BackgroundOut-of-hospital cardiac arrest is a life threatening situation where the first person performing cardiopulmonary resuscitation (CPR) most often is a bystander without medical training. Some existing smartphone apps can call the emergency number and provide for example global positioning system (GPS) location like Hjelp 113-GPS App by the Norwegian air ambulance. We propose to extend functionality of such apps by using the built in camera in a smartphone to capture video of the CPR performed, primarily to estimate the duration and rate of the chest compression executed, if any.MethodsAll calculations are done in real time, and both the caller and the dispatcher will receive the compression rate feedback when detected. The proposed algorithm is based on finding a dynamic region of interest in the video frames, and thereafter evaluating the power spectral density by computing the fast fourier transform over sliding windows. The power of the dominating frequencies is compared to the power of the frequency area of interest. The system is tested on different persons, male and female, in different scenarios addressing target compression rates, background disturbances, compression with mouth-to-mouth ventilation, various background illuminations and phone placements. All tests were done on a recording Laerdal manikin, providing true compression rates for comparison.ResultsOverall, the algorithm is seen to be promising, and it manages a number of disturbances and light situations. For target rates at 110 cpm, as recommended during CPR, the mean error in compression rate (Standard dev. over tests in parentheses) is 3.6 (0.8) for short hair bystanders, and 8.7 (6.0) including medium and long haired bystanders.ConclusionsThe presented method shows that it is feasible to detect the compression rate of chest compressions performed by a bystander by placing the smartphone close to the patient, and using the built-in camera combined with a video processing algorithm performed real-time on the device.

A system for utilizing an artificial neural network to predict splice sites in genes has been studied. The neural network uses a sliding window of nucleotides over a gene and predicts possible splice sites. Based on the neural network output, the exact location of the splice site is found using a curve fitting of a parabolic function. The splice site location is predicted without prior knowledge of any sensor signals, like 'GT' or 'GC' for the donor splice sites, or 'AG' for the acceptor splice sites. The neural network has been trained using backpropagation on a set of 16965 genes of the model plant Arabidopsis thaliana. The performance is then measured using a completely distinct gene set of 5000 genes, and verified at a set of 20 genes. The best measured performance on the verification data set of 20 genes, gives a sensitivity of 0.891, a specificity of 0.816 and a correlation coefficient of 0.552.

The use of overcomplete dictionaries, or frames, has received increased attention in low bit rate compression. Several vector selection algorithms, such as Matching Pursuit, Orthogonal Matching Pursuit and FOCUSS have been developed to get sparse representations of signals. In these algorithms, continuous.valued coeflcients are found and subsequently quantized. The latter part can cause unwanted effects on the quality of the reconstructed signal. We propose an algorithm that merges the selection and quantization procedures by using 0-1 integer programming. The object is to minimize the distortion measured by the 11 -norm, subject to a bound on the number of "ones" in a binary representation of the frame coefJicients. This bound is an indirect measure of the bit rate. Our new algorithm finds the globally optimal solution based on the above-mentioned criteria.

No abstract

No abstract