Search

• Machine learning models tailored for compressive strength of GD-reinforced CCs (GDRCCs). • Unique dataset covering GD properties and fabrication details of GDRCCs was compiled. • Artificial neural network excelled in prediction performance among investigated models. • The influencing factors on the compressive strength of GDRCCs were analyzed. • GDs lateral size, together with w/c and curing time, were the most influential factors. The potential reinforcement effect of graphene derivatives (GDs) on cementitious composites (CCs) has attracted significant attention. Previous studies, however, have produced varied results regarding the impact of GDs on CCs. This can be attributed to differences in the properties of GDs and the fabrication details of CCs reinforced by GDs. Experiments to explore these factors are both time-consuming and cost-ineffective. Additionally, no predictive model currently exists for assessing the influence of GDs on the compressive strength of CCs. In terms of Machine Learning (ML), most existing models focus on continuous parameters, including mixture design properties of CCs and reinforcing filler content, but ignore discontinuous parameters such as dispersion technique of GDs in CCs, curing type, and type of GDs. Compiling a unique dataset, this study tailors ML models to comprehensively explore the effect of GDs inclusion on the compressive strength of CCs, considering continuous and discontinuous parameters, including GD properties, fabrication details, and mixture design properties. The most used dispersion techniques and types of GDs were divided into different categories in this study. Moreover, the dataset included cement strength grade and fineness modules to distinguish between the effect of cement types' variety and GDs. Finally, the backwards elimination technique confirmed the necessity of such a customized dataset for trustworthy predictions. Artificial neural networks (ANN), decision trees, and support vector regressors could successfully investigate the impact of GDs' inclusion on the compressive strength of CCs, with ANN demonstrating superior prediction performance. Among the GDs properties, sensitivity analysis revealed that lateral size had the highest effect. Among the fabrication conditions, the dispersion technique had the greatest effect. Considering all investigated parameters, the water-to-cement ratio was considered the most influential, followed by lateral size and curing time. A low w/c significantly reduces the strength growth rate due to poor dispersion of GDs. [ABSTRACT FROM AUTHOR]

Since the labels for medical images are challenging to collect in real scenarios, especially for rare diseases, fully supervised methods cannot achieve robust performance for clinical anomaly detection. Recent research tried to tackle this problem by training the anomaly detection framework using only normal data. Reconstruction-based methods, e.g., auto-encoder, achieved impressive performances in the anomaly detection task. However, most existing methods adopted the straightforward backbone architecture (i.e., encoder-and-decoder) for image reconstruction. The design of a skip connection, which can directly transfer information between the encoder and decoder, is rarely used. Since the existing U-Net has demonstrated the effectiveness of skip connections for image reconstruction tasks, in this paper, we first use the dynamic gating strategy to achieve the usage of skip connections in existing reconstruction-based anomaly detection methods and then propose a novel gating highway connection module to adaptively integrate skip connections into the framework and boost its anomaly detection performance, namely GatingAno. Furthermore, we formulate an auxiliary task, namely histograms of oriented gradients (HOG) prediction, to encourage the framework to exploit contextual information from fundus images in a self-driven manner, which increases the robustness of feature representation extracted from the healthy samples. Last but not least, to improve the model generalization for anomalous data, we introduce an adversarial strategy for the training of our multi-task framework. Experimental results on the publicly available datasets, i.e., IDRiD and ADAM, validate the superiority of our method for detecting abnormalities in retinal fundus images. The source code is available at https://github.com/WentianZhang-ML/GatingAno. • A gating highway connection module is proposed, which can be applied in a reconstruction-based anomaly detection method to exploit and constrain the intrinsic features during training for better anomaly detection performance. • We integrate histograms of oriented gradients (HOG) prediction as an auxiliary task in the reconstruction network to improve its sensitivity to abnormal images. • Adversarial learning is combined in our multi-task encoder–decoder network in the training phase to improve generalization ability for anomaly detection. • We evaluate our model on two publicly available datasets. The experimental results on image-level and pixel-level tasks prove the effectiveness of gating highway connections, HOG prediction, and adversarial learning for anomaly detection. [ABSTRACT FROM AUTHOR]

In current Optical Coherence Tomography (OCT)-based fingerprint recognition systems, Presentation Attack Detection (PAD) and subsurface fingerprint reconstruction are treated as two independent branches, resulting in high computation and system building complexity. Therefore, this paper proposes a uniform representation model for simultaneous PAD and subsurface fingerprint reconstruction. A novel semantic segmentation network using attention mechanisms was designed to extract and segment multiple subsurface structures from real finger slices (i.e., B-scans). The latent codes derived from the network are directly used to effectively detect PA because they contain abundant subsurface biological information, which is independent of PA materials and has strong robustness for unknown PAs. Segmented subsurface structures were adopted to reconstruct multiple subsurface 2D fingerprints. Extensive experiments were carried out on an in-house database, which is the largest public OCT-based fingerprint database with 2449 volumes. PAD performance was evaluated by comparing the results of the existing methods with those of other segmentation networks. The proposed uniform representation model can obtain an accuracy (Acc) of 96.63%, which achieves a state-of-the-art performance. The effectiveness of subsurface reconstruction was evaluated from the segmentation and recognition results. In the segmentation experiments, the proposed method achieved the best results with an Intersection of Union (mIOU) of 0.834 and a Pixel Accuracy of 0.937. By comparing the recognition performance on surface 2D fingerprints (e.g., commercial and high-resolution), the lowest results with an Equal Error Rate (EER) of 2.25% by minutiae matching and an EER of 5.42% by pore matching are achieved, which indicates the excellent reconstruction capability of the proposed uniform representation model. • A uniform representation model for OCT-based fingerprint PAD and reconstruction. • A novel semantic segmentation network applying biological structure features. • High security PAD is achieved without using any presentation attacks for training. • Reliable reconstruction result is obtained with subsurface biological information. [ABSTRACT FROM AUTHOR]