Analytics and Applications of Audio and Image Sensing Techniques.

Nowadays, with numerous sensors placed everywhere around us, we can obtain signals collected from a variety of environment-based sensors, including the ones placed on the ground, cased in the air or water, etc [...].

The Dependence of Flue Pipe Airflow Parameters on the Proximity of an Obstacle to the Pipe's Mouth.

This paper describes the influence of the presence of an obstacle near the flue pipe's mouth on the air jet, which directly affects the parameters of the sound generated by the flue pipe. Labial pipes of the most common types of mouth were tested. The method of interval calculus was used instead of invasive measuring instruments. The obtained results prove that the proximity of an obstacle affects the sound's fundamental frequency, as the airflow speed coming out of the flue pipe's mouth changes. The relationship between the airflow speed, the value of the Reynolds number, and the Strouhal number was also established. The thesis of the influence of the proximity of an obstacle on the fundamental frequency of the sound of a flue pipe was generalized, and formulas for calculating the untuning of the sound of the pipe were presented for various types of mouth.

Prototyping Mobile Storytelling Applications for People with Aphasia.

Aphasia is a partial or total loss of the ability to articulate ideas or comprehend spoken language, resulting from brain damage, in a person whose language skills were previously normal. Our goal was to find out how a storytelling app can help people with aphasia to communicate and share daily experiences. For this purpose, the Aphasia Create app was created for tablets, along with Aphastory for the Google Glass device. These applications facilitate social participation and enhance quality of life by using visual storytelling forms composed of photos, drawings, icons, etc., that can be saved and shared. We performed usability tests (supervised by a neuropsychologist) on six participants with aphasia who were able to communicate. Our work contributes (1) evidence that the functions implemented in the Aphasia Create tablet app suit the needs of target users, but older people are often not familiar with tactile devices, (2) reports that the Google Glass device may be problematic for persons with right-hand paresis, and (3) a characterization of the design guidelines for apps for aphasics. Both applications can be used to work with people with aphasia, and can be further developed. Aphasic centers, in which the apps were presented, expressed interest in using them to work with patients. The Aphasia Create app won the Enactus Poland National Competition in 2015.

Platelets level variability during the first year after liver transplantation in the risk prediction model for recipients mortality.

INTRODUCTION AND OBJECTIVES: Many scoring systems in liver diseases use static values of liver function parameters. These parameters may change significantly in liver transplant (LTx) recipients over time due to various processes. The study was aimed at building a new model for survival prediction after LTx based on variability of selected parameters. MATERIALS AND METHODS: The study included 450 LTx recipients who survived a minimum one year after transplantation. We analyzed liver enzymes and hematology parameters static values and their variability during the first year after transplantation. Modeling patients' survival was performed using Cox regression. Various sets of parameters (both static and variability and trends values) were tested to predict survival in our study group. Models' performance was measured using the concordance index. RESULTS: The single predictors of the patients survival were the static values of AST with C-index 0.706 (0.5883-0.7494), ALT 0.6102 (0.4843-0.6857) and bilirubin 0.6224 (0.5537-0.6695). High prediction scores were observed for variability in creatinine 0.6023 (0.5409-0.6451), PLT 0.6350 (0.5491-0.7043), RBC 0.5689 (0.5065-0.6213) and WBC 0.6506 (0.5095-0.7124). Our best-fitted and proposed model for patients survival after LTx has C-index 0.8273 (IQR 0.7767-0.8649). The model uses the following indicators for mortality prediction: the static value of AST, variability measure of PLT and trend measures of WBC and PLT. CONCLUSIONS: Adding variability and trend measures increases predictive accuracy in modeling patients survival after LTx. We propose a high-accuracy survival model in which variability and trend of PLT measures in the first year after transplantation are strong predictors of long-term mortality.

Discovering Speed Changes of Vehicles from Audio Data.

In this paper, we focus on detection of speed changes from audio data, representing recordings of cars passing a microphone placed near the road. The goal of this work is to observe the behavior of drivers near control points, in order to check whether their driving is safe both when approaching the speed camera and after passing it. The audio data were recorded in controlled conditions, and they are publicly available for downloading. They represent one of three classes: car accelerating, decelerating, or maintaining constant speed. We used SVM, random forests, and artificial neural networks as classifiers, as well as the time series based approach. We also tested several approaches to audio data representation, namely: average values of basic audio features within the analyzed time segment, parametric description of the time evolution of these features, and parametric description of curves (lines) in the spectrogram. Additionally, the combinations of these representations were used in classification experiments. As a final step, we constructed an ensemble classifier, consisting of the best models. The proposed solution achieved an accuracy of almost 95%, without mistaking acceleration with deceleration, and very rare mistakes between stable speed and speed changes. The outcomes of this work can become a basis for campaigns aiming at improving traffic safety.