Wearable Bluetooth Triage Healthcare Monitoring System

Triage is the first interaction between a patient and a nurse/paramedic. This assessment, usually performed at Emergency departments, is a highly dynamic process and there are international grading systems that according to the patient condition initiate the patient journey. Triage requires an initial rapid assessment followed by routine checks of the patients’ vitals, including respiratory rate, temperature, and pulse rate. Ideally, these checks should be performed continuously and remotely to reduce the workload on triage nurses; optimizing tools and monitoring systems can be introduced and include a wearable patient monitoring system that is not at the expense of the patient’s comfort and can be remotely monitored through wireless connectivity. In this study, we assessed the suitability of a small ceramic piezoelectric disk submerged in a skin-safe silicone dome that enhances contact with skin, to detect wirelessly both respiration and cardiac events at several positions on the human body. For the purposes of this evaluation, we fitted the sensor with a respiratory belt as well as a single lead ECG, all acquired simultaneously. To complete Triage parameter collection, we also included a medical-grade contact thermometer. Performances of cardiac and respiratory events detection were assessed. The instantaneous heart and respiratory rates provided by the proposed sensor, the ECG and the respiratory belt were compared via statistical analyses. In all considered sensor positions, very high performances were achieved for the detection of both cardiac and respiratory events, except for the wrist, which provided lower performances for respiratory rates. These promising yet preliminary results suggest the proposed wireless sensor could be used as a wearable, hands-free monitoring device for triage assessment within emergency departments. Further tests are foreseen to assess sensor performances in real operating environments.

Calibration of Spatial Rain Scanner using Rainfall Depth of Rain Gauges

A spatial rain scanner has been developed based on a marine radar to satisfy the demand for spatial rain information for hydrological applications. Since the coverage of the rain scanner is 44 km in radius, it is necessary to expand the coverage by installing it in two sites that intersect each other performing a radar network. For this purpose, the first rain scanner has been installed at the Center for Atmospheric Science and Technology (PSTA) in Bandung and the second one at the Space and Atmospheric Observation Center (BPAA) Tanjungsari in Sumedang. This paper focuses on the calibration of radar observations with rainfall data from 7 rain gauges installed in Bandung area and its surroundings. The calibration method calculates rainfall depth (three parameters) instead of only the intensity of rainfall. The data period used for this research is from March to November 2020. The rain scanners have better rainfall events detection over basin area, such as Dayeuh Kolot and Cidurian, than over highland area, such as Lembang. Two calibration methods are used, and the results show that the calibration by calculating three parameters (accumulated reflectivity, duration, and intensity) in the linear model is able to measure rainfall estimation better than using a linear model with one parameter (accumulated reflectivity) for rainfall depth more than 10 mm. Rainfall estimation calculation using scheme 1 tends to underestimate while scheme 2 tends to overestimate.

Evaluating Temporal Bias in Time Series Event Detection Methods

The detection of events in time series is an important task in several areas of knowledge where operations monitoring is essential. Experts often have to deal with choosing the most appropriate event detection method for a time series, which can be a complex task. There is a demand for benchmarking different methods in order to guide this choice. For this, standard classification accuracy metrics are usually adopted. However, they are insufficient for a qualitative analysis of the tendency of a method to precede or delay event detections. Such analysis is interesting for applications in which tolerance for "close" detections is important rather than focusing only on accurate ones. In this context, this paper proposes a more comprehensive event detection benchmark process, including an analysis of temporal bias of detection methods. For that, metrics based on the time distance between event detections and identified events (detection delay) are adopted. Computational experiments were conducted using real-world and synthetic datasets from Yahoo Labs and resources from the Harbinger framework for event detection. Adopting the proposed detection delay-based metrics helped obtain a complete overview of the performance and general behavior of detection methods.