Diurnal and seasonal variation of cough episodes in healthy young adults.

SETTING: We developed an algorithm to assess recorded cough episodes and differentiate them from similar, non-cough sounds. OBJECTIVE: To measure cough episodes in healthy young adults, cigarette smokers and non-smokers over a 24-hour recording period, during the course of normal activity. DESIGN: The study subjects were students, aged 20-40 years old. 24-hour sound recordings were conducted by a portable recorder. Analysis used an algorithm that was developed and tested in the study. RESULTS: Seventy students were recruited. Recordings included 2628 cough episodes in 1704 h of recording. The cough detection algorithm successfully detected 88.5% of recorded cough episodes and 95.6% of non-cough sounds. There was a clear tendency for more coughs among smokers. Autumn was the season with the highest mean cough episodes per day (58.65), while summer had the lowest (14.19). There was a strong correlation between self-reported cough episodes and recorded coughs. Cough episodes were significantly more frequent between noon and midnight (p < 0.0001). CONCLUSION: There is a very large range in daily coughs among healthy young adults. During sleeping hours there are less cough episodes. In autumn and spring there are more cough episodes compared to summer and winter, probably secondary to environmental factors. In smokers, the coughing rate is relatively high. If the cough detection device will be able to discriminate between cough variants (i.e., healthy versus patient), and stringent validation will confirm sensitivity and specificity, valuable data from this device may ease the decision regarding medications, or any other changes in order to improve outcome.

Dynamic Spatial Predicted Background.

We present a novel method for online background modeling for static video cameras - Dynamic Spatial Predicted Background (DSPB). Our unique method employs a small subset of image pixels to predict the whole scene by exploiting pixel correlations (distant and close). DSPB acts as a hybrid model combining successful elements taken from two major approaches: local-adaptive that propose to fit a distribution pixelwise, and global-linear that reconstruct the background by finding a lowrank version of the scene. To our knowledge, this is the first attempt to combine these approaches in a unified system. DSPB models the scene as a superposition of illumination effects and predicts each pixel's value by a linear estimator comprised of only 5 pixels of the scene and can initialize the background starting from the 5th frame. By doing so, we keep the computational load low, allowing our method to be used in many real-time applications using simple hardware. The suggested prediction model of scene appearance is novel, and the scheme is very accurate and efficient computationally. We show the method merits on an application for video FG-BG separation, and how some of the main existing approaches may be challenged and how their drawbacks are less dominant in our model. Experimental results validate our findings, by computation speed and mean F-measure values on several public datasets. We also examine how results may improve by analyzing each video individually according to its content. DSPB can be successfully incorporated in other image processing tasks like change detection, video compression and video inpainting.