TY - JOUR A1 - Lapenna, M. A1 - Tsamos, Athanasios A1 - Faglion, F. A1 - Fioresi, R. A1 - Zanchetta, F. A1 - Bruno, Giovanni T1 - Vision GNN (ViG) architecture for a fine‑tuned segmentation of a complex Al–Si metal matrix composite XCT volume N2 - In this paper, we implement a vision graph neural network (ViG) architecture to segment microstructures in X-ray computed tomography 3D data. Our ViG architecture is first trained on a synthetic augmented dataset, and then fine-tuned on experimental data to obtain an improved segmentation. Successively, we assess the accuracy of the segmentation on manually-labeled experimental slices. We exemplarily use the approach on a complex microstructure: a metal matrix composite, reinforced with two ceramic phases, intermetallic inclusions and a silicon network, in order to show the generality of our method. ViG model proves to be more efficient than U-Nets in adapting to new data when fine-tuned on a small portion of the experimental data. The fine-tuned ViG shows comparable performance to U-Nets, while largely reducing the number of trainable parameters, with the potential of greater adaptability and efficiency. KW - X-ray Computed tomography KW - Machine Learning KW - Virtual XCT KW - Segmentation KW - Composites PY - 2025 UR - https://nbn-resolving.org/urn:nbn:de:kobv:b43-630593 DO - https://doi.org/10.1007/s10853-025-10834-5 SN - 1573-4803 VL - 60 SP - 6907 EP - 6921 PB - Springer AN - OPUS4-63059 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER - TY - JOUR A1 - Tsamos, Athanasios A1 - Evsevleev, Sergei A1 - Fioresi, R. A1 - Faglioni, F. A1 - Bruno, Giovanni T1 - Synthetic Data Generation for Automatic Segmentation of X-ray Computed Tomography Reconstructions of Complex Microstructures N2 - The greatest challenge when using deep convolutional neural networks (DCNNs) for automatic segmentation of microstructural X-ray computed tomography (XCT) data is the acquisition of sufficient and relevant data to train the working network. Traditionally, these have been attained by manually annotating a few slices for 2D DCNNs. However, complex multiphase microstructures would presumably be better segmented with 3D networks. However, manual segmentation labeling for 3D problems is prohibitive. In this work, we introduce a method for generating synthetic XCT data for a challenging six-phase Al–Si alloy composite reinforced with ceramic fibers and particles. Moreover, we propose certain data augmentations (brightness, contrast, noise, and blur), a special in-house designed deep convolutional neural network (Triple UNet), and a multi-view forwarding strategy to promote generalized learning from synthetic data and therefore achieve successful segmentations. We obtain an overall Dice score of 0.77. Lastly, we prove the detrimental effects of artifacts in the XCT data on achieving accurate segmentations when synthetic data are employed for training the DCNNs. The methods presented in this work are applicable to other materials and imaging techniques as well. Successful segmentation coupled with neural networks trained with synthetic data will accelerate scientific output. KW - Automatic segmentation KW - 3D deep convolutional neural network (3D DCNN) KW - Dice score KW - Metal matrix composite (MMC) KW - Modified U-Net architectures KW - Multi-phase materials PY - 2023 UR - https://nbn-resolving.org/urn:nbn:de:kobv:b43-571243 DO - https://doi.org/10.3390/jimaging9020022 VL - 9 IS - 2 SP - 1 EP - 23 PB - MDPI AN - OPUS4-57124 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER - TY - JOUR A1 - Bruno, Giovanni A1 - Lapenna, M. A1 - Faglioni, F. A1 - Fioresi, R. T1 - Temperature-based pruning for input features in Graph Neural Networks N2 - In the presentwork,we employ the concept of neural network temperature to prune unimportant features in input to aGraph Neural Network (GNN) architecture. In benchmark datasets for node and graph property prediction, each node comes equipped with a vector of numerous features. It is paramount to understand which information is actually necessary and which can be discarded, both for efficiency and explainability. The temperature is linked to the gradient activity due to the loss function minimization and leads to pruning of weight structures associated with small gradients. This study is done on different GNN architectures, one for node classification and another one for link prediction, and several benchmark datasets are employed.We compare the results with similar experiments previously conducted on the filters of Convolutional Neural Networks. Although still at the proof-of-concept stage, our temperature-based pruning technique stands as a promising alternative to state-of-the-art magnitude-based pruning techniques. KW - Temperature-based pruning KW - Graph Neural Networks PY - 2025 UR - https://nbn-resolving.org/urn:nbn:de:kobv:b43-640886 VL - 140 SP - 1 EP - 20 PB - Springer AN - OPUS4-64088 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER - TY - JOUR A1 - Tsamos, Athanasios A1 - Evsevleev, Sergei A1 - Fioresi, R. A1 - Faglioni, F. A1 - Bruno, Giovanni T1 - A novel iterative algorithm to improve segmentations with deep convolutional neural networks trained with synthetic X-ray computed tomography data (i.S.Sy.Da.T.A) N2 - We propose a novel iterative segmentation algorithm (i.S.Sy.Da.T.A: Iterative Segmentation Synthetic Data Training Algorithm) employing Deep Convolutional Neural Networks and synthetic training data for X-ray tomographic reconstructions of complex microstructures. In our method, we reinforce the synthetic training data with experimental XCT datasets that were automatically segmented in the previous iteration. This strategy produces better segmentations in successive iterations. We test our algorithm with experimental XCT re constructions of a 6-phase Al-Si Matrix Composite reinforced with ceramic fibers and particles. We perform the analysis in 3D with a special network architecture that demonstrates good generalization with synthetic training data. We show that our iterative algorithm returns better segmentations compared to the standard single training approach. More specifically, phases possessing similar attenuation coefficients can be better segmented: for Al2O3 fibers, SiC particles, and Intermetallics, we see an increase of the Dice score with respect to the classic approach: from 0.49 to 0.54, from 0.66 to 0.72, and from 0.55 to 0.66 respectively. Furthermore, the overall Dice score increases from 0.77 to 0.79. The methods presented in this work are also applicable to other materials and imaging techniques. KW - Metal matrix composites (MMC) KW - Multi-phase materials KW - 3D imaging KW - Dice score KW - Automatic segmentation KW - Deep convolutional neural network (DCNN) KW - Modified U-net architectures PY - 2023 DO - https://doi.org/10.1016/j.commatsci.2023.112112 SN - 0927-0256 VL - 223 SP - 1 EP - 8 PB - Elsevier B.V. AN - OPUS4-57482 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER -