Search results

Multi-infusion with integrated multiple pressure sensing allows earlier detection of line occlusions

BackgroundOcclusions of intravenous (IV) tubing can prevent vital and time-critical medication or solutions from being delivered into the bloodstream of patients receiving IV therapy. At low flow rates (≤ 1 ml/h) the alarm delay (time to an alert to the user) can be up to 2 h using conventional pressure threshold algorithms. In order to reduce alarm delays we developed and evaluated the performance of two new real-time occlusion detection algorithms and one co-occlusion detector that determines the correlation in trends in pressure changes for multiple pumps.MethodsBench-tested experimental runs were recorded in triplicate at rates of 1, 2, 4, 8, 16, and 32 ml/h. Each run consisted of 10 min of non-occluded infusion followed by a period of occluded infusion of 10 min or until a conventional occlusion alarm at 400 mmHg occurred. The first algorithm based on binary logistic regression attempts to detect occlusions based on the pump’s administration rate Q(t) and pressure sensor readings P(t). The second algorithm continuously monitored whether the actual variation in the pressure exceeded a threshold of 2 standard deviations (SD) above the baseline pressure. When a pump detected an occlusion using the SD algorithm, a third algorithm correlated the pressures of multiple pumps to detect the presence of a shared occlusion. The algorithms were evaluated using 6 bench-tested baseline single-pump occlusion scenarios, 9 single-pump validation scenarios and 7 multi-pump co-occlusion scenarios (i.e. with flow rates of 1 + 1, 1 + 2, 1 + 4, 1 + 8, 1 + 16, and 1 + 32 ml/h respectively). Alarm delay was the primary performance measure.ResultsIn the baseline single-pump occlusion scenarios, the overall mean ± SD alarm delay of the regression and SD algorithms were 1.8 ± 0.8 min and 0.4 ± 0.2 min, respectively. Compared to the delay of the conventional alarm this corresponds to a mean time reduction of 76% (P = 0.003) and 95% (P = 0.001), respectively. In the validation scenarios the overall mean ± SD alarm delay of the regression and SD algorithms were respectively 1.8 ± 1.6 min and 0.3 ± 0.2 min, corresponding to a mean time reduction of 77% and 95%. In the multi-pump scenarios a correlation > 0.8 between multiple pump pressures after initial occlusion detection by the SD algorithm had a mean ± SD alarm delay of 0.4 ± 0.2 min. In 2 out of the 9 validation scenarios an occlusion was not detected by the regression algorithm before a conventional occlusion alarm occurred. Otherwise no occlusions were missed.ConclusionsIn single pumps, both the regression and SD algorithm considerably reduced alarm delay compared to conventional pressure limit-based detection. The SD algorithm appeared to be more robust than the regression algorithm. For multiple pumps the correlation algorithm reliably detected co-occlusions. The latter may be used to localize the segment of tubing in which the occlusion occurs.

Deep learning with Python for crack detection

An illustrative non-technical review was published on Towards Data Science regarding our recent Journal paper “Automatic crack classification and segmentation on masonry surfaces using convolutional neural networks and transfer learning”.While new technologies have changed almost every aspect of our lives, the construction field seems to be struggling to catch up. Currently, the structural condition of a building is still predominantly manually inspected. In simple terms, even nowadays when a structure needs to be inspected for any damage, an engineer will manually check all the surfaces and take a bunch of photos while keeping notes of the position of any cracks. Then a few more hours need to be spent at the office to sort all the photos and notes trying to make a meaningful report out of it. Apparently this a laborious, costly, and subjective process. On top of that, safety concerns arise since there are parts of structures with access restrictions and difficult to reach. To give you an example, the Golden Gate Bridge needs to be periodically inspected. In other words, up to very recently there would be specially trained people who would climb across this picturesque structure and check every inch of it.

Long-Range Human Detection in Drone Camera Images

In recent years, drones have increasingly supported First Responders (FRs) in monitoring incidents and providing additional information. However, analysing drone footage is time-intensive and cognitively demanding. In this research, we investigate the use of AI models for the detection of humans in drone footage to aid FRs in tasks such as locating victims. Detecting small-scale objects, particularly humans from high altitudes, poses a challenge for AI systems. We present first steps of introducing and evaluating a series of YOLOv8 Convolutional Neural Networks (CNNs) for human detection from drone images. The models are fine-tuned on a created drone image dataset of the Dutch Fire Services and were able to achieve a 53.1% F1-Score, identifying 439 out of 825 humans in the test dataset. These preliminary findings, validated by an incident commander, highlight the promising utility of these models. Ongoing efforts aim to further refine the models and explore additional technologies.

Spatial anomaly detection in sensor networks using neighborhood information

Anomaly detection in sensor systems using lightweight machine learning

The maturing field of Wireless Sensor Networks (WSN) results in long-lived deployments that produce large amounts of sensor data. Lightweight online on-mote processing may improve the usage of their limited resources, such as energy, by transmitting only unexpected sensor data (anomalies). We detect anomalies by analyzing sensor reading predictions from a linear model. We use Recursive Least Squares (RLS) to estimate the model parameters, because for large datasets the standard Linear Least Squares Estimation (LLSE) is not resource friendly. We evaluate the use of fixed-point RLS with adaptive thresholding, and its application to anomaly detection in embedded systems. We present an extensive experimental campaign on generated and real-world datasets, with floating-point RLS, LLSE, and a rule-based method as benchmarks. The methods are evaluated on prediction accuracy of the models, and on detection of anomalies, which are injected in the generated dataset. The experimental results show that the proposed algorithm is comparable, in terms of prediction accuracy and detection performance, to the other LS methods. However, fixed-point RLS is efficiently implementable in embedded devices. The presented method enables online on-mote anomaly detection with results comparable to offline LS methods. © 2013 IEEE.

Detection of Botnet Command and Control Traffic by the Identification of Untrusted Destinations

We present a novel anomaly-based detection approach capable of detecting botnet Command and Control traffic in an enterprise network by estimating the trustworthiness of the traffic destinations. A traffic flow is classified as anomalous if its destination identifier does not origin from: human input, prior traffic from a trusted destination, or a defined set of legitimate applications. This allows for real-time detection of diverse types of Command and Control traffic. The detection approach and its accuracy are evaluated by experiments in a controlled environment.