Search results

Identification and Quantification of Activities Common to Intensive Care Patients

The aim of this study was to develop and validate an algorithm that can identify the type, frequency, and duration of activities common to intensive care (IC) patients. Ten healthy participants wore two accelerometers on their chest and leg while performing 14 activities clustered into four protocols (i.e., natural, strict, healthcare provider, and bed cycling). A video served as the reference standard, with two raters classifying the type and duration of all activities. This classification was reliable as intraclass correlations were all above 0.76 except for walking in the healthcare provider protocol, (0.29). The data of four participants were used to develop and optimize the algorithm by adjusting body-segment angles and rest-activity-threshold values based on percentage agreement (%Agr) with the reference. The validity of the algorithm was subsequently assessed using the data from the remaining six participants. %Agr of the algorithm versus the reference standard regarding lying, sitting activities, and transitions was 95%, 74%, and 80%, respectively, for all protocols except transitions with the help of a healthcare provider, which was 14-18%. For bed cycling, %Agr was 57-76%. This study demonstrated that the developed algorithm is suitable for identifying and quantifying activities common for intensive care patients. Knowledge on the (in)activity of these patients and their impact will optimize mobilization.

Quinolizidine Alkaloid Quantification in Lupin Beans

Lupin plants can grow on marginal lands and in the cold regions of Europe. They produce lupin beans, which contain around 30-40 % proteins and 20 % fats [1]. The high protein and fat content puts the lupin plant into direct competition with soy, which is mostly imported. Despite these promising nutritional values, the potential toxic quinolizidine alkaloid content of up to 4 % leads to prior testing before consumption. Therefore, four different extraction methods were tested and compared.

Quantification of damage evolution in masonry walls subjected to induced seismicity

This paper aims to quantify the evolution of damage in masonry walls under induced seismicity. A damage index equation, which is a function of the evolution of shear slippage and opening of the mortar joints, as well as of the drift ratio of masonry walls, was proposed herein. Initially, a dataset of experimental tests from in-plane quasi-static and cyclic tests on masonry walls was considered. The experimentally obtained crack patterns were investigated and their correlation with damage propagation was studied. Using a software based on the Distinct Element Method, a numerical model was developed and validated against full-scale experimental tests obtained from the literature. Wall panels representing common typologies of house façades of unreinforced masonry buildings in Northern Europe i.e. near the Groningen gas field in the Netherlands, were numerically investigated. The accumulated damage within the seismic response of the masonry walls was investigated by means of representative harmonic load excitations and an incremental dynamic analysis based on induced seismicity records from Groningen region. The ability of this index to capture different damage situations is demonstrated. The proposed methodology could also be applied to quantify damage and accumulation in masonry during strong earthquakes and aftershocks too.

Inefficient use of charging stations

Behavioral quantification of audiomotor transformations in improvising and score-dependent musicians

The historically developed practice of learning to play a music instrument from notes instead of by imitation or improvisation makes it possible to contrast two types of skilled musicians characterized not only by dissimilar performance practices, but also disparate methods of audiomotor learning. In a recent fMRI study comparing these two groups of musicians while they either imagined playing along with a recording or covertly assessed the quality of the performance, we observed activation of a right-hemisphere network of posterior superior parietal and dorsal premotor cortices in improvising musicians, indicating more efficient audiomotor transformation. In the present study, we investigated the detailed performance characteristics underlying the ability of both groups of musicians to replicate music on the basis of aural perception alone. Twenty-two classically trained improvising and score-dependent musicians listened to short, unfamiliar two-part excerpts presented with headphones. They played along or replicated the excerpts by ear on a digital piano, either with or without aural feedback. In addition, they were asked to harmonize or transpose some of the excerpts either to a different key or to the relative minor. MIDI recordings of their performances were compared with recordings of the aural model. Concordance was expressed in an audiomotor alignment score computed with the help of music information retrieval algorithms. Significantly higher alignment scores were found when contrasting groups, voices, and tasks. The present study demonstrates the superior ability of improvising musicians to replicate both the pitch and rhythm of aurally perceived music at the keyboard, not only in the original key, but also in other tonalities. Taken together with the enhanced activation of the right dorsal frontoparietal network found in our previous fMRI study, these results underscore the conclusion that the practice of improvising music can be associated with enhanced audiomotor transformation in response to aurally perceived music.

An End-to-End Pipeline for Uncertainty Quantification and Remaining Useful Life Estimation: An Application on Aircraft Engines

Estimating the remaining useful life (RUL) of an asset lies at the heart of prognostics and health management (PHM) of many operations-critical industries such as aviation. Mod- ern methods of RUL estimation adopt techniques from deep learning (DL). However, most of these contemporary tech- niques deliver only single-point estimates for the RUL without reporting on the confidence of the prediction. This practice usually provides overly confident predictions that can have severe consequences in operational disruptions or even safety. To address this issue, we propose a technique for uncertainty quantification (UQ) based on Bayesian deep learning (BDL). The hyperparameters of the framework are tuned using a novel bi-objective Bayesian optimization method with objectives the predictive performance and predictive uncertainty. The method also integrates the data pre-processing steps into the hyperparameter optimization (HPO) stage, models the RUL as a Weibull distribution, and returns the survival curves of the monitored assets to allow informed decision-making. We vali- date this method on the widely used C-MAPSS dataset against a single-objective HPO baseline that aggregates the two ob- jectives through the harmonic mean (HM). We demonstrate the existence of trade-offs between the predictive performance and the predictive uncertainty and observe that the bi-objective HPO returns a larger number of hyperparameter configurations compared to the single-objective baseline. Furthermore, we see that with the proposed approach, it is possible to configure models for RUL estimation that exhibit better or comparable performance to the single-objective baseline when validated on the test sets.