A Comparison Study of Machine Learning Based Algorithms for Fatigue Crack Growth Calculation.
Ontology highlight
ABSTRACT:
SUBMITTER: Wang H
PROVIDER: S-EPMC5459084 | biostudies-other | 2017 May
REPOSITORIES: biostudies-other
Similar Datasets
Project description:Data-driven models based on deep learning have led to tremendous breakthroughs in classical computer vision tasks and have recently made their way into natural sciences. However, the absence of domain knowledge in their inherent design significantly hinders the understanding and acceptance of these models. Nevertheless, explainability is crucial to justify the use of deep learning tools in safety-relevant applications such as aircraft component design, service and inspection. In this work, we train convolutional neural networks for crack tip detection in fatigue crack growth experiments using full-field displacement data obtained by digital image correlation. For this, we introduce the novel architecture ParallelNets-a network which combines segmentation and regression of the crack tip coordinates-and compare it with a classical U-Net-based architecture. Aiming for explainability, we use the Grad-CAM interpretability method to visualize the neural attention of several models. Attention heatmaps show that ParallelNets is able to focus on physically relevant areas like the crack tip field, which explains its superior performance in terms of accuracy, robustness, and stability.
| S-EPMC9184622 | biostudies-literature
Project description:The fatigue strength, effects of a notch on the fatigue strength, and fatigue crack growth rate of Ti-15Zr-4Nb-4Ta alloy were compared with those of other implantable metals. Zr, Nb, and Ta are important alloying elements for Ti alloys for attaining superior long-term corrosion resistance and biocompatibility. The highly biocompatible Ti-15Zr-4Nb-4Ta alloy exhibited an excellent balance between strength and ductility. Its notched tensile strength was much higher than that of a smooth specimen. The strength of 20% cold-worked commercially pure (C.P.) grade 4 Ti was close to that of Ti alloy. The tension-to-tension fatigue strength of an annealed Ti-15Zr-4Nb-4Ta rod at 107 cycles was approximately 740 MPa. The fatigue strength of this alloy was much improved by aging treatment after solution treatment. The fatigue strengths of C.P. grade 4 Ti and stainless steel were markedly improved by 20% cold working. The fatigue strength of Co-Cr-Mo alloy was markedly increased by hot forging. The notch fatigue strengths of 20% cold-worked C.P. grade 4 Ti, and annealed and aged Ti-15Zr-4Nb-4Ta, and annealed Ti-6Al-4V alloys were less than those of the smooth specimens. The fatigue crack growth rate of Ti-15Zr-4Nb-4Ta was the same as that of Ti-6Al-4V. The fatigue crack growth rate in 0.9% NaCl was the same as that in air. Stainless steel and Co-Cr-Mo-Ni-Fe alloy had a larger stress-intensity factor range (?K) than Ti alloy.
| S-EPMC5449063 | biostudies-literature
Project description:In this work, plasma samples of 5 metabolic syndrome patients and 5 healthy volunteers were collected. Then, high-throughput RNA sequencing was performed to detect the expression of plasma coding RNA.
2021-05-29 | E-MTAB-10494 | biostudies-arrayexpress
Project description:PurposeTo develop and test machine learning classifiers (MLCs) for determining visual field progression.MethodsIn total, 90,713 visual fields from 13,156 eyes were included. Six different progression algorithms (linear regression of mean deviation, linear regression of the visual field index, Advanced Glaucoma Intervention Study algorithm, Collaborative Initial Glaucoma Treatment Study algorithm, pointwise linear regression [PLR], and permutation of PLR) were applied to classify each eye as progressing or stable. Six MLCs were applied (logistic regression, random forest, extreme gradient boosting, support vector classifier, convolutional neural network, fully connected neural network) using a training and testing set. For MLC input, visual fields for a given eye were divided into the first and second half and each location averaged over time within each half. Each algorithm was tested for accuracy, sensitivity, positive predictive value, and class bias with a subset of visual fields labeled by a panel of three experts from 161 eyes.ResultsMLCs had similar performance metrics as some of the conventional algorithms and ranged from 87% to 91% accurate with sensitivity ranging from 0.83 to 0.88 and specificity from 0.92 to 0.96. All conventional algorithms showed significant class bias, meaning each individual algorithm was more likely to grade uncertain cases as either progressing or stable (P ≤ 0.01). Conversely, all MLCs were balanced, meaning they were equally likely to grade uncertain cases as either progressing or stable (P ≥ 0.08).ConclusionsMLCs showed a moderate to high level of accuracy, sensitivity, and specificity and were more balanced than conventional algorithms.Translational relevanceMLCs may help to determine visual field progression.
| S-EPMC8237084 | biostudies-literature
Project description:Non-isothermal conditions during flight cycles have long led to the requirement for thermo-mechanical fatigue (TMF) evaluation of aerospace materials. However, the increased temperatures within the gas turbine engine have meant that the requirements for TMF testing now extend to disc alloys along with blade materials. As such, fatigue crack growth rates are required to be evaluated under non-isothermal conditions along with the development of a detailed understanding of related failure mechanisms. In the current work, a TMF crack growth testing method has been developed utilising induction heating and direct current potential drop techniques for polycrystalline nickel-based superalloys, such as RR1000. Results have shown that in-phase (IP) testing produces accelerated crack growth rates compared with out-of-phase (OOP) due to increased temperature at peak stress and therefore increased time dependent crack growth. The ordering of the crack growth rates is supported by detailed fractographic analysis which shows intergranular crack growth in IP test specimens, and transgranular crack growth in 90° OOP and 180° OOP tests. Isothermal tests have also been carried out for comparison of crack growth rates at the point of peak stress in the TMF cycles.
| S-EPMC5344579 | biostudies-other
Project description:PurposeAbsorbed dose calculation by kernel convolution requires the prior determination of dose point kernels (DPK). This study reports on the design, implementation, and test of a multi-target regressor approach to generate the DPKs for monoenergetic sources and a model to obtain DPKs for beta emitters.MethodsDPK for monoenergetic electron sources were calculated using the FLUKA Monte Carlo (MC) code for many materials of clinical interest and initial energies ranging from 10 to 3000 keV. Regressor Chains (RC) with three different coefficients regularization/shrinkage models were used as base regressors. Electron monoenergetic scaled DPKs (sDPKs) were used to assess the corresponding sDPKs for beta emitters typically used in nuclear medicine, which were compared against reference published data. Finally, the beta emitters sDPK were applied to a patient-specific case calculating the Voxel Dose Kernel (VDK) for a hepatic radioembolization treatment with [Formula: see text]Y.ResultsThe three trained machine learning models demonstrated a promising capacity to predict the sDPK for both monoenergetic emissions and beta emitters of clinical interest attaining differences lower than [Formula: see text] in the mean average percentage error (MAPE) as compared with previous studies. Furthermore, differences lower than [Formula: see text] were obtained for the absorbed dose in patient-specific dosimetry comparing against full stochastic MC calculations.ConclusionAn ML model was developed to assess dosimetry calculations in nuclear medicine. The implemented approach has shown the capacity to accurately predict the sDPK for monoenergetic beta sources in a wide range of energy in different materials. The ML model to calculate the sDPK for beta-emitting radionuclides allowed to obtain VDK useful to achieve reliable patient-specific absorbed dose distributions required short computation times.
| S-EPMC10293553 | biostudies-literature
Project description:Deformation-induced martensitic transformation (DIMT) has been used for designing high-performance alloys to prevent structural failure under static loads. Its effectiveness against fatigue, however, is unclear. This limits the application of DIMT for parts that are exposed to variable loads, although such scenarios are the rule and not the exception for structural failure. Here we reveal the dual role of DIMT in fatigue crack growth through in situ observations. Two antagonistic fatigue mechanisms mediated by DIMT are identified, namely, transformation-mediated crack arresting, which prevents crack growth, and transformation-mediated crack coalescence, which promotes crack growth. Both mechanisms are due to the hardness and brittleness of martensite as a transformation product, rather than to the actual transformation process itself. In fatigue crack growth, the prevalence of one mechanism over the other critically depends on the crack size and the mechanical stability of the parent austenite phase. Elucidating the two mechanisms and their interplay allows for the microstructure design and safe use of metastable alloys that experience fatigue loads. The findings also generally reveal how metastable alloy microstructures must be designed for materials to be fatigue-resistant.
| S-EPMC8892355 | biostudies-literature
Project description:Motorcycle crash severity is under-researched in Ghana. Thus, the probable risk factors and association between these factors and motorcycle crash severity outcomes is not known. Traditional statistical models have intrinsic assumptions and pre-defined correlations that, if flouted, can generate inaccurate results. In this study, machine learning based algorithms were employed to predict and classify motorcycle crash severity. Machine learning based techniques are non-parametric models without the presumption of relationships between endogenous and exogenous variables. The main aim of this research is to evaluate and compare different approaches to modeling motorcycle crash severity as well as investigating the effect of risk factors on the injury outcomes of motorcycle crashes. Motorcycle crash dataset between 2011 and 2015 was extracted from the National Road Traffic Crash Database at the Building and Road Research Institute (BRRI) in Ghana. The dataset was classified into four injury severity categories: fatal, hospitalized, injured, and damage-only. Three machine learning based models were developed: J48 Decision Tree Classifier, Random Forest (RF) and Instance-Based learning with parameter k (IBk) were employed to model the severity of injury in a motorcycle crash. These machine learning algorithms were validated using 10-fold cross-validation technique. The three machine learning based algorithms were compared with one another and the statistical model: multinomial logit model (MNLM). Also, the relative importance analysis of the attribute was conducted to determine the impact of these attributes on injury severity outcomes. The results of the study reveal that the predictions of machine learning algorithms are superior to the MNLM in accuracy and effectiveness, and the RF-based algorithms show the overall best agreement with the experimental data out of the three machine learning algorithms, for its global optimization and extrapolation ability. Location type, time of the crash, settlement type, collision partner, collision type, road separation, road surface type, the day of the week, and road shoulder condition were found as the critical determinants of motorcycle crash injury severity.
| S-EPMC6448880 | biostudies-literature
Project description:An accurate step length estimation can provide valuable information to different applications such as indoor positioning systems or it can be helpful when analyzing the gait of a user, which can then be used to detect various gait impairments that lead to a reduced step length (caused by e.g., Parkinson's disease or multiple sclerosis). In this paper, we focus on the estimation of the step length using machine learning techniques that could be used in an indoor positioning system. Previous step length algorithms tried to model the length of a step based on measurements from the accelerometer and some tuneable (user-specific) parameters. Machine-learning-based step length estimation algorithms eliminate these parameters to be tuned. Instead, to adapt these algorithms to different users, it suffices to provide examples of the length of multiple steps for different persons to the machine learning algorithm, so that in the training phase the algorithm can learn to predict the step length for different users. Until now, these machine learning algorithms were trained with features that were chosen intuitively. In this paper, we consider a systematic feature selection algorithm to be able to determine the features from a large collection of features, resulting in the best performance. This resulted in a step length estimator with a mean absolute error of 3.48 cm for a known test person and 4.19 cm for an unknown test person, while current state-of-the-art machine-learning-based step length estimators resulted in a mean absolute error of 4.94 cm and 6.27 cm for respectively a known and unknown test person.
| S-EPMC7038475 | biostudies-literature
Project description:OBJECTIVE:The Centers for Disease Control and Prevention (CDC) coordinates a labor-intensive process to measure the prevalence of autism spectrum disorder (ASD) among children in the United States. Random forests methods have shown promise in speeding up this process, but they lag behind human classification accuracy by about 5%. We explore whether more recently available document classification algorithms can close this gap. MATERIALS AND METHODS:Using data gathered from a single surveillance site, we applied 8 supervised learning algorithms to predict whether children meet the case definition for ASD based solely on the words in their evaluations. We compared the algorithms' performance across 10 random train-test splits of the data, using classification accuracy, F1 score, and number of positive calls to evaluate their potential use for surveillance. RESULTS:Across the 10 train-test cycles, the random forest and support vector machine with Naive Bayes features (NB-SVM) each achieved slightly more than 87% mean accuracy. The NB-SVM produced significantly more false negatives than false positives (P = 0.027), but the random forest did not, making its prevalence estimates very close to the true prevalence in the data. The best-performing neural network performed similarly to the random forest on both measures. DISCUSSION:The random forest performed as well as more recently available models like the NB-SVM and the neural network, and it also produced good prevalence estimates. NB-SVM may not be a good candidate for use in a fully-automated surveillance workflow due to increased false negatives. More sophisticated algorithms, like hierarchical convolutional neural networks, may not be feasible to train due to characteristics of the data. Current algorithms might perform better if the data are abstracted and processed differently and if they take into account information about the children in addition to their evaluations. CONCLUSION:Deep learning models performed similarly to traditional machine learning methods at predicting the clinician-assigned case status for CDC's autism surveillance system. While deep learning methods had limited benefit in this task, they may have applications in other surveillance systems.
| S-EPMC6760799 | biostudies-literature