The aim of this paper is to investigate different machine learning based forecasting techniques for forecasting of blood pressure and heart rate. Forecasting of blood pressure could potentially help a clinician to take preventative steps even before dangerous medical situations occur. This paper examines forecasting blood pressure 30 minutes in advance. Univariate and multivariate forecast models are considered. Different forecast strategies are also considered. To compare different forecast strategies, LSTM and BI-LSTM machine learning algorithms were included. Then univariate and multivariate LSTM, BI-LSTM and CNN machine learning algorithms were compared using the two best forecasting strategies. Comparative analysis between forecasting strategies suggest that MIMO and DIRMO forecast strategies provide the best accuracy in forecasting physiological time series data. Results also appear to show that multivariate forecast models for blood pressure and heart rate are more reliable compared to blood pressure alone. Comparative analysis between MIMO and DIRMO forecasting strategies appear to show that DIRMO is more reliable for both univariate and multivariate cases. Results also appear to show that the forecast model that uses BI-LSTM with the DIRMO strategy is the best overall.
Abstract-This paper presents a high speed configurable FPGAbased wideband channel sounder with signal bandwidths up to 200 MHz and results of a study of dynamic urban picocell channel. The use of FPGA allows the sounder to be adaptable for measurements in different scenarios. Adaptable options include changes to the waveform, bandwidth, channel sampling rate and real-time averaging to improve signal-to-noise ratio in weak signal conditions. The implemented architecture has led to a 60% reduction in size and weight compared to sounders in use elsewhere making it ideal for mobile channel measurements. The study of an urban picocell channel has shown that dynamic variation due to automotive traffic introduces average signal fades of up to 5 dB but causes frequency selective fading with depths of up to 40 dB. Existing channel models assume antenna heights of more than 6 m and path lengths of more than 30 m. Therefore there is a need for shorter path models and this paper proposes a linear picocell channel model for static and dynamic urban environment.
Intensive care units (ICUs) are busy and noisy areas where patients and professional staff can be exposed to acoustic noise for long periods of time. In many cases, noise levels significantly exceed the levels recommended by the official health organisations. This situation can affect not only patient recovery but also professional staff, making ICUs unhealthy work and treatment environments. To introduce the measures and reduce the acoustic noise in the ICU, acoustic noise levels should first be measured and then appropriately analysed. However, in most studies dealing with this problem, measurements have been performed manually over short periods, leading to limited data being collected. They are usually followed by insufficient analysis, which in turn results in inadequate measures and noise reduction. This paper reviews recent works dealing with the problem of excessively high noise levels in ICUs and proposes a more thorough analysis of measured data both in the time and frequency domains. Applied frequency domain analysis identifies the cyclic behaviour of the measured sound pressure levels (SPLs) and detects the dominant frequency components in the SPL time series. Moreover, statistical analyses are produced to depict the patterns and SPLs to which patients in ICUs are typically exposed during their stay in the ICU. It has been shown that the acoustic environment is very similar every night, while it can vary significantly during the day or evening periods. However, during most of the observed time, recorded SPLs were significantly above the prescribed values, indicating an urgent need for their control and reduction. To effectively tackle this problem, more detailed information about the nature of noise during each of the analysed periods of the day is needed. This issue will be addressed in the continuation of this project.
In recent years, Secondary Substations (SSs) are being provided with equipment that allows their full management. This is particularly useful not only for monitoring and planning purposes but also for detecting erroneous measurements, which could negatively affect the performance of the SS. On the other hand, load forecasting is extremely important since they help electricity companies to make crucial decisions regarding purchasing and generating electric power, load switching, and infrastructure development. In this regard, Short Term Load Forecasting (STLF) allows the electric power load to be predicted over an interval ranging from one hour to one week. However, important issues concerning error detection by employing STLF has not been specifically addressed until now. This paper proposes a novel STLF-based approach to the detection of gain and offset errors introduced by the measurement equipment. The implemented system has been tested against real power load data provided by electricity suppliers. Different gain and offset error levels are successfully detected.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.