Abstract.This paper proposes a hierarchical measurement model for silicon carbide (SiC) epilayer thickness based on infrared interferometry. Under a dual-beam interference framework, an analytical relationship between thickness and fringe spacing is established by integrating the Sellmeier dispersion model, resolving systematic errors caused by the constant refractive index assumption. For measured spectra, a five-step inversion algorithm is designed: Savitzky-Golay filtering suppresses noise, while robust spacing statistics combined with significant peak identification enable high-precision SiC … Read more →

Abstract.Non-invasive prenatal testing (NIPT) has gained widespread clinical adoption as an efficient and precise method for screening fetal chromosomal abnormalities. However, there remains room for optimization in balancing the timing of testing with diagnostic accuracy. This study employs multilevel modeling of clinical data from 267 pregnant women to propose a personalized testing timing optimization strategy based on multiple factors, including fetal cell-free DNA concentration and maternal body mass index (BMI). Using a mixed-effects model, we analyzed the relationship between BMI, gestational … Read more →

Abstract.Non-invasive prenatal testing (NIPT) screens for abnormalities in chromosomes 21, 18, and 13 by analyzing fetal cell-free DNA in maternal blood. Conducted between 10 and 25 weeks of gestation, it aids in early assessment of fetal health, with accuracy contingent upon male fetuses exhibiting Y chromosome concentrations ≥4% and female fetuses displaying normal X chromosome concentrations. This paper focuses on constructing a data model for NIPT data preprocessing. A quality-weighted linear regression model was developed to establish associations between fetal Y concentration and gestational … Read more →

Abstract.This paper systematically reviews the technological evolution, classification systems, and application scenarios of AI chips. Research shows AI chips have evolved from traditional general - purpose processors to a diverse ecosystem including GPUs, FPGAs, ASICs (e.g., TPUs, NPUs), and brain - inspired chips. By analyzing architectural features, algorithm adaptation, and performance limits of various chips, a multidimensional classification framework is constructed, categorizing AI chips from three dimensions: technical architecture, functional positioning, and application scenarios. The study … Read more →

Abstract.The sparse Mixture of Experts (MoE) architecture has evolved as a powerful approach for scaling deep learning models to more parameters with comparable computation cost. As an important branch of large language model (LLM), MoE model only activate a subset of experts based on a routing network. This sparse conditional computation mechanism significantly improves computational efficiency, paving a promising path for greater scalability and cost-efficiency. It not only enhance downstream applications such as natural language processing, computer vision, and multimodal in various horizontal … Read more →

Abstract.Colorectal cancer has been recognized as the third most prevalent and the third most lethal cancer worldwide. Accurate detection and segmentation of polyps in the colorectal region play a crucial role in facilitating the early diagnosis of cancer. The rapid advancement of deep learning has greatly accelerated the development of automated polyp detection and segmentation techniques. Although the well-established U-Net and its variants have long been regarded as classical architectures in medical image segmentation, the basic U-Net model still exhibits several limitations when applied to complex … Read more →