Filippo Pietroni

A novel approach for features extraction in physiological signals

The authors have investigated a novel processing technique, which allows to measure possibly relevant features in the ECG (Electrocardiogram) signal according to the morphology of its waveform. The aim of this work is to prove its efficacy in the assessment of the subjects Heart Rate (HR) and to broaden its use to signals coming from different biomedical sensors (based on optical, acoustical and mechanical principles) for the computation of HR. The analysis technique proposed for the identification of the main feature (R-peak) in ECG signal provides results that are comparable to those obtained with traditional approaches. The approach has also been applied to other signals related to blood flow, such as PCG (Phonocardiography), PPG (Photoplethysmography) and VCG (Vibrocardiography), where standard algorithms (i.e. Pan & Tompkins) could not be widely applied. HR results from a measurement campaign on 8 healthy subjects have shown, respect to ECG, a deviation (calculated as 2σ) of ±3.3 bpm, ±2.3 bpm and ±1.5 bpm for PCG, PPG and VCG. Future work will involve the extraction of additional features from the previous signals, with the aim of a deeper characterization of them to better describe the subjects health status.

A Low-Cost Sensor for Real-Time Monitoring of Indoor Thermal Comfort for Ambient Assisted Living

The present paper illustrates an innovative low cost solution for the monitoring of indoor thermal comfort by means of Predictive Mean Vote (PMV) index for multiple positions. This is particularly interesting in an Ambient Assisted Living environment as replacement of typical thermostat used for the climate control. In fact, the system proposed considers also personal parameters, as metabolic rate (M) and clothing level (\(I_{cl}\)), instead of the merely environmental parameters. If this is important for normal living conditions, it becomes crucial in case of elderly people and long-term care patients where a reduction of M or \(I_{cl}\) causes a high sensitivity to thermal conditions (especially for cold sensation), or where the disability does not allow the subject re-action (e.g. shading opening/closing when solar radiation occurs). The device proposed uses a set of low-cost non-contact sensors to determine, based on algorithms provided by ISO 7726 and 7730, Mean Radiant Temperature (MRT) and PMV, which are provided as output of the device through wireless or wired connection. The capability of predicting thermal comfort conditions for multiple positions of the occupant in the room has been tested and validated in a real case study: it resulted in a discrepancy of \(\pm 0.5\,^{\circ }\mathrm{C}\) in the MRT measurement and \(\pm \)0.1 for the PMV with respect to a reference measurement system (microclimate station). The sensitivity to the metabolic rate and clothing level for AAL applications is also discussed together with a procedure for an estimation of these parameters. The accuracy achieved allows a better measurement of the real thermal sensation for a more comfortable environment with lower energy consumption.

Integration of Real-Time Metabolic Rate Measurement in a Low-Cost Tool for the Thermal Comfort Monitoring in AAL Environments

The work presented illustrates a methodology to integrate the continuous estimation of metabolic rate in a monitoring tool for the indoor thermal comfort in AAL environments. The monitoring tool adopts an infrared (IR) sensor to retrieve indoor temperatures and evaluate the mean radiant temperature for multiple positions in the space. Other sensors embedded in the central unit allow the estimation of the PMV (Predicted Mean Vote) index. Beyond the ambient quantities, an accurate estimation of the personal parameters (clothing insulation and metabolic rate) allows a reliable assessment of the indoor thermal conditions. According to standards, heart rate measurement can provide an accurate estimation of metabolic rate, but the need of measuring it continuously made this method not applicable in real scenarios. However, in Ambient Assisting Living applications it is easy to monitor vital signs from the existing equipment, e.g. wearable sensors. Therefore, this paper presents the results of the integration of low-cost heart rate sensors in a tool for the monitoring of thermal comfort. The solution turned out to have an uncertainty for the metabolic rate of ±7 % of the reading in a range from 0.7 to 3.4 m, considering that the sensor used has a discrepancy of ±1.3 bpm with respect to a reference measurement system. An accuracy of ±0.05 in the PMV computation was found as result of the uncertainty in the estimation of M.

Implementation of an “at-home” e-Health system using heterogeneous devices

Systems enabling long-term monitoring of physiological data and everyday activities has been the subject of considerable research efforts in the last years, in order to improve the quality of life of patients, elderly people and common citizens at home, out of the hospitalization. With the availability of inexpensive, low power, wireless and integrated devices, current smart homes are typically equipped with a large amount of sensors, which collaboratively process and make deductions from the acquired data on the state of the home, as well as the activities and behaviors of its residents. According to the field of application and the end-users involved (healthy people, elderly, people with disabilities), the definition of the parameters (e.g. heart rate, blood pressure, activity, body mass, etc.) and the appropriate sensors (electrocardiogram, sphygmomanometer, glucometer, etc.) for their acquisition assume a fundamental role. One of the goals of the Italian project Health@Home is to create a network of health sensors and home automation devices to monitor the users status within the home environment. We present a candidate implementation of such a system, describing the software architecture and the selected components, and a testbed of the architecture, realized in a lab room and used for a preliminary experimental study involving seven users.

A Smart Sensing Architecture for Domestic Monitoring: Methodological Approach and Experimental Validation

Smart homes play a strategic role for improving life quality of people, enabling to monitor people at home with numerous intelligent devices. Sensors can be installed to provide a continuous assistance without limiting the resident’s daily routine, giving her/him greater comfort, well-being and safety. This paper is based on the development of domestic technological solutions to improve the life quality of citizens and monitor the users and the domestic environment, based on features extracted from the collected data. The proposed smart sensing architecture is based on an integrated sensor network to monitor the user and the environment to derive information about the user’s behavior and her/his health status. The proposed platform includes biomedical, wearable, and unobtrusive sensors for monitoring user’s physiological parameters and home automation sensors to obtain information about her/his environment. The sensor network stores the heterogeneous data both locally and remotely in Cloud, where machine learning algorithms and data mining strategies are used for user behavior identification, classification of user health conditions, classification of the smart home profile, and data analytics to implement services for the community. The proposed solution has been experimentally tested in a pilot study based on the development of both sensors and services for elderly users at home.