In this paper, a machine learning (ML) approach to estimate blood pressure (BP) using photoplethysmography (PPG) is presented. The final aim of this paper was to develop ML methods for estimating blood pressure (BP) in a non-invasive way that is suitable in a telemedicine health-care monitoring context. The training of regression models useful for estimating systolic blood pressure (SBP) and diastolic blood pressure (DBP) was conducted using new extracted features from PPG signals processed using the Maximal Overlap Discrete Wavelet Transform (MODWT). As a matter of fact, the interest was on the use of the most significant features obtained by the Minimum Redundancy Maximum Relevance (MRMR) selection algorithm to train eXtreme Gradient Boost (XGBoost) and Neural Network (NN) models. This aim was satisfactorily achieved by also comparing it with works in the literature; in fact, it was found that XGBoost models are more accurate than NN models in both systolic and diastolic blood pressure measurements, obtaining a Root Mean Square Error (RMSE) for SBP and DBP, respectively, of 5.67 mmHg and 3.95 mmHg. For SBP measurement, this result is an improvement compared to that reported in the literature. Furthermore, the trained XGBoost regression model fulfills the requirements of the Association for the Advancement of Medical Instrumentation (AAMI) as well as grade A of the British Hypertension Society (BHS) standard.

In this paper, new features relevant to blood pressure (BP) estimation using photoplethysmography (PPG) are presented. A total of 195 features, including the proposed ones and those already known in the literature, have been calculated on a set composed of 50,000 pulses from 1080 different patients. Three feature selection methods, namely Correlation-based Feature Selection (CFS), RReliefF and Minimum Redundancy Maximum Relevance (MRMR), have then been applied to identify the most significant features for BP estimation. Some of these features have been extracted through a novel PPG signal enhancement method based on the use of the Maximal Overlap Discrete Wavelet Transform (MODWT). As a matter of fact, the enhanced signal leads to a reliable identification of the characteristic points of the PPG signal (e.g., systolic, diastolic and dicrotic notch points) by simple means, obtaining results comparable with those from purposely defined algorithms. For systolic points, mean and std of errors computed as the difference between the locations obtained using a purposely defined already known algorithm and those using the MODWT enhancement are, respectively, 0.0097 s and 0.0202 s; for diastolic points they are, respectively, 0.0441 s and 0.0486 s; for dicrotic notch points they are 0.0458 s and 0.0896 s. Hence, this study leads to the selection of several new features from the MODWT enhanced signal on every single pulse extracted from PPG signals, in addition to features already known in the literature. These features can be employed to train machine learning (ML) models useful for estimating systolic blood pressure (SBP) and diastolic blood pressure (DBP) in a non-invasive way, which is suitable for telemedicine health-care monitoring.

Electrocardiogram (ECG) is a simple and fast test that provides the graphical representation of the electrical signal obtained by the heart’s activity and is widely employed to detect functional heart conditions. Indeed, its segmentation and feature extraction are required to produce a diagnosis. This paper focuses on the development of a segmentation technique for QRS complex, T waves and P waves of ECG waveforms that can be useful to detect cardiac diseases. Despite many proposed algorithms have already successfully accomplished the problem of QRS complex ECG segmentation, there is still a lack in the performance of these detectors on P and T waves. The proposed algorithm is based on a threshold calculated using the Otsu’s method to detect R-peaks in the ECG waveform. In addition, it allows to segment all the other ECG waves using wavelet filters and local maxima algorithm. Each segmented wave is defined with onset and offset. This method is evaluated on ECG signals acquired with a single lead heart rate monitor verifying if a wave and its onset and offset are detected or not and using mean time error and IOU (Intersection over Union) as evaluation metrics.Experimental results presented in this paper demonstrate that the proposed algorithm achieves 100 % of R-peak and QRS complex detection for the considered dataset. In addition, also P peak and T peak are detected with a score of 100 %.

The increasing need to monitor Heart Rate (HR) has led to the development of dedicated devices based on miniaturized sensors and to their consequent pervasiveness. In this paper the aim is to develop a DAQ based system supervised by a LabVIEW Virtual Instrument to estimate the HR by means of an accelerometric plethysmograph. Two different signal processing techniques have been developed to estimate the HR respectively in time- and frequency domain, to study and compare their performances and to monitor the patient's heartbeat at rest by identifying possible pathological conditions or diseases.

Nowadays telemedicine has become increasingly pervasive with the use of wearable sensors and miniaturized devices to monitor vital signs in patients at home and in non-structured places. Various telemedicine system architectures for remote monitoring of vital parameters have been proposed, also given the increase in population aging. In this paper the implementation of a web-based telemedicine solution with an Android app for a medical device, Checkme Pro, is presented. The proposed solution permits to download data to the developed Android app exploiting the Bluetooth connectivity of the medical device, that is a certified multi-parametric vital sign monitor, and to manage their presentation to users. The Android app permits to view data and measurements and store them in the database, while the developed web app, in addition to data visualization, permits to visualize graphic trends of different vital signs by querying the database. The functionality that allows visualization of the time trends of vital parameters is of enormous interest for telemedicine applications ensuring greater patient safety and a better understanding of the patient's health situation. Then a validation of the telemedicine system has been done to test the correctness and consistency of the data stored in the Android app, saved in the database, and displayed in both apps.