Sparse-view Reconstruction can be used to provide accelerated low dose CT imaging with both accelerated scan and reduced projection/back-projection calculation. Despite the rapid developments, image noise and artifacts still remain a major issue in the low dose protocol. In this paper, a deep learning based method named Improved GoogLeNet is proposed to remove streak artifacts due to projection missing in sparse-view CT reconstruction. Residual learning is used in GoogLeNet to study the artifacts of sparse-view CT reconstruction, and then subtracts the artifacts obtained by learning from the sparse reconstructed images, finally recovers a clear correction image. The intensity of reconstruction using the proposed method is very close to the full-view projective reconstructed image. The results indicate that the proposed method is practical and effective for reducing the artifacts and preserving the quality of the reconstructed image.

The influence of fuel pressure fluctuation on multi-injection fuel mass deviation has been studied a lot, but the fuel pressure fluctuation at injector inlet is still not eliminated efficiently. In this paper, a new type of hydraulic filter consisting of a damping hole and a chamber is developed for elimination of fuel pressure fluctuation and multi-injection fuel mass deviation. Linear model of the improved high pressure common-rail system(HPCRS) including injector, the pipe connecting common-rail with injector and the hydraulic filter is built. Fuel pressure fluctuation at injector inlet, on which frequency domain analysis is conducted through fast Fourier transformation, is acquired at different target pressure and different damping hole diameter experimentally. The linear model is validated and can predict the natural frequencies of the system. Influence of damping hole diameter on fuel pressure fluctuation is analyzed qualitatively based on the linear model, and it can be inferred that an optimal diameter of the damping hole for elimination of fuel pressure fluctuation exists. Fuel pressure fluctuation and fuel mass deviation under different damping hole diameters are measured experimentally, and it is testified that the amplitude of both fuel pressure fluctuation and fuel mass deviation decreases first and then increases with the increasing of damping hole diameter. The amplitude of main injection fuel mass deviation can be reduced by 73% at most under pilot-main injection mode, and the amplitude of post injection fuel mass deviation can be reduced by 92% at most under main-post injection mode. Fuel mass of a single injection increases with the increasing of the damping hole diameter. The hydraulic filter proposed by this research can be potentially used to eliminate fuel pressure fluctuation at injector inlet and improve the stability of HPCRS fuel injection.

Within the domain of Artificial Intelligence Generated Content (AIGC), technological strides in image generation have been marked, resulting in the proliferation of deepfake images that pose substantial security threats. The current landscape of deepfake detection technologies is marred by limited generalization across diverse generative models and a subpar detection rate for images generated through diffusion processes. In response to these challenges, this paper introduces a novel detection model designed for high generalizability, leveraging multiscale frequency and spatial domain features. Our model harnesses an array of specialized filters to extract frequency‐domain characteristics, which are then integrated with spatial‐domain features captured by a Feature Pyramid Network (FPN). The integration of the Attentional Feature Fusion (AFF) mechanism within the feature fusion module allows for the optimal utilization of the extracted features, thereby enhancing detection capabilities. We curated an extensive dataset encompassing deepfake images from a variety of GANs and diffusion models for rigorous evaluation. The experimental findings reveal that our proposed model achieves superior accuracy and generalization compared to existing baseline models when confronted with deepfake images from multiple generative sources. Notably, in cross‐model detection scenarios, our model outperforms the next best model by a significant margin of 29.1% for diffusion‐generated images and 15.1% for GAN‐generated images. This accomplishment presents a viable solution to the pressing issues of generalization and adaptability in the field of deepfake detection.

Objectives: In order to assist junior doctors in better diagnosing apical periodontitis (AP), an artificial intelligence AP grading system was developed based on deep learning (DL) and its reliability and accuracy were evaluated. Methods: One hundred and twenty cone-beam computed tomography (CBCT) images were selected to construct a classification dataset with four categories, which were divided by CBCT periapical index (CBCTPAI), including normal periapical tissue, CBCTPAI 1-2, CBCTPAI 3-5, and young permanent teeth. Three classic algorithms (ResNet50/101/152) as well as one self-invented algorithm (PAINet) were compared with each other. PAINet were also compared with two recent Transformer-based models and three attention models. Their performance was evaluated by accuracy, precision, recall, balanced F score (F1-score), and the area under the macro-average receiver operating curve (AUC). Reliability was evaluated by Cohen’s kappa to compare the consistency of model predicted labels with expert opinions. Results: PAINet performed best among the four algorithms. The accuracy, precision, recall, F1-score, and AUC on the test set were 0.9333, 0.9415, 0.9333, 0.9336, and 0.9972, respectively. Cohen’s kappa was 0.911, which represented almost perfect consistency. Conclusions: PAINet can accurately distinguish between normal periapical tissues, CBCTPAI 1-2, CBCTPAI 3-5, and young permanent teeth. Its results were highly consistent with expert opinions. It can help junior doctors diagnose and score AP, reducing the burden. It can also be promoted in areas where experts are lacking to provide professional diagnostic opinions.

Objectives: This study was aimed to investigate the accuracy of soft-tissue measurements obtained by two imaging modalities, three-dimensional (3D) photogrammetry and cone beam CT (CBCT) when confounded by influence factors (facial deformities and partitions). Methods: 60 wax facial models from facially deformed patients were captured by 3D photogrammetry and CBCT. 19 linear distances on each image were measured and juxtaposed to reference values attained via a coordinate-measuring machine (CMM) as the gold-standard. Paired t-tests were used to compare linear accuracy of the test and reference systems. The influence of deformities and partitions (created by dividing the face with three vertical and five horizontal lines) on the measurement errors were analyzed by independent sample t-test and one-way ANOVA. Results: Statistically significant differences were found between linear accuracy of the test and reference systems. The test values obtained by 3D photogrammetry were closer to the reference values than CBCT’s. 3D photogrammetry’s measurement errors were significantly higher in deformed areas, unlike CBCT’s. Both systems reported significantly lower errors within partitions 8 and 13 compared to other partitions; for CBCT, aside from partitions 8 and 13, the differences in the errors for partitions 6 and 10 were significant compared to partitions 8, 12, 13, 14. Conclusion: 3D photogrammetry showed a higher linear accuracy than CBCT in patients with facial deformities due to protuberances. Facial reconstruction by both test modalities was significantly influenced in different facial partitions, but facial deformities extensively affected the results from 3D photogrammetry.