Prognosis Analysis of Heart Failure Based on Recurrent Attention Model

ObjectivesHeart failure is a group of complex clinical syndromes that lead to ventricular filling or impaired ejection ability due to abnormal heart structure or function. Difficult treatment, poor prognosis and high mortality are the main characteristics of heart failure. According to admission data and past medical use, the 30-day mortality rate of patients with heart failure was obtained and the main characteristics affecting the 30-day mortality of patients with heart failure were determined.Material and methodsBased on the data of April 2016 to July 2018 of Shanxi Acadeny of Medical Sciences, and we chose 4,682 information on heart failure patients, of which 539 died in the hospital by screening. We built a 30-day mortality prediction model for patients with heart failure. The model can fuse clinical data and text data through multiple kernel learning, and input the fused data into the recurrent attention model. It can not only predict the 30-day mortality of patients with heart failure, but also the influencing factors of prognosis of patients with heart failure were also obtained.ResultsThe prediction accuracy of the recurrent attention network is obviously higher than that of other machine learning models, and the accuracy rate reaches 93.4%. The AUC value of the area under the ROC curve of the model reaches 87%, which is obviously higher than that of the traditional machine learning models such as decision tree, naive Bayesian and support vector machine. In addition, the model can also reach a conclusion that New York heart function classification, age, NT—ProBNP, LVEF, β-blockers, ventricular arrhythmia, high blood pressure, coronary heart disease (CHD) and bronchitis were independent risk factors for death. And patients with revascularization, ACEI/ARB drugs, β-blockers, spironolactone have a better prognosis than non-users. This provides an important reference for doctors to better treat and manage patients with heart failure.ConclusionExperiments show that the prognostic effect of the recurrent attention model is significantly higher than that of other traditional machine learning models. Because the model increases the attention mechanism, the important features affecting the prognostic results are obtained, which enables doctors to prescribe drugs according to the symptoms, take timely precautions and help patients to treat in time.     

An Approach Based on Mutually Informed Neural Networks to Optimize the Generalization Capabilities of Decision Support Systems Developed for Heart Failure Prediction

Available clinical methods for heart failure (HF) diagnosis are expensive and require a high-level of experts intervention. Recently, various machine learning models have been developed for the prediction of HF where most of them have an issue of over-fitting. Over-fitting occurs when machine learning based predictive models show better performance on the training data yet demonstrate a poor performance on the testing data and the other way around. Developing a machine learning model which is able to produce generalization capabilities (such that the model exhibits better performance on both the training and the testing data sets) could overall minimize the prediction errors. Hence, such prediction models could potentially be helpful to cardiologists for the effective diagnose of HF. This paper proposes a two-stage decision support system to overcome the over-fitting issue and to optimize the generalization factor. The first stage uses a mutual information based statistical model while the second stage uses a neural network. We applied our approach to the HF subset of publicly available Cleveland heart disease database. Our experimental results show that the proposed decision support system has optimized the generalization capabilities and has reduced the mean percent error (MPE) to 8.8% which is significantly less than the recently published studies. In addition, our model exhibits a 93.33% accuracy rate which is higher than twenty eight recently developed HF risk prediction models that achieved accuracy in the range of 57.85% to 92.31%. We can hope that our decision support system will be helpful to cardiologists if deployed in clinical setup.     

An exploration of the genetic and environmental etiology of heart rate in infancy and middle childhood.

Heart rate was recorded on 210 MZ and 174 DZ same sex twin pairs participating in the MacArthur Longitudinal Twin Study (MALTS) at age 14, 20, 24, 36 months and 7 years. Heart rate was monitored in the laboratory at all ages. At ages 14 to 36 months, heart rate was monitored prior to a set of cognitive tasks. At age 7 years heart rate was recorded during a mood-eliciting videotaped presentation. At this age only heart rate monitored during neutral portions of the presentation were used. Mean heart rate declines substantially across this age range, but is similar in boys and girls and for MZ and DZ twins at each age. Heart rate is moderately correlated across all time points suggesting that individual differences in heart rate are relatively stable over this age range. Multivariate genetic and environmental models were fitted to the raw data. In general, genetic factors contribute to the stability of individual differences over time. Shared and non-shared environment factors tended to be occasion specific, with non-shared environment contributing substantially to the individual variation at each age. Shared environment and non-shared environment also contributed a modicum to the stability across time. Thus, individual differences in resting heart rate is a relatively stable, heritable trait from infancy to early childhood.     

Model of the sino-atrial and atrio-ventricular nodes of the conduction system of the human heart.

The coupled sino-atrial and atrio-ventricular nodes of the heart are discussed using a dedicated non-linear oscillator model. Several modes by which the oscillations cease in the system are obtained (asystole models). The oscillations of the model are compared with heart rate variability in heart block patients.     

Theoretical and empirical derivation of cardiovascular allometric relationships in children.

Basic fluid dynamic principles were used to derive a theoretical model of optimum cardiovascular allometry, the relationship between somatic and cardiovascular growth. The validity of the predicted models was then tested against the size of 22 cardiovascular structures measured echocardiographically in 496 normal children aged 1 day to 20 yr, including valves, pulmonary arteries, aorta and aortic branches, pulmonary veins, and left ventricular volume. Body surface area (BSA) was found to be a more important determinant of the size of each of the cardiovascular structures than age, height, or weight alone. The observed vascular and valvar dimensions were in agreement with values predicted from the theoretical models. Vascular and valve diameters related linearly to the square root of BSA, whereas valve and vascular areas related to BSA. The relationship between left ventricular volume and body size fit a complex model predicted by the nonlinear decrease of heart rate with growth. Overall, the relationship between cardiac output and body size is the fundamental driving factor in cardiovascular allometry.     

Deep learning-based multi-task prediction system for plant disease and species detection

The manual prediction of plant species and plant diseases is expensive, time-consuming, and requires expertise that is not always available. Automated approaches, including machine learning and deep learning, are increasingly being applied to surmount these challenges. For this, accurate models are needed to provide reliable predictions and guide the decision-making process. So far, these two problems have been addressed separately, and likewise, separate models have been developed for each of these two problems, but considering that plant species and plant disease prediction are often related tasks, they can be considered together. We therefore propose and validate a novel approach based on the multi-task learning strategy, using shared representations between these related tasks, because they perform better than individual models. We apply a multi-input network that uses raw images and transferred deep features extracted from a pre-trained deep model to predict each plant's type and disease. We develop an end-to-end multi-task model that carries out more than one learning task at a time and combines the Convolutional Neural Network (CNN) features and transferred features. We then evaluate this model using public datasets. The results of our experiments demonstrated that this Multi-Input Multi-Task Neural Network model increases efficiency and yields faster learning for similar detection tasks.     

大鼠腹腔异位心脏移植模型建立及改进

目的:建立一种快速有效的大鼠腹腔异位心脏移植模型。方法:采用SD大鼠作为受体,Wistar大鼠作为供体,行同种异位腹腔心脏移植,术后给以CsA5 mg/kg/d灌服,心脏移植手术方法采用改良的Ono术式,观察改良的腹腔异位心脏移植各步骤所需时间、术后成功率及主要并发症发生率。结果:共建立40只大鼠异位心脏移植模型,手术成功率92.5%。动脉吻合时间12.5±2.3min,静脉吻合时间12.3±1.5 min,供心缺血时间37±3.5 min,受体血管阻断时间34.2±2.6 min,总手术时间90.2±4.8 min,出现的主要并发症为出血和供心复跳失败(各占5%、2.5%)。结论:改进的大鼠腹部异位心脏移植技术是一种简便、快速、有效、成功率高的模型制作方法。     

A generating model of realistic synthetic heart sounds for performance assessment of phonocardiogram processing algorithms

A new model which is capable of generating realistic synthetic phonocardiogram (PCG) signals is introduced based on three coupled ordinary differential equations. The new PCG model takes into account the respiratory frequency, the heart rate variability and the time splitting of first and second heart sounds. This time splitting occurs with each cardiac cycle and varies with inhalation and exhalation. Clinical PCG statistics and the close temporal relationship between events in ECG and PCG are used to deduce values of PCG model parameters.In comparison with published PCG models, the proposed model allows a larger number of known PCG features to be taken into consideration. Moreover it is able to generate both normal and abnormal realistic synthetic heart sounds. Results show that these synthetic PCG signals have the closest features to those of a conventional heart sound in both time and frequency domains. Additionally, a sound quality test carried out by eight cardiologists demonstrates that the proposed model outperforms the existing models.This new PCG model is promising and useful in assessing signal processing techniques which are developed to help clinical diagnosis based on PCG.     

Type 5 adenylyl cyclase plays a major role in stabilizing heart rate in response to microgravity induced by parabolic flight.

It is well known that autonomic nervous activity is altered under microgravity, leading to disturbed regulation of cardiac function, such as heart rate. Autonomic regulation of the heart is mostly determined by beta-adrenergic receptors/cAMP signal, which is produced by adenylyl cyclase, in cardiac myocytes. To examine a hypothesis that a major cardiac isoform, type 5 adenylyl cyclase (AC5), plays an important role in regulating heart rate during parabolic flights, we used transgenic mouse models with either disrupted (AC5KO) or overexpressed AC5 in the heart (AC5TG) and analyzed heart rate variability. Heart rate had a tendency to decrease gradually in later phases within one parabola in each genotype group, but the magnitude of decrease was smaller in AC5KO than that in the other groups. The inverse of heart rate, i.e., the R-R interval, was much more variable in AC5KO and less variable in AC5TG than that in wild-type controls. The standard deviation of normal R-R intervals, a marker of total autonomic variability, was significantly greater in microgravity phase in each genotype group, but the magnitude of increase was much greater in AC5KO than that in the other groups, suggesting that heart rate regulation became unstable in the absence of AC5. In all, AC5 plays a major role in stabilizing heat rate under microgravity.     

Evaluation of data transformations used with the square root and schoolfield models for predicting bacterial growth rate.

A comparison was made between mathematical variations of the square root and Schoolfield models for predicting growth rate as a function of temperature. The statistical consequences of square root and natural logarithm transformations of growth rate use in several variations of the Schoolfield and square root models were examined. Growth rate variances of Yersinia enterocolitica in brain heart infusion broth increased as a function of temperature. The ability of the two data transformations to correct for the heterogeneity of variance was evaluated. A natural logarithm transformation of growth rate was more effective than a square root transformation at correcting for the heterogeneity of variance. The square root model was more accurate than the Schoolfield model when both models used natural logarithm transformation.     

Evaluation of data transformations used with the square root and schoolfield models for predicting bacterial growth rate.

A comparison was made between mathematical variations of the square root and Schoolfield models for predicting growth rate as a function of temperature. The statistical consequences of square root and natural logarithm transformations of growth rate use in several variations of the Schoolfield and square root models were examined. Growth rate variances of Yersinia enterocolitica in brain heart infusion broth increased as a function of temperature. The ability of the two data transformations to correct for the heterogeneity of variance was evaluated. A natural logarithm transformation of growth rate was more effective than a square root transformation at correcting for the heterogeneity of variance. The square root model was more accurate than the Schoolfield model when both models used natural logarithm transformation.     

The missing link between cardiovascular rhythm control and myocardial cell modeling.

Cardiac arrhythmia is currently investigated from two different points of view. One considers ECG bio-signal analysis and investigates heart rate variability, baroreflex control, heart rate turbulence, alternans phenomena, etc. The other involves building computer models of the heart based on ion channels, bio-domain models and forward calculations to finally reach ECG and body surface potential maps. Both approaches aim to support the cardiologist in better understanding of arrhythmia, improving diagnosis and reliable risk stratification, and optimizing therapy. This article summarizes recent results and aims to trigger new research to bridge the different views.     

Precision medicine in human heart modeling

Precision medicine is a new frontier in healthcare that uses scientific methods to customize medical treatment to the individual genes, anatomy, physiology, and lifestyle of each person. In cardiovascular health, precision medicine has emerged as a promising paradigm to enable cost-effective solutions that improve quality of life and reduce mortality rates. However, the exact role in precision medicine for human heart modeling has not yet been fully explored. Here, we discuss the challenges and opportunities for personalized human heart simulations, from diagnosis to device design, treatment planning, and prognosis. With a view toward personalization, we map out the history of anatomic, physical, and constitutive human heart models throughout the past three decades. We illustrate recent human heart modeling in electrophysiology, cardiac mechanics, and fluid dynamics and highlight clinically relevant applications of these models for drug development, pacing lead failure, heart failure, ventricular assist devices, edge-to-edge repair, and annuloplasty. With a view toward translational medicine, we provide a clinical perspective on virtual imaging trials and a regulatory perspective on medical device innovation. We show that precision medicine in human heart modeling does not necessarily require a fully personalized, high-resolution whole heart model with an entire personalized medical history. Instead, we advocate for creating personalized models out of population-based libraries with geometric, biological, physical, and clinical information by morphing between clinical data and medical histories from cohorts of patients using machine learning. We anticipate that this perspective will shape the path toward introducing human heart simulations into precision medicine with the ultimate goals to facilitate clinical decision making, guide treatment planning, and accelerate device design.

     

Coexistence of Reward and Unsupervised Learning During the Operant Conditioning of Neural Firing Rates

A fundamental goal of neuroscience is to understand how cognitive processes, such as operant conditioning, are performed by the brain. Typical and well studied examples of operant conditioning, in which the firing rates of individual cortical neurons in monkeys are increased using rewards, provide an opportunity for insight into this. Studies of reward-modulated spike-timing-dependent plasticity (RSTDP), and of other models such as R-max, have reproduced this learning behavior, but they have assumed that no unsupervised learning is present (i.e., no learning occurs without, or independent of, rewards). We show that these models cannot elicit firing rate reinforcement while exhibiting both reward learning and ongoing, stable unsupervised learning. To fix this issue, we propose a new RSTDP model of synaptic plasticity based upon the observed effects that dopamine has on long-term potentiation and depression (LTP and LTD). We show, both analytically and through simulations, that our new model can exhibit unsupervised learning and lead to firing rate reinforcement. This requires that the strengthening of LTP by the reward signal is greater than the strengthening of LTD and that the reinforced neuron exhibits irregular firing. We show the robustness of our findings to spike-timing correlations, to the synaptic weight dependence that is assumed, and to changes in the mean reward. We also consider our model in the differential reinforcement of two nearby neurons. Our model aligns more strongly with experimental studies than previous models and makes testable predictions for future experiments.     

A mathematical model of human heart including the effects of heart contractility varying with heart rate changes

Incorporating the intrinsic variability of heart contractility varying with heart rate into the mathematical model of human heart would be useful for addressing the dynamical behaviors of human cardiovascular system, but models with such features were rarely reported. This study focused on the development and evaluation of a mathematical model of the whole heart, including the effects of heart contractility varying with heart rate changes. This model was developed based on a paradigm and model presented by Ottesen and Densielsen, which was used to model ventricular contraction. A piece-wise function together with expressions for time-related parameters were constructed for modeling atrial contraction. Atrial and ventricular parts of the whole heart model were evaluated by comparing with models from literature, and then the whole heart model were assessed through coupling with a simple model of the systemic circulation system and the pulmonary circulation system. The results indicated that both atrial and ventricular parts of the whole heart model could reasonably reflect their contractility varying with heart rate changes, and the whole heart model could exhibit major features of human heart. Results of the parameters variation studies revealed the correlations between the parameters in the whole heart model and performances (including the maximum pressure and the stroke volume) of every chamber. These results would be useful for helping users to adjust parameters in special applications.     

Heart Rate Differences between Targets and Nontargets in Intuitive Tasks

This study reports the results of one experiment and a replication, aimed at investigating heart rate changes related to a purely intuitive task. In each experiment, 12 subjects were required to guess which of four pictures presented in sequence for about 10 s was the target. Each subject performed 20 trials. In each trial, the target was automatically selected using a pseudorandom algorithm. The heart rate was recorded during the picture presentation. In the first experiment, a statistically significant heart rate increment associated with targets with respect to nontargets was observed. The replication experiment with 12 new subjects confirmed the data obtained in the main experiment. These findings support the hypothesis that heart rate is related not only to conscious but also to unconscious cognitive activity such as that involved in intuitive tasks, giving convergent evidence for the models describing human intuitive cognitive activity as a double, partially independent information processing system.     

Is Model Fitting Necessary for Model-Based fMRI?

Model-based analysis of fMRI data is an important tool for investigating the computational role of different brain regions. With this method, theoretical models of behavior can be leveraged to find the brain structures underlying variables from specific algorithms, such as prediction errors in reinforcement learning. One potential weakness with this approach is that models often have free parameters and thus the results of the analysis may depend on how these free parameters are set. In this work we asked whether this hypothetical weakness is a problem in practice. We first developed general closed-form expressions for the relationship between results of fMRI analyses using different regressors, e.g., one corresponding to the true process underlying the measured data and one a model-derived approximation of the true generative regressor. Then, as a specific test case, we examined the sensitivity of model-based fMRI to the learning rate parameter in reinforcement learning, both in theory and in two previously-published datasets. We found that even gross errors in the learning rate lead to only minute changes in the neural results. Our findings thus suggest that precise model fitting is not always necessary for model-based fMRI. They also highlight the difficulty in using fMRI data for arbitrating between different models or model parameters. While these specific results pertain only to the effect of learning rate in simple reinforcement learning models, we provide a template for testing for effects of different parameters in other models.     

Heart rate differences between targets and non targets in intuition tasks

This study reports the results of one experiment and a replication, aimed at investigating heart rate changes related to a pure intuition task. In each experiment, twelve subjects were required to guess which of the four pictures presented in sequence for about 10 seconds, was the target. Each subject performed 20 trials. In each trial the target was automatically selected using a pseudo-random algorithm. Heart rate was recorded (see Method section for details) during the pictures presentation. In the first experiment, a statistical significant increment of heart rate associated to targets with respect non targets was observed. The replication experiment with new twelve subjects confirmed the data obtained in the main experiment. These findings support the hypothesis that heart rate is related not only to overt but also to covert cognitive activity such as that involved in intuition tasks, giving convergent evidence to the models describing our intuitive cognitive activity as a double, partial independent information processing system.     

The effects of the distribution of training on learning feedback-assisted cardiac acceleration

In order to test a hypothesis derived from a motor skills learning model of cardiac acceleration control, groups of subjects were given biofeedback training for four sessions to learn cardiac acceleration under four different training schedules: (1) all sessions in one day, (2) daily sessions, (3) sessions every other day, and (4) weekly sessions. Ability to accelerate heart rate both with and without feedback was determined at each session. Also ability to accelerate heart rate without feedback was determined 1 week after the last training session as a measure of retention. Although there was highly significant (p less than.0001) evidence of heart rate control both with and without feedback, there were no differences in degree of control attributable to distribution of training sessions. There was, however, a trend (p less than .10) for subjects trained under the most distributed training schedule (weekly) to show more retention than subjects trained under a less distributed schedule (daily).