TemporalRI: subgraph isomorphism in temporal networks with multiple contacts

Temporal networks are graphs where each edge is associated with a timestamp denoting when two nodes interact. Temporal Subgraph Isomorphism (TSI) aims at retrieving all the subgraphs of a temporal network (called target) matching a smaller temporal network (called query), such that matched target edges appear in the same chronological order of corresponding query edges. Few algorithms have been proposed to solve the TSI problem (or variants of it) and most of them are applicable only to small or specific queries. In this paper we present TemporalRI, a new subgraph isomorphism algorithm for temporal networks with multiple contacts between nodes, which is inspired by RI algorithm. TemporalRI introduces the notion of temporal flows and uses them to filter the search space of candidate nodes for the matching. Our algorithm can handle queries of any size and any topology. Experiments on real networks of different sizes show that TemporalRI is very efficient compared to the state-of-the-art, especially for large queries and targets.

Assessment of OTT Pluvio2 Rain Intensity Measurements

This study investigates the OTT Pluvio2 weighing precipitation gauge’s random and systematic error components as well as stabilization of the measurements on time varying rainfall intensities (RI) under laboratory conditions. A highly precise programmable peristaltic pump that provided both constant and time varying RI was utilized in the experiments. Abrupt, gradual step, and cyclic step changes in the RI values were evaluated. RI readings were taken in real-time (RT) at different time resolution (6-60s) for the RI range of 6-70mm/h. Our analysis indicates that the lower threshold for the OTT Pluvio2’s real-time RI measurements should be redefined as 7mm/h at a one-minute resolution. Tolerance intervals containing 95% of the repeated measurements with probability 0.95 are given. It is shown that the measurement variances are unequal over the range of RI and the measurement repeatability is not uniform. A statistically significant negative bias was observed for the RI values of 7 and 8mm/h, while there was not a statistically significant linearity problem. Through the use of statistical control limits, it is shown that means of the RI measurements stabilized on the actual RI value. A detailed investigation on RT bucket weight measurements revealed a time delay in bucket weight measurements, which causes notable errors in reported RI measurements under dynamic rainfall conditions. To demonstrate the potentiality of large errors in Pluvio2’s real-time RI measurements, a set of equations was developed that faithfully reproduced the Pluvio2’s internal (hidden) algorithm, and results from dynamic laboratory and in-situ rainfall scenarios were simulated. The results of this investigation show the necessity of modifying the present Pluvio2 RI algorithm to enhance its performance and show the possibility of post-processing the existing Pluvio2 RI datasets for improved measurement accuracies.

Algorithm for Quantifying Frontal EMG Responsiveness for Sedation Monitoring

IntroductionTo study stimulation-related facial electromyographic (FEMG) activity in intensive care unit (ICU) patients, develop an algorithm for quantifying the FEMG activity, and to optimize the algorithm for monitoring the sedation state of ICU patients.MethodsFirst, the characteristics of FEMG response patterns related to vocal stimulation of 17 ICU patients were studied. Second, we collected continuous FEMG data from 30 ICU patients. Based on these data, we developed the Responsiveness Index (RI) algorithm that quantifies FEMG responses. Third, we compared the RI values with clinical sedation level assessments and adjusted algorithm parameters for best performance.ResultsIn patients who produced a clinically observed response to the vocal stimulus, the poststimulus FEMG power was 0.33 µV higher than the prestimulus power. In nonresponding patients, there was no difference. The sensitivity and specificity of the developed RI for detecting deep sedation in the subgroup with low probability of encephalopathy were 0.90 and 0.79, respectively.ConclusionConsistent FEMG patterns were found related to standard stimulation of ICU patients. A simple and robust algorithm was developed and good correlation with clinical sedation scores achieved in the development data.

Development of an Inverse Algorithm for Resonance Inspection

Resonance inspection (RI), which employs the natural frequency spectra shift between the good and the anomalous part populations to detect defects, is a nondestructive evaluation (NDE) technique with many advantages, such as low inspection cost, high testing speed, and broad applicability to structures with complex geometry compared to other contemporary NDE methods. It has already been widely used in the automobile industry for quality inspections of safety critical parts. Unlike some conventionally used NDE methods, the current RI technology is unable to provide details, i.e., location, dimension, or types, of the flaws for the discrepant parts. Such limitation severely hinders its widespread applications and further development. In this study, an inverse RI algorithm based on maximum correlation function is proposed to quantify the location and size of flaws for a discrepant part. A dog-bone-shaped stainless steel sample with and without controlled flaws is used for algorithm development and validation. The results show that multiple flaws can be accurately pinpointed back, using the algorithms developed, and the prediction accuracy decreases with increasing flaw numbers and decreasing distance between flaws.