Thesis (M.S.)--University of Michigan, 1993.

Thesis (M.S.)--University of Michigan, 1993.

Thesis (M.S.)--University of Michigan, 2000. / Running title: Measurement of quality-of-life in heart failure. Includes bibliographical references.

Chronic heart failure (HF) is a condition that affects more than 900,000 people in the UK. Mortality rates associated with the condition are high, with nearly 20% of patients dying within one year of diagnosis. Continuous monitoring and risk stratification can help identify patients at risk of deterioration and may consequently improve patients' likelihood of survival. Current repeated-measure risk stratification techniques for HF patients often rely on subjective perception of symptoms, such as breathlessness, and markers of fluid retention in the body (e.g. weight). Despite the common use of such markers, studies have shown that they offer limited effectiveness in predicting HF-related events. This thesis set out to identify and evaluate new markers for repeated-measure risk stratification of HF patients. It started with an exploration of traditional HF measurements, including weight, blood pressure, heart rate and symptom scores, and aimed to improve the performance of these measurements using a data-driven approach. A multi-variate model was developed from data acquired during a randomised controlled trial of remotely-monitored HF patients. The rare occurrence of HF-related adverse events during the trial required the developement of a careful methodology. This methodology helped identify the markers with most predictive ability, which achieved moderate performance at identifying patients at risk of HF-related adverse events, clearly outperforming commonly-used thresholds. Subsequently, this thesis explored the potential value of additional, accelerometer-derived physical activity (PA) and sleep markers. For this purpose, the ability of accelerometer-derived markers to differentiate between individuals with and without HF was evaluated. It was found that markers that summarise the frequency and duration of different PA intensities performed best at differentiating between the two groups and may therefore be most suitable for future use in repeated-measure applications. As part of the analysis of accelerometer-derived HF markers, a gap in the methodology of automated accelerometer processing was identified, namely the need for self-reported sleep-onset and wake-up information. As a result, Chapter 5 of this thesis describes the development and evaluation of a data-driven solution for this problem. In summary, this thesis explored both traditional and new, accelerometer-derived markers for the early detection of HF deterioration. It utilised sound methodology to overcome limitations faced by sparse and unbalanced datasets and filled a methodological gap in the processing of signals from wrist-worn accelerometers.

京都大学 / 0048 / 新制・課程博士 / 博士(医学) / 甲第22042号 / 医博第4527号 / 新制||医||1039(附属図書館) / 京都大学大学院医学研究科医学専攻 / (主査)教授 佐藤 俊哉, 教授 湊谷 謙司, 教授 稲垣 暢也 / 学位規則第4条第1項該当 / Doctor of Medical Science / Kyoto University / DFAM

Computational ECG (electrocardiogram) analysis enables accurate and faster diagnosis and early prediction of heart failure related symptoms (heart failure worsening). Machine learning, particularly deep learning, has been applied for ECG data successfully. The previous applications, however, either mainly focused on classifying occurrent, known patterns of on-going heart failure or heart failure related diseases such arrhythmia, which have undesirable predictability beforehand, or emphasizing on data from pre-processed public database data. In this dissertation, we developed an approach, however, does not fully capitalize on the potential of deep learning, which directly learns important features from raw input data without relying on a priori knowledge. Here, we present a deep transfer learning pipeline which combines an image-based pretrained deep neural network model with manifold learning to predict the precursors of heart failure (heart failure-worsening and recurrent heart failure related re-hospitalization) using raw ECG time series from wearable devices. In this dissertation, we used the unprocessed real-life ECG data from the SENTINEL-HF study by Dovancescu, et al. to predict the precursors of heart failure worsening. To extract rich features from ECG time series, we took a deep transfer learning approach where 1D time-series of five heartbeats were transformed to 2D images by Gramian Angular Summation Field (GASF) and then the pretrained models, VGG19 were used for feature extraction. Then, we applied UMAP (Uniform Manifold Approximation and Projection) to capture the manifold of the standardized feature space and reduce the dimension, followed by SVM (Support Vector Machine) training. Using our pipeline, we demonstrated that our classifier was able to predict heart failure worsening with 92.1% accuracy, 92.9% precision, 92.6% recall and F1 score of 0.93 bypassing the detection of known abnormal ECG patterns. In conclusion, we demonstrate the feasibility of early alerts of heart failure by predicting the precursor of heart failure worsening based on raw ECG signals. We expected that our approached provided an innovative method to assess the recovery and successfulness for the treatment patient received during the first hospitalization, to predict whether recurrent heart failure is likely to occur, and to evaluate whether the patient should be discharged.

No description available.

No description available.

Reducing heart failure risk standardized readmissions rates (RSRRs) continues to be a challenge in the United States. Among Medicare beneficiaries, the U.S. national rate for heart failure RSRRs is 23, and Georgia only has 3 hospitals with heart failure RSRRs that are better than the national rate. The hospital component of the chronic care model (CCM) was the theoretical framework used in this study because the model was designed to assist heath care organizations in improving chronic care outcomes. Researchers have indicated that the Hospital to Home Initiative (H2H), a national quality improvement campaign launched in 2009, is effective in reducing RSSRs. However, very little research has been conducted to determine which specific H2H strategies and categories of strategies are associated with reducing heart failure RSRRs in Georgia. The purpose of this nonexperimental, cross-sectional quantitative research study was to address this gap. The H2H Survey used in this study is a valid instrument that was previously used in a national study. Surveys were sent to 35 hospitals in Georgia participating in the H2H. A series of one-way ANOVAs were used to test the hypotheses. Key findings were as follows: (a) heart failure RSRRs were reduced when hospitals implemented the H2H, (b) the number of implemented H2H strategies was associated with a reduction in heart failure RSRRs, and (c) categories of strategies were associated with a reduction in heart failure RSRRs. These findings can be used for promoting positive social change because hospital administrators can implement changes using effective strategies to reduce both heart failure RSRRs and government penalties associated with these readmissions.

<p>  </p> <p><strong>Background: </strong></p> <p>Torsades de pointes (TdP) is a polymorphic ventricular tachycardia (VT) associated with heart rate-corrected QT interval (QTc) prolongation on the electrocardiogram (ECG). TdP can cause sudden cardiac arrest (SCA), a catastrophic outcome. The antiarrhythmic drugs dofetilide and sotalol can cause QTc prolongation and arrhythmias, as can more than 200 other medications available on global markets. Heart failure (HF) with reduced ejection fraction (HFrEF) is a risk factor for drug-induced TdP, and HFrEF heightens sensitivity to drug-induced QTc lengthening. However, ~55% of patients with HF have preserved, rather than reduced, ejection fraction. It remains unknown whether patients with HF with preserved ejection fraction (HFpEF) are at increased risk for drug-induced VT/SCA. Assessment of the risk of drug-induced VT/SCA in HFpEF patients is important, so that recommendations can be made regarding the safety of QTc-prolonging drugs and need for enhanced ECG monitoring in this population. </p> <p><strong>Objective:</strong></p> <p>In aim 1, we sought to determine the risk of VT and SCA associated with dofetilide and sotalol in patients with HFpEF. In aim 2, we were able to use QTc interval to determine the odds of dofetilide/sotalol-associated QT interval prolongation in patients with HFpEF. In Aim 3, we investigated the influence of HFpEF on VT and SCA associated with a broader group of drugs known to cause TdP (“known “TdP drugs”), as designated by the QT drugs list at www.crediblemeds.org. </p> <p><strong>Methods:</strong></p> <p>In aim 1, we used Medicare claims (2014-2016) and ICD-9/10 codes to identify patients taking the QT interval-prolonging drugs dofetilide or sotalol, which are used commonly in patients with HF and atrial fibrillation, as well as non-dofetilide or sotalol users among 3 groups: HFpEF, HFrEF, and no HF. Multinomial propensity score-matching was performed. Cochran–Mantel–Haenszel statistics and standardized differences were used to compare baseline characteristics. A generalized Cox proportional hazards model was used to estimate hazard ratios (HRs) and test the association of VT and SCA among dofetilide/sotalol users, HFpEF, HFrEF, and no HF.</p> <p>In Aim 2, the data source was electronic health records from the Indiana Network for Patient Care (February 2010 to May 2021). After removing patients with overlapping diagnoses of HFpEF and HFrEF, no diagnosis code, absence of QT interval records, and no validated record of using dofetilide or sotalol, we identified patients taking dofetilide or sotalol among three groups: HFrEF, HFpEF, and no HF. Cochran–Mantel–Haenszel statistics were used to compare baseline characteristics. QT interval prolongation was defined as heart rate-corrected QT (QTc) > 500 ms during dofetilide/sotalol therapy. Unadjusted odds ratios (OR) of QT interval prolongation were determined by univariate analysis, and adjusted ORs were determined by generalized estimating equations (GEE) with logit link to account for an individual cluster with different times of hospitalization and covariates.</p> <p>In aim 3, we used Medicare enrollment in fee-for-service medical and pharmacy benefits (2014 to 2016) and ICD-9/10 codes, we identified patients taking drugs known to cause torsades de pointes (TdP drugs; www.crediblemeds.org) and non-TdP drug users among three groups: HFrEF, HFpEF, and no HF. Multinomial propensity score-matching was performed to minimize baseline differences in covariates (patient demographics, comorbidities, health care utilization and drug history). Cochran–Mantel–Haenszel statistics and standardized differences were used to compare baseline characteristics. A generalized Cox proportional hazards model was used to estimate HRs and test the association of VT and SCA among TdP drug users with HFpEF, HFrEF, and no HF.</p> <p><strong>Results:</strong></p> <p>In Aim 1, VT and SCA occurred in 166 (10.68%) and 16 (1.03%), respectively, of 1,554 dofetilide/sotalol users with HFpEF, 543 (38.76%) and 40 (2.86%) of 1,401 dofetilide/sotalol users with HFrEF, and 245 (5.06%) and 13 (0.27%) of 4,839 dofetilide/sotalol users with no HF. The adjusted HR for VT in patients with HFrEF was 7.00 (95% CI 6.12-8.02) and in patients with HFpEF was 1.99 (1.71-2.32). The risk of VT associated with dofetilide/sotalol was increased across the overall study population (HR: 2.47 [1.89-3.23]). Use of dofetilide/sotalol increased the risk of VT in patients with HFrEF (HR: 1.53 [1.07-2.20]) and in those with HFpEF (HR: 2.34 [1.11-4.95]). However, while the overall risk of SCA was increased in patients with HFrEF (HR: 5.19 [4.10-6.57]) and HFpEF (HR: 2.53 [1.98-3.23]) compared to patients with no HF, dofetilide/sotalol use was not significantly associated with an increased risk of SCA.</p> <p>In Aim 2, QTc prolongation associated with dofetilide/sotalol occurred in 51.2% of patients with HFpEF, 70.1% of patients with HFrEF, and 29.4% of patients with no HF. After adjusting for age, sex, race, serum potassium and magnesium concentrations, kidney function, concomitant drug therapy, and comorbid conditions, the adjusted odds of having QTc interval larger than 500ms during the hospital stay were 5.23 [3.15-8.67] for HFrEF and 1.98 [1.17-3.33] for HFpEF with no HF as the reference group. </p> <p>In Aim 3, of 23,910 known TdP drug users with HFrEF, VT and SCA occurred in 4,263 (17.8%) and 493 (2.1%) patients, respectively. In comparison, among 31,359 known TdP drug users with HFpEF, VT and SCA occurred in 1,570 (5.0%) and 340 (1.1%) patients. VT and SCA occurred in 3,154 (0.8%) and 528 (0.1%) of 384,824 known TdP drug users without HF. The overall HR of both VT and SCA was increased in patients with HFrEF (HR: 7.18 [6.13-8.40])  and in those with HFpEF (HR: 2.09 [1.80-2.42]). The risk of VT associated with known TdP drugs was increased across the overall population (HR: 1.34 [1.20-1.51]). Use of known TdP drugs significantly increased the risk of VT and SCA in patients with HFrEF (HR: 1.34 [1.07-1.67]), but not in patients with HFpEF.</p> <p><strong>Conclusion:</strong></p> <p>HFpEF may exhibit an enhanced response to drug-associated VT, and is associated with a higher risk of drug-associated QTc interval prolongation. Further study is needed to identify methods to minimize this risk for patients with HFpEF requiring therapy with dofetilide, sotalol, or drugs known to cause TdP. </p>