Background: Blood pressure (BP) measurements have been used widely in clinical and private environments. Recently, the use of ECG monitors has proliferated; however, they are not enabled with BP estimation. We have developed a method for BP estimation using only electrocardiogram (ECG) signals. Methods: Raw ECG data are filtered and segmented, and, following this, a complexity analysis is performed for feature extraction. Then, a machine-learning method is applied, combining a stacking-based classification module and a regression module for building systolic BP (SBP), diastolic BP (DBP), and mean arterial pressure (MAP) predictive models. In addition, the method allows a probability distribution-based calibration to adapt the models to a particular user. Results: Using ECG recordings from 51 different subjects, 3129 30-s ECG segments are constructed, and seven features are extracted. Using a train-validation-test evaluation, the method achieves a mean absolute error (MAE) of 8.64 mmHg for SBP, 18.20 mmHg for DBP, and 13.52 mmHg for the MAP prediction. When models are calibrated, the MAE decreases to 7.72 mmHg for SBP, 9.45 mmHg for DBP and 8.13 mmHg for MAP. Conclusion: The experimental results indicate that, when a probability distribution-based calibration is used, the proposed method can achieve results close to those of a certified medical device for BP estimation.
Diabetes is one of today’s greatest global problems, and it is only becoming bigger. Constant measuring of blood glucose level is a prerequisite for monitoring glucose blood level and establishing diabetes treatment procedures. The usual way of glucose level measuring is by an invasive procedure that requires finger pricking with the lancet and might become painful and obeying, especially if this becomes a daily routine. In this study, we analyze noninvasive glucose measurement approaches and present several classification dimensions according to different criteria: size, invasiveness, analyzed media, sensing properties, applied method, activation type, response delay, measurement duration, and access to results. We set the focus on using machine learning and neural network methods and correlation with heart rate variability and electrocardiogram, as a new research and development trend.
Chronic heart failure represents a global pandemic, currently affecting over 26 million of patients worldwide. It is a major contributor in the death rate of patients with cardiovascular diseases and results in more than 1 million hospitalizations annually in Europe and North America. Methods for chronic heart failure detection can be utilized to act preventive, improve early diagnosis and avoid hospitalizations or even life-threatening situations, thus highly enhance the quality of patient's life. In this paper, we present a machine-learning method for chronic heart failure detection from heart sounds. The method consists of: filtering, segmentation, feature extraction and machine learning. The method was tested with a leave-one-subject-out evaluation technique on data from 122 subjects, gathered in the study. The method achieved 96% accuracy, outperforming a majority classifier for 15 percentage points. More specifically, it detects (recalls) 87% of the chronic heart failure subjects with a precision of 87%. The study confirmed that advanced machine learning applied on real-life sounds recorded with an unobtrusive digital stethoscope can be used for chronic heart failure detection.
Cloud model's ability of sharing hardware resources and services among multiple tenants sets new challenges and issues. Up to now, most of the surveys mostly relate to the security issues in a multitenant cloud. In this paper, we measure the performance for the multitenant cloud model and aim to analyze the CPU utilization for applying the same load to different web services hosted in multitenant environments; and finally compare it to the single-tenant environment. Two parameters will be analyzed: 1) the number of concurrent messages and 2) the message size. Test cases consist of two web services, the first is the memory demand only web service (Concat) and the second is both computation intensive and memory demanding web service (Sort). The conclusions will analyze the results of the performance in terms of CPU utilization in all cloud environments using the same amount of resources, but orchestrated differently. We have set a hypothesis that single-tenant cloud environment achieves smaller CPU utilization than both the multitenant environments we will test. However, the results show that there are particular regions where multitenant environment achieves smaller CPU utilization than a single-tenant environment. Moreover, we have performed additional tests to make better conclusions which multitenant environment achieves smaller utilization in comparison one to another, and showed that at most cases, for both Concat and Sort web services, multitenant (4×1) environment achieves greater utilization than multitenant (2×2) environment, except for Sort web service where multitenant (2×2) achieves smaller utilization than multitenant (4×1) environment when exactly one of the input parameters is small.