Explore the vast range of titles available.

Die casting is a suitable process for producing complex and high precision parts, but it faces challenges in terms of quality degradation due to inevitable defects. The casting parameters play a significant role in quality, and in many cases, producers rely on their experience to manage these parameters. In order to address this, domestic small and medium sized die casting companies have established smart factories (MES) and collected data. This study aims to utilize this data to construct a machine learning based optimal casting parameter model to enhance quality. During the model development process, distinct important features were identified for each company. This indicates the necessity of deriving tailored models for each site, aligning with the make to order (MTO) environment, rather than a generalized model.

To examine the association of depression after breast cancer diagnosis with long-term mortality risk in a Korean cohort. We conducted a retrospective population-based cohort study using data from the National Health Insurance Service (NHIS) cancer patient’s cohort of South Korea. We included women aged 40 years or older who were diagnosed with breast cancer between 2007 and 2013, survived at least 5 years, and had no history of depression prior to the breast cancer diagnosis. Depression was defined as hospitalization for more than 2 days with a primary diagnosis of depression (ICD-10 codes F32–F33). We evaluated all-cause, cancer-specific, and non-cancer-specific mortality using the Cox proportional hazards model while adjusting for covariates. Among 30,873 eligible women (mean age, 56.5 years), 502 were diagnosed with new-onset depression during the 5-year survival period, whereas 30,371 did not develop depression. During follow-up after the 5-year survivor period, a total of 1,904 deaths occurred. New-onset depression was associated with higher risks of all-cause mortality (adjusted hazard ratio [aHR], 1.38; 95% CI, 1.03–1.86; p  = 0.033) and non-cancer mortality (aHR, 1.81; 95% CI, 1.14–2.86; p  = 0.011), while no significant association was observed for cancer-specific mortality. The association was particularly pronounced among patients aged 65 years or older (aHR, 1.96; 95% CI, 1.27–3.03; p  = 0.002). Depression was linked to increased mortality among 5-year breast cancer survivors, especially for non-cancer causes. This study implies the need for depression screening and treatment among breast cancer patients, especially in non-Western settings where depression may be underdiagnosed and undertreated.

This paper presents a Cosine Similarity-Based Knowledge Distillation (CSKD) for robust, lightweight object detectors. Knowledge Distillation (KD) has been effective in enhancing the performance of compact models in image classification by leveraging deep CNN models. However, the complex and multifaceted nature of object detection, characterized by its modular design and multitasking requirements, poses significant challenges for traditional KD techniques. These challenges are further compounded by the conventional reliance on the Mean Squared Error (MSE) loss function and the limited application of enhanced feature representations to the training phase. Addressing these limitations, the proposed CSKD method combines cosine similarity guidance with MSE loss to facilitate a more effective knowledge transfer from the teacher model to the student model. This is achieved by distilling both intermediate features and prediction outputs, aided by an assistant prediction branch designed to learn directly from the teacher’s predictions. This dual-faceted distillation strategy enables the student model to better mimic the teacher model’s behavior, leading to improved performance. The proposed method demonstrates versatility and robustness across various object detector architectures without the need for additional feature enhancement layers during training. Notably, employing ResNet-50 as the teacher model and ResNet-18 as the student model, we achieve new benchmarks in KD for object detection across several architectures, including Faster-RCNN, RetinaNet, FCOS, and GFL, with respective mAP scores of 36.6, 35.2, 35.9, and 38.9. These results highlights the effectiveness of CSKD in advancing the state-of-the-art in KD for object detection, offering a compelling solution to the challenges previously faced by traditional KD methods in this domain. The code of the proposed CSKD is available at https://github.com/swkdn16/CSKD .

Simultaneous imaging of various facets of intact biological systems across multiple spatiotemporal scales is a long-standing goal in biology and medicine, for which progress is hindered by limits of conventional imaging modalities. Here we propose using the refractive index (RI), an intrinsic quantity governing light–matter interaction, as a means for such measurement. We show that major endogenous subcellular structures, which are conventionally accessed via exogenous fluorescence labelling, are encoded in three-dimensional (3D) RI tomograms. We decode this information in a data-driven manner, with a deep learning-based model that infers multiple 3D fluorescence tomograms from RI measurements of the corresponding subcellular targets, thereby achieving multiplexed microtomography. This approach, called RI2FL for refractive index to fluorescence, inherits the advantages of both high-specificity fluorescence imaging and label-free RI imaging. Importantly, full 3D modelling of absolute and unbiased RI improves generalization, such that the approach is applicable to a broad range of new samples without retraining to facilitate immediate applicability. The performance, reliability and scalability of this technology are extensively characterized, and its various applications within single-cell profiling at unprecedented scales (which can generate new experimentally testable hypotheses) are demonstrated. Jo et al. develop a broadly applicable deep-learning approach to predict fluorescence (FL) based on label-free refractive index (RI) measurements, ‘RI2FL’ (RI to FL). The trained model can be used across cell types without retraining.

In the semiconductor industry, achieving a high production yield is a very important issue. Wafer bin maps (WBMs) provide critical information for identifying anomalies in the manufacturing process. A WBM forms a certain defect pattern according to the error occurring during the process, and by accurately classifying the defect pattern existing in the WBM, the root causes of the anomalies that have occurred during the process can be inferred. Therefore, WBM defect pattern recognition and classification tasks are important for improving yield. In this paper, we propose a deep convolutional generative adversarial network (DCGAN)-based data augmentation method to improve the accuracy of a convolutional neural network (CNN)-based defect pattern classifier in the presence of extremely imbalanced data. The proposed method forms various defect patterns compared to the data augmentation method by using a convolutional autoencoder (CAE), and the formed defect patterns are classified into the same pattern as the original pattern through a CNN-based defect pattern classifier. Here, we introduce a new quantitative index called PGI to compare the effectiveness of the augmented models, and propose a masking process to refine the augmented images. The proposed method was tested using the WM-811k dataset. The proposed method helps to improve the classification performance of the pattern classifier by effectively solving the data imbalance issue compared to the CAE-based augmentation method. The experimental results showed that the proposed method improved the accuracy of each defect pattern by about 5.31% on average compared to the CAE-based augmentation method.

An efficient data-driven prediction strategy for multi-antenna frequency-selective channels must operate based on a small number of pilot symbols. This paper proposes novel channel-prediction algorithms that address this goal by integrating transfer and meta-learning with a reduced-rank parametrization of the channel. The proposed methods optimize linear predictors by utilizing data from previous frames, which are generally characterized by distinct propagation characteristics, in order to enable fast training on the time slots of the current frame. The proposed predictors rely on a novel long short-term decomposition (LSTD) of the linear prediction model that leverages the disaggregation of the channel into long-term space-time signatures and fading amplitudes. We first develop predictors for single-antenna frequency-flat channels based on transfer/meta-learned quadratic regularization. Then, we introduce transfer and meta-learning algorithms for LSTD-based prediction models that build on equilibrium propagation (EP) and alternating least squares (ALS). Numerical results under the 3GPP 5G standard channel model demonstrate the impact of transfer and meta-learning on reducing the number of pilots for channel prediction, as well as the merits of the proposed LSTD parametrization.

Treatment for malignant primary brain tumors, including glioblastoma, remains a significant challenge despite advances in therapy. CAR-T cell immunotherapy represents a promising alternative to conventional treatments. This review discusses the landscape of clinical trials for CAR-T cell therapy targeting brain tumors, highlighting key advancements like novel target antigens and combinatorial strategies designed to address tumor heterogeneity and immunosuppression, with the goal of improving outcomes for patients with these aggressive cancers.

Spiking neural networks (SNNs) are recurrent models that can leverage sparsity in input time series to efficiently carry out tasks such as classification. Additional efficiency gains can be obtained if decisions are taken as early as possible as a function of the complexity of the input time series. The decision on when to stop inference and produce a decision must rely on an estimate of the current accuracy of the decision. Prior work demonstrated the use of conformal prediction (CP) as a principled way to quantify uncertainty and support adaptive-latency decisions in SNNs. In this paper, we propose to enhance the uncertainty quantification capabilities of SNNs by implementing ensemble models for the purpose of improving the reliability of stopping decisions. Intuitively, an ensemble of multiple models can decide when to stop more reliably by selecting times at which most models agree that the current accuracy level is sufficient. The proposed method relies on different forms of information pooling from ensemble models and offers theoretical reliability guarantees. We specifically show that variational inference-based ensembles with p-variable pooling significantly reduce the average latency of state-of-the-art methods while maintaining reliability guarantees.

In addition to the known link between cholecystectomy and depression, the risk of developing short-term and long-term depression after surgery and whether such mental health issues leads to suicide were not known. Therefore, this study aimed to address these questions. Using data from the National Health Insurance Service of Korea (2002–2019), we conducted a retrospective cohort study including 6,688 cholecystectomy patients matched with 66,880 individuals without a history of cholecystectomy for suicide analysis and 6,694 cholecystectomy patients matched with 66,940 individuals for depression analysis. The non-cholecystectomy group was matched at a 1:10 ratio for sex and age. The incidence of depression and suicide were followed from the day of cholecystectomy to December 31, 2019. Adjusted hazard ratios (aHRs) and 95% confidence intervals (CIs) were estimated using multivariable Cox proportional hazards regression. Short-term depression risk within three years of cholecystectomy was significantly elevated (aHR 1.38, 95% CI 1.19–1.59), while the long-term depression risk beyond three years was not significantly greater (aHR 1.09, 95% CI 0.98–1.22). Cholecystectomy was not associated with an increased risk of suicide in any period. These findings highlight the importance of monitoring and providing postoperative mental health support for patients at risk of short-term depression after cholecystectomy. However, no association was observed with long-term depression or suicide risk.

With the recent rapid development of IT technology, the demand for multifunctional semiconductor devices capable of high performance has increased rapidly, and the miniaturization of such devices has also faced limitations. To overcome these limitations, various studies have investigated three-dimensional packaging methods of stacking devices, and among them, hybrid bonding is being actively conducted during the bonding process. studies of hybrid bonding during the bonding process are active. In this study, Cu bonding using a nano passivation layer was carried out for Cu/SiO2 hybrid bonding applications, with Au and Ag deposited on Cu at the nano level and used as a protective layer to prevent Cu oxidation and to achieve low-temperature Cu bonding. Au was deposited at about 12 nm, and Ag was deposited at about 15 nm, with Cu bonding carried out at 180 °C for 30 min, after which an annealing process was conducted at 200 °C for one hour. After bonding, the specimen was diced into a 1 cm × 1 cm chip, and the bonding interface was analyzed using SEM and TEM. Additionally, the 1 cm × 1 cm chip was diced into 2 mm × 2 mm specimens to measure the shear strength of the bonded chip, and the average shear strength of Au and Ag was found to be 5.4 and 6.6 MPa, respectively. The degree of diffusion between Au-Cu and Ag-Cu was then investigated; the diffusion activation energy when Au diffuses to Cu was 6369.52 J/mol, and the diffusion activation energy when Ag diffuses to Cu was 17,933.21 J/mol.