Time-based Threshold Research Articles

Automated Patient Discomfort Detection Using Deep Learning

The Internet of Things (IoT) has been transformed almost all fields of life, but its impact on the healthcare sector has been notable. Various IoT-based sensors are used in the healthcare sector and offer quality and safe care to patients. This work presents a deep learning-based automated patient discomfort detection system in which patients’ discomfort is non-invasively detected. To do this, the overhead view patients’ data set has been recorded. For testing and evaluation purposes, we investigate the power of deep learning by choosing a Convolution Neural Network (CNN) based model. The model uses confidence maps and detects 18 different key points at various locations of the body of the patient. Applying association rules and part affinity fields, the detected key points are later converted into six main body organs. Furthermore, the distance of subsequent key points is measured using coordinates information. Finally, distance and the time-based threshold are used for the classification of movements associated with discomfort or normal conditions. The accuracy of the proposed system is assessed on various test sequences. The experimental outcomes reveal the worth of the proposed system’ by obtaining a True Positive Rate of 98% with a 2% False Positive Rate.

Transition time threshold for Double Cantilever Beam specimens under high loading rates

Inertial force effects complicate the determination of the mode I interlaminar fracture toughness under high loading rates in composites. Although using a quasi-static analysis is a common approach, it is not always valid. A time-based threshold criterion is proposed to determine when dynamic effects might be neglected during the analysis of a high loading rate Double Cantilever Beam test. The criterion compares a transition time, time after which inertia effects can be neglected, versus the time for the initiation of fracture propagation. Three different methods are considered for the transition time. Good agreement is found when comparing the expressions with numerical simulations. It is also demonstrated that the transition time is affected by the velocity profile. The proposed criterion and approach to determine the transition time are useful tools to define when a quasi-static data-reduction scheme can be used.

Late-rejection, a strategy to perform an overflow policy

Motivated by overflow policies implemented in service systems, we consider a multi-server queue with customers’ abandonment where rejection control is exercised on customers currently waiting in the queue. Our aim is to find a good balance between conflicting goals, namely, the rate of rejected customers and a cost function which may involve wait and abandonment metrics like percentiles of the waiting time or rate of abandonment. We develop a Markov decision process approach where the waiting time of the first customer in line is used in a discretized form to define the system state. We show that a time-based threshold policy is optimal, and develop a procedure to compute the optimal threshold. Our analysis explains some known behaviors in practice. For instance, if the cost function is constant in the system state like with wait percentiles, then the optimal threshold is one of the time limits defining the percentiles. Also, abandonment is shown to have beneficial or detrimental effect depending on the system manager’s objective.

Multiple Sleeping States for Energy Saving in CNC Machining Centers

Energy efficient control policies that switch off/on the machine have been proposed in recent literature. However, such policies control the whole machine simultaneously and consider only one stand-by state. This paper extends a time-based threshold policy by modeling several sleeping states and controlling machine components. Machine idle time is stochastic and the expected value of the energy consumed per produced part is minimized while assuring a certain productivity target. Technological constraints for control applicability are discussed. Results are based on a 5-axes Computer Numerical Control (CNC) machining center and savings are significant compared with that of state-of-the-art policies.

A Generalized Method to Extract Visual Time-Sharing Sequences From Naturalistic Driving Data

Indicators based on visual time-sharing have been used to investigate drivers’ visual behaviour during additional task execution. However, visual time-sharing analyses have been restricted to additional tasks with well-defined temporal start and end points and a dedicated visual target area. We introduce a method to automatically extract visual time-sharing sequences directly from eye tracking data. This facilitates investigations of systems, providing continuous information without well-defined start and end points. Furthermore, it becomes possible to investigate time-sharing behavior with other types of glance targets such as the mirrors. Time-sharing sequences are here extracted based on between-glance durations. If glances to a particular target are separated by less than a time-based threshold value, we assume that they belong to the same information intake event. Our results indicate that a 4-s threshold is appropriate. Examples derived from 12 drivers (about 100 hours of eye tracking data), collected in an on-road investigation of an in-vehicle information system, are provided to illustrate sequence-based analyses. This includes the possibility to investigate human-machine interface designs based on the number of glances in the extracted sequences, and to increase the legibility of transition matrices by deriving them from time-sharing sequences instead of single glances. More object-oriented glance behavior analyses, based on additional sensor and information fusion, are identified as the next future step. This would enable automated extraction of time-sharing sequences not only for targets fixed in the vehicle’s coordinate system, but also for environmental and traffic targets that move independently of the driver’s vehicle.

How Reliable Is Perfusion MR in Acute Stroke?

Perfusion magnetic resonance imaging (pMR) is increasingly used in acute stroke, but its physiologic significance is still debated. A reasonably good correlation between pMR and positron emission tomography (PET) has been reported in normal subjects and chronic cerebrovascular disease, but corresponding validation in acute stroke is still lacking. We compared the cerebral blood flow (CBF), cerebral blood volume, and mean transit time (MTT) maps generated by pMR (deconvolution method) and PET ((15)O steady-state method) in 5 patients studied back-to-back with the 2 modalities at a mean of 16 hours (range, 7 to 21 hours) after stroke onset. We also determined the penumbra thresholds for pMR-derived MTT, time to peak (TTP), and Tmax against the previously validated probabilistic PET penumbra thresholds. In all patients, the PET and pMR relative distribution images were remarkably similar, especially for CBF and MTT. Within-patient correlations between pMR and PET were strong for absolute CBF (average r(2)=0.45) and good for MTT (r(2)=0.35) but less robust for cerebral blood volume (r(2)=0.24). However, pMR overestimated absolute CBF and underestimated MTT, with substantial variability in individual slopes. Removing individual differences by normalization to the mean resulted in much stronger between-patient correlations. Penumbra thresholds of approximately 6, 4.8, and 5.5 seconds were obtained for MTT delay, TTP delay, and Tmax, respectively. Although derived from a small sample studied relatively late after stroke onset, our data show that pMR tends to overestimate absolute CBF and underestimate MTT, but the relative distribution of the perfusion variables was remarkably similar between pMR and PET. pMR appears sufficiently reliable for clinical purposes and affords reliable detection of the penumbra from normalized time-based thresholds.