management of patients with ventricular assist devices.

Mechanical circulatory support is now widely accepted as a viable long-term treatment option for patients with end-stage heart failure (HF). As the range of indications for the implantation of ventricular assist devices grows, so does the number of patients living in the community with durable support. Because of their underlying disease and comorbidities, in addition to the presence of mechanical support, these patients are at a high risk for medical urgencies and emergencies (Table 1). Thus, it is the responsibility of clinicians to understand the basics of their emergency care. This consensus document represents a collaborative effort by the Heart Failure Society of America, the Society for Academic Emergency Medicine, and the International Society for Heart and Lung Transplantation (ISHLT) to educate practicing clinicians about the emergency management of patients with ventricular assist devices. The target audience includes HF specialists and emergency medicine physicians, as well as general cardiologists and community-based providers.

Clinically significant ventricular arrhythmias and progression of depression and anxiety following an acute coronary syndrome.

BACKGROUND: Depression and anxiety are common and associated with worse clinical outcomes in patients who experience an acute coronary syndrome (ACS). We investigated the association between major ventricular arrhythmias (VAs) with the progression of depression and anxiety among hospital survivors of an ACS. METHODS: Patients were interviewed in hospital and by telephone up to 12 months after hospital discharge. The primary outcome was the presence of moderate/severe symptoms of depression and anxiety defined as a Patient Health Questionnaire (PHQ)-9 score ≥ 10 and a Generalized Anxiety Disorder (GAD)-7 score ≥ 10 at baseline and 1 month and PHQ-2 ≥ 3 and GAD-2 ≥ 3 at 3, 6, and 12 months. We used marginal models to examine the association between major VAs and the symptoms of depression or anxiety over time. RESULTS: The average age of the study population (n = 2074) was 61.1 years, 33.5% were women, and 78.3% were white. VAs developed in 105 patients (5.1%). Symptoms of depression and anxiety were present in 22.2% and 23.5% of patients at baseline, respectively, and declined to 14.1% and 12.6%, respectively, at 1-month post-discharge. VAs were not significantly associated with the progression of symptoms of depression (adjusted relative risk [aRR] = 1.29, 95% confidence interval [CI] = 0.94-1.77) and anxiety (aRR = 1.22, 95% CI = 0.86-1.72), or with change in average scores of PHQ-2 and GAD-2 over time, both before and after risk adjustment. CONCLUSION: The prevalence of symptoms of depression and anxiety was high after an ACS but declined thereafter and may not be associated with the occurrence of major in-hospital VAs.

Twenty-five year trends (1986-2011) in hospital incidence and case-fatality rates of ventricular tachycardia and ventricular fibrillation complicating acute myocardial infarction.

BACKGROUND: Long-term trends in the incidence rates (IRs) and hospital case-fatality rates (CFRs) of ventricular tachycardia (VT) and ventricular fibrillation (VF) among patients hospitalized with acute myocardial infarction (AMI) have not been recently examined. METHODS: We used data from 11,825 patients hospitalized with AMI at all 11 medical centers in central Massachusetts on a biennial basis between 1986 and 2011. Multivariable adjusted logistic regression modeling was used to examine trends in hospital IRs and CFRs of VT and VF complicating AMI. RESULTS: The median age of the study population was 71 years, 57.9% were men, and 94.7% were white. The hospital IRs declined from 14.3% in 1986/1988 to 10.5% in 2009/2011 for VT and from 8.2% to 1.7% for VF. The in-hospital CFRs declined from 27.7% to 6.9% for VT and from 49.6% to 36.0% for VF between 1986/1988 and 2009/2011, respectively. The IRs of both early (<48 hours) and late VT and VF declined over time, with greater declines in those of late VT and VF. The incidence rates of VT declined similarly for patients with either an ST-segment elevation myocardial infarction (STEMI) or non-STEMI, whereas they only declined in those with VF and a STEMI. CONCLUSIONS: The hospital IRs and CHRs of VT and VF complicating AMI have declined over time, likely because of changes in acute monitoring and treatment practices. Despite these encouraging trends, efforts remain needed to identify patients at risk for these serious ventricular arrhythmias so that preventive and treatment strategies might be implemented as necessary.

Hyperglycemia and risk of ventricular tachycardia among patients hospitalized with acute myocardial infarction.

BACKGROUND: Little is known about the association of hyperglycemia with the development of ventricular tachycardia (VT) in patients hospitalized with acute myocardial infarction (AMI) which we examined in the present study. The objectives of this community-wide observational study were to examine the relation between elevated serum glucose levels at the time of hospital admission for AMI and occurrence of VT, and time of occurrence of VT, during the patient's acute hospitalization. METHODS: We used data from a population-based study of patients hospitalized with AMI at all central Massachusetts medical centers between 2001 and 2011. Hyperglycemia was defined as a serum glucose level ≥ 140 mg/dl at the time of hospital admission. The development of VT was identified from physicians notes and electrocardiographic findings by our trained team of data abstractors. RESULTS: The average age of the study population was 70 years, 58.0% were men, and 92.7% were non-Hispanic whites. The mean and median serum glucose levels at the time of hospital admission were 171.4 mg/dl and 143.0, respectively. Hyperglycemia was present in 51.9% of patients at the time of hospital admission; VT occurred in 652 patients (15.8%), and two-thirds of these episodes occurred during the first 48 h after hospital admission (early VT). After multivariable adjustment, patients with hyperglycemia were at increased risk for developing VT (adjusted OR = 1.48, 95% CI = 1.23-1.78). The presence of hyperglycemia was significantly associated with early (multivariable adjusted OR = 1.39, 95% CI = 1.11-1.73) but not with late VT. Similar associations were observed in patients with and without diabetes and in patients with and without ST-segment elevation AMI. CONCLUSIONS: Efforts should be made to closely monitor and treat patients who develop hyperglycemia, especially early after hospital admission, to reduce their risk of VT.

Online health information seeking by adults hospitalized for acute coronary syndromes: Who looks for information, and who discusses it with healthcare providers?

OBJECTIVE: To describe characteristics associated with online health information-seeking and discussing resulting information with healthcare providers among adults with acute coronary syndromes (ACS). METHODS: Consecutive patients hospitalized with ACS in 6 hospitals in Massachusetts and Georgia who reported Internet use in the past 4 weeks (online patients) were asked about online health information-seeking and whether they discussed information with healthcare providers. Participants reported demographic and psychosocial characteristics; clinical characteristics were abstracted from medical records. Logistic regression models estimated associations with information-seeking and provider communication. RESULTS: Online patients (N = 1142) were on average aged 58.8 (SD: 10.6) years, 30.3% female, and 82.8% non-Hispanic white; 56.7% reported online health information-seeking. Patients with higher education and difficulty accessing medical care were more likely to report information-seeking; patients hospitalized with myocardial infarction, and those with impaired health numeracy and limited social networks were less likely. Among information-seekers, 33.9% discussed information with healthcare providers. More education and more frequent online information-seeking were associated with provider discussions. CONCLUSION: Over half of online patients with ACS seek health information online, but only 1 in 3 of these discuss information with healthcare providers. PRACTICE IMPLICATIONS: Clinician awareness of patient information-seeking may enhance communication including referral to evidence-based online resources.

Using support vector machines on photoplethysmographic signals to discriminate between hypovolemia and euvolemia.

Identifying trauma patients at risk of imminent hemorrhagic shock is a challenging task in intraoperative and battlefield settings given the variability of traditional vital signs, such as heart rate and blood pressure, and their inability to detect blood loss at an early stage. To this end, we acquired N = 58 photoplethysmographic (PPG) recordings from both trauma patients with suspected hemorrhage admitted to the hospital, and healthy volunteers subjected to blood withdrawal of 0.9 L. We propose four features to characterize each recording: goodness of fit (r2), the slope of the trend line, percentage change, and the absolute change between amplitude estimates in the heart rate frequency range at the first and last time points. Also, we propose a machine learning algorithm to distinguish between blood loss and no blood loss. The optimal overall accuracy of discriminating between hypovolemia and euvolemia was 88.38%, while sensitivity and specificity were 88.86% and 87.90%, respectively. In addition, the proposed features and algorithm performed well even when moderate blood volume was withdrawn. The results suggest that the proposed features and algorithm are suitable for the automatic discrimination between hypovolemia and euvolemia, and can be beneficial and applicable in both intraoperative/emergency and combat casualty care.

A Historical Perspective on Presentations of Hypertensive Acute Heart Failure.

BACKGROUND: The initial systolic blood pressure (SBP) in patients presenting to the hospital with acute heart failure (AHF) informs prognosis, diagnosis, and guides initial treatment. However, over time AHF presentations with elevated SBP appear to have declined. The present study examined whether the frequency of AHF presentations with systolic hypertension (SBP >160 mmHg) declined over a nearly two-decade time interval. METHODS: This study compares four historical, cross-sectional cohorts with AHF who were admitted to tertiary care medical centres in the North-eastern USA in 1995, 2000, 2006, and 2011-13. The main outcome was the proportion of AHF patients presenting with an initial SBP >160 mmHg. RESULTS: 2,366 patients comprised the study sample. The average age was 77 years, 55% were female, 94% white, and 75% had prior heart failure. In 1995, 34% of AHF patients presented with an initial SBP >160 mmHg compared to 20% in 2011-2013 (p<0.01). Multivariate logistic regression demonstrated reduced odds of presenting with a SBP >160 mmHg in 2006 (0.64, 95% CI 0.42-0.96) and 2011-13 (0.46, 95% CI 0.28-0.74) compared with patients in 1995. CONCLUSION: The proportion of patients with AHF and initial SBP >160 mmHg significantly declined over the study time period. There are several potential reasons for this observation and these findings highlight the need for ongoing surveillance of patients with AHF as changing clinical characteristics can impact early treatment decisions.

A Robust Motion Artifact Detection Algorithm for Accurate Detection of Heart Rates From Photoplethysmographic Signals Using Time-Frequency Spectral Features.

Motion and noise artifacts (MNAs) impose limits on the usability of the photoplethysmogram (PPG), particularly in the context of ambulatory monitoring. MNAs can distort PPG, causing erroneous estimation of physiological parameters such as heart rate (HR) and arterial oxygen saturation (SpO2). In this study, we present a novel approach, "TifMA," based on using the time-frequency spectrum of PPG to first detect the MNA-corrupted data and next discard the nonusable part of the corrupted data. The term "nonusable" refers to segments of PPG data from which the HR signal cannot be recovered accurately. Two sequential classification procedures were included in the TifMA algorithm. The first classifier distinguishes between MNA-corrupted and MNA-free PPG data. Once a segment of data is deemed MNA-corrupted, the next classifier determines whether the HR can be recovered from the corrupted segment or not. A support vector machine (SVM) classifier was used to build a decision boundary for the first classification task using data segments from a training dataset. Features from time-frequency spectra of PPG were extracted to build the detection model. Five datasets were considered for evaluating TifMA performance: (1) and (2) were laboratory-controlled PPG recordings from forehead and finger pulse oximeter sensors with subjects making random movements, (3) and (4) were actual patient PPG recordings from UMass Memorial Medical Center with random free movements and (5) was a laboratory-controlled PPG recording dataset measured at the forehead while the subjects ran on a treadmill. The first dataset was used to analyze the noise sensitivity of the algorithm. Datasets 2-4 were used to evaluate the MNA detection phase of the algorithm. The results from the first phase of the algorithm (MNA detection) were compared to results from three existing MNA detection algorithms: the Hjorth, kurtosis-Shannon entropy, and time-domain variability-SVM approaches. This last is an approach recently developed in our laboratory. The proposed TifMA algorithm consistently provided higher detection rates than the other three methods, with accuracies greater than 95% for all data. Moreover, our algorithm was able to pinpoint the start and end times of the MNA with an error of less than 1 s in duration, whereas the next-best algorithm had a detection error of more than 2.2 s. The final, most challenging, dataset was collected to verify the performance of the algorithm in discriminating between corrupted data that were usable for accurate HR estimations and data that were nonusable. It was found that on average 48% of the data segments were found to have MNA, and of these, 38% could be used to provide reliable HR estimation.

Relation of Atrial Fibrillation in Acute Myocardial Infarction to In-Hospital Complications and Early Hospital Readmission.

Atrial fibrillation (AF) is a common complication of acute myocardial infarction (AMI) and contributes to high rates of in-hospital adverse events. However, there are few contemporary studies examining rates of AF in the contemporary era of AMI or the impact of new-onset AF on key in-hospital and postdischarge outcomes. We examined trends in AF in 6,384 residents of Worcester, Massachusetts, who were hospitalized with confirmed AMI during 7 biennial periods between 1999 and 2011. Multivariate logistic regression analysis was used to examine associations between occurrence of AF and various in-hospital and postdischarge complications. The overall incidence of AF complicating AMI was 10.8%. Rates of new-onset AF increased from 1999 to 2003 (9.8% to 13.2%), and decreased thereafter. In multivariable adjusted models, patients developing new-onset AF after AMI were at a higher risk for in-hospital stroke (odds ratio [OR] 2.5, 95% confidence interval [CI] 1.6 to 4.1), heart failure (OR 2.0, 95% CI 1.7 to 2.4), cardiogenic shock (OR 3.7, 95% CI 2.8 to 4.9), and death (OR 2.3, 95% CI 1.9 to 3.0) than patients without AF. Development of AF during hospitalization for AMI was associated with higher rates of readmission within 30 days after discharge (21.7% vs 16.0%), but no significant difference was noted in early postdischarge 30-day all-cause mortality rates (8.3% vs 5.1%). In conclusion, new-onset AF after AMI is strongly related to in-hospital complications of AMI and higher short-term readmission rates.

PULSE-SMART: Pulse-Based Arrhythmia Discrimination Using a Novel Smartphone Application.

BACKGROUND: Atrial fibrillation (AF) is a common and dangerous rhythm abnormality. Smartphones are increasingly used for mobile health applications by older patients at risk for AF and may be useful for AF screening. OBJECTIVES: To test whether an enhanced smartphone app for AF detection can discriminate between sinus rhythm (SR), AF, premature atrial contractions (PACs), and premature ventricular contractions (PVCs). METHODS: We analyzed two hundred and nineteen 2-minute pulse recordings from 121 participants with AF (n = 98), PACs (n = 15), or PVCs (n = 15) using an iPhone 4S. We obtained pulsatile time series recordings in 91 participants after successful cardioversion to sinus rhythm from preexisting AF. The PULSE-SMART app conducted pulse analysis using 3 methods (Root Mean Square of Successive RR Differences; Shannon Entropy; Poincare plot). We examined the sensitivity, specificity, and predictive accuracy of the app for AF, PAC, and PVC discrimination from sinus rhythm using the 12-lead EKG or 3-lead telemetry as the gold standard. We also administered a brief usability questionnaire to a subgroup (n = 65) of app users. RESULTS: The smartphone-based app demonstrated excellent sensitivity (0.970), specificity (0.935), and accuracy (0.951) for real-time identification of an irregular pulse during AF. The app also showed good accuracy for PAC (0.955) and PVC discrimination (0.960). The vast majority of surveyed app users (83%) reported that it was "useful" and "not complex" to use. CONCLUSION: A smartphone app can accurately discriminate pulse recordings during AF from sinus rhythm, PACs, and PVCs.

Magnitude of and Prognostic Factors Associated With 1-Year Mortality After Hospital Discharge for Acute Decompensated Heart Failure Based on Ejection Fraction Findings.

BACKGROUND: Limited data exist about the magnitude of and the factors associated with prognosis within 1 year for patients discharged from the hospital after acute decompensated heart failure. Data are particularly limited from the more generalizable perspective of a population-based investigation and should be further stratified according to currently recommended ejection fraction (EF) findings. METHODS AND RESULTS: The hospital medical records of residents of the Worcester, Massachusetts, metropolitan area who were discharged after acute decompensated heart failure from all 11 medical centers in central Massachusetts during 1995, 2000, 2002, 2004, and 2006 were reviewed. The average age of the 4025 study patients was 75 years, 93% were white, and 44% were men. Of these, 35% (n=1414) had reduced EF (≤40%), 13% (n=521) had borderline preserved EF (41-49%), and 52% (n=2090) had preserved EF (≥50%); at 1 year after discharge, death rates were 34%, 30%, and 29%, respectively (P=0.03). Older age, a history of chronic obstructive pulmonary disease, systolic blood pressure findings <150 mm Hg on admission, and hyponatremia were important predictors of 1-year mortality for all study patients, whereas several comorbidities and physiological factors were differentially associated with 1-year death rates in patients with reduced, borderline preserved, and preserved EF. CONCLUSIONS: This population-based study highlights the need for further contemporary research into the characteristics, treatment practices, natural history, and long-term outcomes of patients with acute decompensated heart failure and varying EF findings and reinforces ongoing discussions about whether different treatment guidelines may be needed for these patients to design more personalized treatment plans.

Detecting Heart Failure Decompensation by Measuring Transthoracic Bioimpedance in the Outpatient Setting: Rationale and Design of the SENTINEL-HF Study.

BACKGROUND: Recurrent hospital admissions are common among patients admitted for acute decompensated heart failure (ADHF), but identification of patients at risk for rehospitalization remains challenging. Contemporary heart failure (HF) management programs have shown modest ability to reduce readmissions, partly because they monitor signs or symptoms of HF worsening that appear late during decompensation. Detecting early stages of HF decompensation might allow for immediate application of effective HF therapies, thereby potentially reducing HF readmissions. One of the earliest indicators of HF decompensation is intrathoracic fluid accumulation, which can be assessed using transthoracic bioimpedance. OBJECTIVE: The SENTINEL-HF study is a prospective observational study designed to test a novel, wearable HF monitoring system as a predictor of HF decompensation among patients discharged after hospitalization for ADHF. METHODS: SENTINEL-HF tests the hypothesis that a decline in transthoracic bioimpedance, as assessed daily with the Philips fluid accumulation vest (FAV) and transmitted using a mobile phone, is associated with HF worsening and rehospitalization. According to pre-specified power calculations, 180 patients admitted with ADHF are enrolled. Participants transmit daily self-assessments using the FAV-mobile phone dyad for 45 days post-discharge. The primary predictor is the deviation of transthoracic bioimpedance for 3 consecutive days from a patient-specific normal variability range. The ADHF detection algorithm is evaluated in relation with a composite outcome of HF readmission, diuretic up-titration, and self-reported HF worsening (Kansas City Cardiomyopathy Questionnaire) during a 90-day follow-up period. Here, we provide the details and rationale of SENTINEL-HF. RESULTS: Enrollment in the SENTINEL-HF study is complete and the 90-days follow-up is currently under way. Once data collection is complete, the study dataset will be used to evaluate our ADHF detection algorithm and the results submitted for publication. CONCLUSION: SENTINEL-HF emerged from our long-term vision that advanced home monitoring technology can improve the management of chronic HF by extending clinical care into patients' homes. Monitoring transthoracic bioimpedance with the FAV may identify patients at risk of recurrent HF decompensation and enable timely preventive measures. TRIAL REGISTRATION: Clinicaltrials.gov NCT01877369: https://clinicaltrials.gov/ct2/show/NCT01877369 (Archived by WebCite at http://www.webcitation.org/6bDYl0dGy).

Patient safety incident capture resulting from incident reports: a comparative observational analysis.

BACKGROUND: Patient safety incident (PSI) discovery is an essential component of quality improvement. When submitted, incident reports may provide valuable opportunities for PSI discovery. However, little objective information is available to date to quantify or demonstrate this value. The objective of this investigation was to assess how often Emergency Department (ED) incident reports submitted by different sources led to the discovery of PSIs. METHODS: A standardized peer review process was implemented to evaluate all incident reports submitted to the ED. Findings of the peer review analysis were recorded prospectively in a quality improvement database. A retrospective analysis of the quality improvement database was performed to calculate the PSI capture rates for incident reports submitted by different source groups. RESULTS: 363 incident reports were analyzed over a period of 18 months; 211 were submitted by healthcare providers (HCPs) and 126 by non-HCPs. PSIs were identified in 108 resulting in an overall capture rate of 31%. HCP-generated reports resulted in a 44% capture rate compared to 10% for non-HCPs (p < 0.001). There was no difference in PSI capture between sub-groups of HCPs and non-HCPs. CONCLUSION: HCP-generated ED incident reports were much more likely to capture PSIs than reports submitted by non-HCPs. However, HCP reports still led to PSI discovery less than half the time. Further research is warranted to develop effective strategies to improve the utility of incident reports from both HCPs and non-HCPs.

Cognitive status in patients hospitalized with acute decompensated heart failure.

BACKGROUND: Cognitive impairment is highly prevalent in patients with heart failure and is associated with adverse outcomes. However, whether specific cognitive abilities (eg, memory vs executive function) are impaired in heart failure has not been fully examined. We investigated the prevalence of impairment in 3 cognitive domains in patients hospitalized with acute decompensated heart failure (ADHF) and the associations of impairment with demographic and clinical characteristics. METHODS: The sample included 744 patients hospitalized with ADHF (mean age 72 years, 46% female) at 5 medical centers. Impairment was assessed in 3 cognitive domains (memory, processing speed, executive function) using standardized measures. Demographic and clinical characteristics were obtained from a structured interview and medical record review. RESULTS: A total of 593 (80%) of 744 patients were impaired in at least 1 cognitive domain; 32%, 31%, and 17% of patients were impaired in 1, 2, or all 3 cognitive domains, respectively. Patients impaired in more than 1 cognitive domain were significantly older, had less formal education, and had more noncardiac comorbidities (all P values < .05). In multivariable adjusted analyses, patients with older age and lower education had higher odds of impairment in 2 or more cognitive domains. Depressed patients had twice the odds of being impaired in all 3 cognitive domains (odds ratio 1.98, 95% CI 1.08-3.64). CONCLUSION: Impairments in executive function, processing speed, and memory are common among patients hospitalized for ADHF. Recognition of these prevalent cognitive deficits is critical for the clinical management of these high-risk patients.

Photoplethysmograph signal reconstruction based on a novel hybrid motion artifact detection-reduction approach. Part I: Motion and noise artifact detection.

Motion and noise artifacts (MNA) are a serious obstacle in utilizing photoplethysmogram (PPG) signals for real-time monitoring of vital signs. We present a MNA detection method which can provide a clean vs. corrupted decision on each successive PPG segment. For motion artifact detection, we compute four time-domain parameters: (1) standard deviation of peak-to-peak intervals (2) standard deviation of peak-to-peak amplitudes (3) standard deviation of systolic and diastolic interval ratios, and (4) mean standard deviation of pulse shape. We have adopted a support vector machine (SVM) which takes these parameters from clean and corrupted PPG signals and builds a decision boundary to classify them. We apply several distinct features of the PPG data to enhance classification performance. The algorithm we developed was verified on PPG data segments recorded by simulation, laboratory-controlled and walking/stair-climbing experiments, respectively, and we compared several well-established MNA detection methods to our proposed algorithm. All compared detection algorithms were evaluated in terms of motion artifact detection accuracy, heart rate (HR) error, and oxygen saturation (SpO2) error. For laboratory controlled finger, forehead recorded PPG data and daily-activity movement data, our proposed algorithm gives 94.4, 93.4, and 93.7% accuracies, respectively. Significant reductions in HR and SpO2 errors (2.3 bpm and 2.7%) were noted when the artifacts that were identified by SVM-MNA were removed from the original signal than without (17.3 bpm and 5.4%). The accuracy and error values of our proposed method were significantly higher and lower, respectively, than all other detection methods. Another advantage of our method is its ability to provide highly accurate onset and offset detection times of MNAs. This capability is important for an automated approach to signal reconstruction of only those data points that need to be reconstructed, which is the subject of the companion paper to this article. Finally, our MNA detection algorithm is real-time realizable as the computational speed on the 7-s PPG data segment was found to be only 7 ms with a Matlab code.

Emergency department patient safety incident characterization: an observational analysis of the findings of a standardized peer review process.

BACKGROUND: Emergency Department (ED) care has been reported to be prone to patient safety incidents (PSIs). Improving our understanding of PSIs is essential to prevent them. A standardized, peer review process was implemented to identify and analyze ED PSIs. The primary objective of this investigation was to characterize ED PSIs identified by the peer review process. A secondary objective was to characterize PSIs that led to patient harm. In addition, we sought to provide a detailed description of the peer review process for others to consider as they conduct their own quality improvement initiatives. METHODS: An observational study was conducted in a large, urban, tertiary-care ED. Over a two-year period, all ED incident reports were investigated via a standardized, peer review process. PSIs were identified and analyzed for contributing factors including systems failures and practitioner-based errors. The classification system for factors contributing to PSIs was developed based on systems previously reported in the emergency medicine literature as well as the investigators' experience in quality improvement and peer review. All cases in which a PSI was discovered were further adjudicated to determine if patient harm resulted. RESULTS: In 24 months, 469 cases were investigated, identifying 152 PSIs. In total, 188 systems failures and 96 practitioner-based errors were found to have contributed to the PSIs. In twelve cases, patient harm was determined to have resulted from PSIs. Systems failures were identified in eleven of the twelve cases in which a PSI resulted in patient harm. CONCLUSION: Systems failures were almost twice as likely as practitioner-based errors to contribute to PSIs, and systems failures were present in the majority of cases resulting in patient harm. To effectively reduce PSIs, ED quality improvement initiatives should focus on systems failure reduction.

Decade-long trends in the magnitude, treatment, and outcomes of patients aged 30 to 54 years hospitalized with ST-segment elevation and non-ST-segment elevation myocardial infarction.

Although acute myocardial infarction (AMI) occurs primarily in the elderly, this disease also affects young adults. Few studies have, however, presented data on relatively young patients hospitalized with AMI. The objectives of this population-based study were to examine recent trends in the magnitude, clinical characteristics, management, and in-hospital and long-term outcomes associated with ST-segment elevation myocardial infarction (STEMI) and non-ST-segment elevation myocardial infarction (NSTEMI) in patients aged 30 to 54 years. We reviewed the medical records of 955 residents of the Worcester (Massachusetts) metropolitan area aged 30 to 54 years who were hospitalized for an initial STEMI or NSTEMI in 6 biennial periods from 1999 to 2009 at 11 greater Worcester medical centers. From 1999 to 2009, the proportion of young adults hospitalized with an STEMI decreased from approximately 2/3 to 2/5 of all patients with an initial AMI. Patients with STEMI were less likely to have a history of heart failure, hypertension, hyperlipidemia, and kidney disease than those with NSTEMI. Both groups received similar effective medical therapies during their acute hospitalization. In-hospital clinical complications and mortality were low and no significant differences in these end points were observed between patients with STEMI and NSTEMI or with regard to 1-year postdischarge death rates (1.9% vs 2.8%). The present results demonstrate recent decreases in the proportion of relatively young patients diagnosed with an initial STEMI. Patients with STEMI and NSTEMI had similar in-hospital outcomes and long-term survival. Trends in these and other important outcomes warrant continued monitoring.