Event Detection for Distributed Acoustic Sensing: Combining Knowledge-Based, Classical Machine Learning, and Deep Learning Approaches.
Ontology highlight
ABSTRACT:
SUBMITTER: Bublin M
PROVIDER: S-EPMC8618866 | biostudies-literature |
REPOSITORIES: biostudies-literature
Similar Datasets
Project description:We present one of the first case studies demonstrating the use of distributed acoustic sensing deployed on regional unlit fiber-optic telecommunication infrastructure (dark fiber) for broadband seismic monitoring of both near-surface soil properties and earthquake seismology. We recorded 7 months of passive seismic data on a 27?km section of dark fiber stretching from West Sacramento, CA to Woodland, CA, densely sampled at 2?m spacing. This dataset was processed to extract surface wave velocity information using ambient noise interferometry techniques; the resulting VS profiles were used to map both shallow structural profiles and groundwater depth, thus demonstrating that basin-scale variations in hydrological state could be resolved using this technique. The same array was utilized for detection of regional and teleseismic earthquakes and evaluated for long period response using records from the M8.1 Chiapas, Mexico 2017, Sep 8th event. The combination of these two sets of observations conclusively demonstrates that regionally extensive fiber-optic networks can effectively be utilized for a host of geoscience observation tasks at a combination of scale and resolution previously inaccessible.
| S-EPMC6362068 | biostudies-literature
Project description:Earthquake focal mechanisms provide critical in-situ insights about the subsurface faulting geometry and stress state. For frequent small earthquakes (magnitude< 3.5), their focal mechanisms are routinely determined using first-arrival polarities picked on the vertical component of seismometers. Nevertheless, their quality is usually limited by the azimuthal coverage of the local seismic network. The emerging distributed acoustic sensing (DAS) technology, which can convert pre-existing telecommunication cables into arrays of strain/strain-rate meters, can potentially fill the azimuthal gap and enhance constraints on the nodal plane orientation through its long sensing range and dense spatial sampling. However, determining first-arrival polarities on DAS is challenging due to its single-component sensing and low signal-to-noise ratio for direct body waves. Here, we present a data-driven method that measures P-wave polarities on a DAS array based on cross-correlations between earthquake pairs. We validate the inferred polarities using the regional network catalog on two DAS arrays, deployed in California and each comprising ~ 5000 channels. We demonstrate that a joint focal mechanism inversion combining conventional and DAS polarity picks improves the accuracy and reduces the uncertainty in the focal plane orientation. Our results highlight the significant potential of integrating DAS with conventional networks for investigating high-resolution earthquake source mechanisms.
| S-EPMC10345142 | biostudies-literature
Project description:Understanding physical processes prior to and during volcanic eruptions has improved significantly in recent years. However, uncertainties about subsurface structures distorting observed signals and undetected processes within the volcano prevent volcanologists to infer subtle triggering mechanisms of volcanic phenomena. Here, we demonstrate that distributed acoustic sensing (DAS) with optical fibres allows us to identify volcanic events remotely and image hidden near-surface volcanic structural features. We detect and characterize strain signals associated with explosions and locate their origin using a 2D-template matching between picked and theoretical wave arrival times. We find evidence for non-linear grain interactions in a scoria layer of spatially variable thickness. We demonstrate that wavefield separation allows us to incrementally investigate the ground response to various excitation mechanisms. We identify very small volcanic events, which we relate to fluid migration and degassing. Those results provide the basis for improved volcano monitoring and hazard assessment using DAS.
| S-EPMC8971480 | biostudies-literature
Project description:Objective: In emergency medicine, airway management is a core skill that includes endotracheal intubation (ETI), a common technique that can result in ineffective ventilation and laryngotracheal injury if executed incorrectly. We present a method for automatically generating performance feedback during ETI simulator training, potentially augmenting training outcomes on robotic simulators. Method: Electret microphones recorded ultrasonic echoes pulsed through the complex geometry of a simulated airway during ETI performed on a full-size patient simulator. As the endotracheal tube is inserted deeper and the cuff is inflated, the resulting changes in geometry are reflected in the recorded signal. We trained machine learning models to classify 240 intubations distributed equally between six conditions: three insertion depths and two cuff inflation states. The best performing models were cross validated in a leave-one-subject-out scheme. Results: Best performance was achieved by transfer learning with a convolutional neural network pre-trained for sound classification, reaching global accuracy above 98% on 1-second-long audio test samples. A support vector machine trained on different features achieved a median accuracy of 85% on the full label set and 97% on a reduced label set of tube depth only. Significance: This proof-of-concept study demonstrates a method of measuring qualitative performance criteria during simulated ETI in a relatively simple way that does not damage ecological validity of the simulated anatomy. As traditional sonar is hampered by geometrical complexity compounded by the introduced equipment in ETI, the accuracy of machine learning methods in this confined design space enables application in other invasive procedures. By enabling better interaction between the human user and the robotic simulator, this approach could improve training experiences and outcomes in medical simulation for ETI as well as many other invasive clinical procedures.
| S-EPMC10642916 | biostudies-literature
Project description:Distributed acoustic sensing based on phase-sensitive optical time-domain reflectometry (?-OTDR) has been widely used in many fields. Phase demodulation of the ?-OTDR signal is essential for undistorted acoustic measurement. Digital coherent detection is a universal method to implement phase demodulation, but it may cause severe computational burden. In this paper, analog I/Q demodulation is introduced into the ?-OTDR based DAS system to solve this problem, which can directly obtain the I and Q components of the beat signal without any digital processing, meaning that the computational cost can be sharply reduced. Besides, the sampling frequency of the data acquisition card can theoretically be lower than the beat frequency as the spectrum aliasing would not affect the demodulation results, thus further reducing the data volume of the system. Experimental results show that the proposed DAS system can demodulate the phase signal with good linearity and wide frequency response range. It can also adequately recover the sound signal sensed by the optical fiber, indicating that it can be a promising solution for computational-cost-sensitive distributed acoustic sensing applications.
| S-EPMC6749300 | biostudies-literature
Project description:Records of Alpine microseismicity are a powerful tool to study landscape-shaping processes and warn against hazardous mass movements. Unfortunately, seismic sensor coverage in Alpine regions is typically insufficient. Here we show that distributed acoustic sensing (DAS) bridges critical observational gaps of seismogenic processes in Alpine terrain. Dynamic strain measurements in a 1?km long fiber optic cable on a glacier surface produce high-quality seismograms related to glacier flow and nearby rock falls. The nearly 500 cable channels precisely locate a series of glacier stick-slip events (within 20-40?m) and reveal seismic phases from which thickness and material properties of the glacier and its bed can be derived. As seismic measurements can be acquired with fiber optic cables that are easy to transport, install and couple to the ground, our study demonstrates the potential of DAS technology for seismic monitoring of glacier dynamics and natural hazards.
| S-EPMC7229220 | biostudies-literature
Project description:Machine learning (ML) techniques offer a promising avenue for improving the integration of remote sensing data into mathematical crop models, thereby enhancing crop growth prediction accuracy. A critical variable for this integration is the leaf area index (LAI), which can be accurately assessed using proximal or remote sensing data based on plant canopies. This study aimed to (1) develop a machine learning-based method for estimating the LAI in rice and soybean crops using proximal sensing data and (2) evaluate the performance of a Remote Sensing-Integrated Crop Model (RSCM) when integrated with the ML algorithms. To achieve these objectives, we analyzed rice and soybean datasets to identify the most effective ML algorithms for modeling the relationship between LAI and vegetation indices derived from canopy reflectance measurements. Our analyses employed a variety of ML regression models, including ridge, lasso, support vector machine, random forest, and extra trees. Among these, the extra trees regression model demonstrated the best performance, achieving test scores of 0.86 and 0.89 for rice and soybean crops, respectively. This model closely replicated observed LAI values under different nitrogen treatments, achieving Nash-Sutcliffe efficiencies of 0.93 for rice and 0.97 for soybean. Our findings show that incorporating ML techniques into RSCM effectively captures seasonal LAI variations across diverse field management practices, offering significant potential for improving crop growth and productivity monitoring.
| S-EPMC10894942 | biostudies-literature
Project description:Telomerase is a ribonucleoprotein enzyme that adds telomeric DNA repeat sequences to the ends of linear chromosomes. The enzyme plays pivotal roles in cellular senescence and aging, and because it provides a telomere maintenance mechanism for approximately 90% of human cancers, it is a promising target for cancer therapy. Despite its importance, a high-resolution structure of the telomerase enzyme has been elusive, although a crystal structure of an N-terminal domain (TEN) of the telomerase reverse transcriptase subunit (TERT) from Tetrahymena has been reported. In this study, we used a comparative strategy, in which sequence-based machine learning approaches were integrated with computational structural modeling, to explore the potential conservation of structural and functional features of TERT in phylogenetically diverse species. We generated structural models of the N-terminal domains from human and yeast TERT using a combination of threading and homology modeling with the Tetrahymena TEN structure as a template. Comparative analysis of predicted and experimentally verified DNA and RNA binding residues, in the context of these structures, revealed significant similarities in nucleic acid binding surfaces of Tetrahymena and human TEN domains. In addition, the combined evidence from machine learning and structural modeling identified several specific amino acids that are likely to play a role in binding DNA or RNA, but for which no experimental evidence is currently available.
| S-EPMC2569851 | biostudies-literature
Project description:The observed increase in extreme weather has prompted recent methodological advances in extreme event attribution. We propose a machine learning-based approach that uses convolutional neural networks to create dynamically consistent counterfactual versions of historical extreme events under different levels of global mean temperature (GMT). We apply this technique to one recent extreme heat event (southcentral North America 2023) and several historical events that have been previously analyzed using established attribution methods. We estimate that temperatures during the southcentral North America event were 1.18° to 1.42°C warmer because of global warming and that similar events will occur 0.14 to 0.60 times per year at 2.0°C above preindustrial levels of GMT. Additionally, we find that the learned relationships between daily temperature and GMT are influenced by the seasonality of the forced temperature response and the daily meteorological conditions. Our results broadly agree with other attribution techniques, suggesting that machine learning can be used to perform rapid, low-cost attribution of extreme events.
| S-EPMC11338235 | biostudies-literature
Project description:We present a simple and efficient hypothesis-free machine learning pipeline for risk factor discovery that accounts for non-linearity and interaction in large biomedical databases with minimal variable pre-processing. In this study, mortality models were built using gradient boosting decision trees (GBDT) and important predictors were identified using a Shapley values-based feature attribution method, SHAP values. Cox models controlled for false discovery rate were used for confounder adjustment, interpretability, and further validation. The pipeline was tested using information from 502,506 UK Biobank participants, aged 37-73 years at recruitment and followed over seven years for mortality registrations. From the 11,639 predictors included in GBDT, 193 potential risk factors had SHAP values ≥ 0.05, passed the correlation test, and were selected for further modelling. Of the total variable importance summed up, 60% was directly health related, and baseline characteristics, sociodemographics, and lifestyle factors each contributed about 10%. Cox models adjusted for baseline characteristics, showed evidence for an association with mortality for 166 out of the 193 predictors. These included mostly well-known risk factors (e.g., age, sex, ethnicity, education, material deprivation, smoking, physical activity, self-rated health, BMI, and many disease outcomes). For 19 predictors we saw evidence for an association in the unadjusted but not adjusted analyses, suggesting bias by confounding. Our GBDT-SHAP pipeline was able to identify relevant predictors 'hidden' within thousands of variables, providing an efficient and pragmatic solution for the first stage of hypothesis free risk factor identification.
| S-EPMC8626442 | biostudies-literature