Predicting forced, long-term radiative feedbacks from internal climate variability has been a decades-long quest in climate science. We train a convolutionalneural network (CNN) to predict annual- and global-mean top...
详细信息
Predicting forced, long-term radiative feedbacks from internal climate variability has been a decades-long quest in climate science. We train a convolutionalneural network (CNN) to predict annual- and global-mean top of the atmosphere radiation anomalies from time-varying maps of near-surface temperature in climate models. Trained on internal variability alone, the nonlinear CNN can predict radiation under strong climate change, outperforms a regularized linear regression approach, and works within and across different climate models. We show with explainable artificial intelligence methods that the CNN draws predictive skill from physically meaningful regions but at much smaller spatial scales than currently assumed.
Doping control screening analyses usually involve visual inspection of extracted ion chromatograms (EIC) by a trained analytical chemist, followed by further investigations if needed. This task is both highly repetiti...
详细信息
Doping control screening analyses usually involve visual inspection of extracted ion chromatograms (EIC) by a trained analytical chemist, followed by further investigations if needed. This task is both highly repetitive and time-consuming, given the hundreds of compounds and metabolites to be screened in tens of thousands of samples per year. With the recent widespread adoption of machine learning in analytical chemistry and the training of high-performance convolutional neural networks (CNN), these operations can be automated with high accuracy and throughput. Applying this technology to doping control is challenging as the false negative rate (FNR) shall be equal to zero. In this study, we demonstrated that implementing a deep learning strategy for chromatogram classification in equine doping control can be feasible and accurate. We illustrated our findings with a CNN scoring model combined with a linear discriminant analysis (LDA) classifier trained on chromatogram images from our ultra-high-pressure liquid chromatography coupled to high-resolution tandem mass spectrometry (UHPLC-HRMS/MS)-based biotherapeutics screening method. We expect that artificial intelligence (AI) will be a valuable tool for doping control laboratories in the near future.
This research introduces an innovative method using convolutional neural networks (CNNs) to identify mass imbalances in wind turbine rotors through a feature fusion strategy. To address the issue of class imbalance, t...
详细信息
This research introduces an innovative method using convolutional neural networks (CNNs) to identify mass imbalances in wind turbine rotors through a feature fusion strategy. To address the issue of class imbalance, the Synthetic Minority Oversampling Technique (SMOTE) is applied. A detailed simulation was carried out using a 1.5 MW three-bladed Wind Turbine model, employing tools such as Turbsim, FAST, and Matlab Simulink, to collect rotor speed data under different wind conditions. Mass imbalances were simulated by modifying blade density in the software. The fusion architecture combines feature extraction with Power Spectral Density analysis, improving the CNNs ability to work across both frequency and time domains. The effectiveness of this approach was confirmed through a comparative analysis with 9 classifiers and 4 different dataset combinations, demonstrating its capability in detecting mass imbalances.
Recently, there has been a proliferation of applied machine learning (ML) research, including the use of convolutional neural networks (CNNs) for direction of arrival (DoA) estimation. With the increasing amount of re...
详细信息
Recently, there has been a proliferation of applied machine learning (ML) research, including the use of convolutional neural networks (CNNs) for direction of arrival (DoA) estimation. With the increasing amount of research in this area, it is important to balance the performance and computational costs of CNNs with classical methods of DoA estimation such as Multiple Signal Classification (MUSIC). We outline the performance of both methods of DoA estimation for single-source and two-source cases for multiple array conditions. The results are also compared to the Cramer-Rao lower bound (CRLB) and conventional beamforming. For each source case, CNNs were trained for a perfect uniform line array (ULA) and tested against data from a perfect ULA, perturbed ULAs, ULAs with missing sensors, and ULAs with muffled sensors. We show that for the single-source case, the CNNs do not offer any performance improvement relative to MUSIC at low signal-to-noise ratio (SNR). For the two-source cases, the CNNs perform better than MUSIC but only at low SNR. For the remaining array cases, the CNNs outperform MUSIC. These results indicate that the performance improvements from CNNs are highest for situations where there is signal model to data mismatch (imperfect information). This work also illustrates that the CNN estimators developed in this work exceed the CRLB and are biased estimators caused by the lack of unbiased constraint in the loss function during training of the CNNs.
Background: The hippocampus plays a crucial role in memory and is one of the first structures affected by Alzheimer's disease. Postmortem MRI offers a way to quantify the alterations by measuring the atrophy of th...
详细信息
Background: The hippocampus plays a crucial role in memory and is one of the first structures affected by Alzheimer's disease. Postmortem MRI offers a way to quantify the alterations by measuring the atrophy of the inner structures of the hippocampus. Unfortunately, the manual segmentation of hippocampal subregions required to carry out these measures is very time-consuming. New Method: In this study, we explore the use of fully automated methods relying on state-of-the-art Deep Learning approaches to produce these annotations. More specifically, we propose a new segmentation framework made of a set of encoder-decoder blocks embedding self-attention mechanisms and atrous spatial pyramidal pooling to produce better maps of the hippocampus and identify four hippocampal regions: the dentate gyrus, the hippocampal head, the hippocampal body, and the hippocampal tail. Results: Trained using slices extracted from 15 postmortem T1-weighted, T2-weighted, and susceptibility- weighted MRI scans, our new approach produces hippocampus parcellations that are better aligned with the manually delineated parcellations provided by neuroradiologists. Comparison with Existing Methods: Four standard deep learning segmentation architectures: UNet, Double UNet, Attention UNet, and Multi-resolution UNet have been utilized for the qualitative and quantitative comparison of the proposed hippocampal region segmentation model. Conclusions: Postmortem MRI serves as a highly valuable neuroimaging technique for examining the effects of neurodegenerative diseases on the intricate structures within the hippocampus. This study opens the way to large sample-size postmortem studies of the hippocampal substructures.
A machine learning (ML) framework is proposed to achieve the automatic and rapid optimization of antenna topologies. A convolutionalneural network (CNN) is utilized as a surrogate model (SM) and is combined with rein...
详细信息
Artificial intelligence advancements have significantly sped up the development of specialized algorithms for diagnosing Papillary Thyroid Cancer from digital images. Several studies demonstrate that AI-based approach...
详细信息
Artificial intelligence advancements have significantly sped up the development of specialized algorithms for diagnosing Papillary Thyroid Cancer from digital images. Several studies demonstrate that AI-based approaches provide highly satisfactory performance, with one of them being convolutional neural networks (CNN). However, there is a noticeable gap in research regarding head-to-head comparisons of CNN architectures for identifying thyroid cancer histopathological imaging. This study seeks to address this gap by providing a thorough evaluation of 13 well-known CNN architectures performance using transfer learning. The selected CNN architecture includes CoAtNet-0, ConvNeXt Tiny, DenseNet121, DenseNet201, InceptionV3, InceptionResNetV2, EfficientNetV2B0, ResNet50, ResNet101, ResNet50V2, VGG19, VGG16, and Xception. The model will be assessed using a comprehensive set of metrics widely employed in medical applications, including accuracy, specificity, sensitivity, F1-score, negative predictive value, and positive predictive value in two different patch sizes of 512 x 512 and 256 x 256. ConvNeXt Tiny, ResNet50, and ResNet101 were proven as the leading models and demonstrated optimal performance across all metrics in both image patch sizes. The results emphasize the significance of model selection and appropriate patch size in identifying thyroid histopathological images.
We present the use of chess filters for the convolutional layers used in computer chess. We compare different types of blocks with and without chess filters. Our comparison uses the Leela Chess Zero (Lc0) T60 dataset ...
详细信息
In this paper, a convolutionalneural network parameter training method based on Hausdorff difference is proposed to solve the problems of gradient vanishing and local optimum in the momentum algorithm. A momentum alg...
详细信息
In the current era of manufacturing digitalization, computer vision is an emerging field of artificial intelligence (AI) that replicates human visual system functions by enabling computers and systems to identify and ...
详细信息
暂无评论