As one of the cancer types with the highest incidence rates, colorectal cancer (CRC) would benefit from treatments with fewer side effects and reduced treatment-resistant potential. One of the options is to harness th...
详细信息
As one of the cancer types with the highest incidence rates, colorectal cancer (CRC) would benefit from treatments with fewer side effects and reduced treatment-resistant potential. One of the options is to harness the anti-CRC potential of natural products. Previous studies have shown that Calamus draco exudate, dragon's blood, has anticancer activity in liver cancer and acute myeloid leukemia, but its bioactivity has not been studied in CRC. Here we conduct a bioinformatics study based on network pharmacology to explore the anti-CRC potential and mechanism of C. draco -derived compounds. The bioinformatics pipeline is composed of compound and target collection, biological network evaluation, and enrichment analysis. We found that there are 43 bioactive compounds from C. draco targeting 91 CRC-related targets, of which most compounds target MEN1, PTGS2, and IDH1. Further analyses show that the targets of C. draco are involved in the cellular response to hypoxia. By inhibiting those targets, C. draco bioactive compounds can potentially hinder angiogenesis and increase treatment response efficacy.
Intrusion Detection systems (IDS) are essential for safeguarding IoT networks against various attacks. Our previously developed ensemble-based IDS model, which combines stacked Long Short-Term Memory (LSTM) networks w...
详细信息
ISBN:
(数字)9798350355079
ISBN:
(纸本)9798350355086
Intrusion Detection systems (IDS) are essential for safeguarding IoT networks against various attacks. Our previously developed ensemble-based IDS model, which combines stacked Long Short-Term Memory (LSTM) networks with ANOVA feature selection, achieved a remarkable accuracy of 97.46% and an overfitting value of 0.22% in classifying network attacks. However, deploying this powerful model on resource-constrained devices like Raspberry Pi poses challenges due to limited computational resources. In this paper, we tackle these challenges by converting the original model to TensorFlow Lite (TFLite) and applying quantization techniques to optimize the model size for deployment on a Raspberry Pi 3b+. Our approach significantly reduces the model size by 90.9% while maintaining an impressive accuracy of 96.96%, only a 0.5% reduction from the original model. The optimized model demonstrates efficient performance on the Raspberry Pi, with a loading time of 100.2ms and an inference time of 1.2ms.
Generative Artificial Intelligence (GenAI) represents a significant milestone in the development of artificial intelligence, bringing sophisticated AI capabilities into daily life and work. As we approach the era of H...
详细信息
ISBN:
(数字)9798331532093
ISBN:
(纸本)9798331532109
Generative Artificial Intelligence (GenAI) represents a significant milestone in the development of artificial intelligence, bringing sophisticated AI capabilities into daily life and work. As we approach the era of Hyper Intelligence (Hyper-I), a variety of critical challenges and emerging issues have come to light, ranging from computational complexity to ethical concerns. This paper explores the evolution of AI from the perspective of human learning, comparing machine and human intelligence, and identifying key considerations for the development of future AI systems. It also highlights the growing importance of regulating advanced AI models, such as Reinforcement Learning-based Long-Term Planning Agents, to ensure that Hyper-I remains under human control. Additionally, the paper discusses the computational complexity of transformer-based models, their applicability to intractable problems, and their role in cognitive building systems and resource-constrained environments through TinyML. By analyzing these pressing challenges, this work provides insights into the future of AI and the path toward responsible innovation in generative and hyper-intelligent systems.
The spin Seebeck effect (SSE) is sensitive to thermally driven magnetic excitations in magnetic insulators. Vanadium dioxide in its insulating low-temperature phase is expected to lack magnetic degrees of freedom, as ...
详细信息
The spin Seebeck effect (SSE) is sensitive to thermally driven magnetic excitations in magnetic insulators. Vanadium dioxide in its insulating low-temperature phase is expected to lack magnetic degrees of freedom, as vanadium atoms are thought to form singlets upon dimerization of the vanadium chains. Instead, we find a paramagnetic SSE response in VO2 films that grows as the temperature decreases below 50 K. The field and temperature-dependent SSE voltage is qualitatively consistent with a general model of paramagnetic SSE response and inconsistent with triplet spin transport. Quantitative estimates find a spin Seebeck coefficient comparable in magnitude to that observed in strongly magnetic materials. The microscopic nature of the magnetic excitations in VO2 requires further examination.
Semantic segmentation enables high-accuracy object classification by assigning a class label to each pixel in an image. However, creating training data for segmentation is labor-intensive as it involves labeling every...
详细信息
ISBN:
(数字)9798350355079
ISBN:
(纸本)9798350355086
Semantic segmentation enables high-accuracy object classification by assigning a class label to each pixel in an image. However, creating training data for segmentation is labor-intensive as it involves labeling every pixel. To alleviate the training cost, computer graphics (CG) datasets are considered as an efficient alternative solution. However, models trained on CG datasets may induce a significant drop in the intersection over union (IoU) when applied to real-world images due to domain differences. In this work, semi-supervised domain adaptation (SSDA) and unsupervised domain adaptation (UDA) training methods are adopted to achieve efficient segmentation models for intersection images. SSDA utilizes a small amount of labeled data from the target domain, while UDA does not use any labeled data from the target domain. Simulation results show that SSDA achieves similar accuracy to supervised learning across most classes, and UDA achieves comparable accuracy to supervised learning in certain classes.
With the spread of high-resolution video such as 8K and 360-degree videos, efficient video coding methods is still highly required. The intra prediction method of versatile video coding (VVC) achieves efficient coding...
详细信息
ISBN:
(数字)9798350379051
ISBN:
(纸本)9798350379068
With the spread of high-resolution video such as 8K and 360-degree videos, efficient video coding methods is still highly required. The intra prediction method of versatile video coding (VVC) achieves efficient coding by referencing neighboring pixel values based on predefined directional modes. However, it is difficult to further improve video coding efficiency using traditional linear models in the intra prediction mode. In this work, we propose a method to increase the coding efficiency by applying the image restoration technique. Specifically, image restoration technique uses Conditional-UNet combined with a stochastic differential equation (SDE) to generate highly accurate predicted pixels. The simulation results show that the proposed algorithm can achieve an improvement of 29.14 % on BD-rate compared to the original VVC algorithm at random access. Moreover, the proposed algorithm reduces the training and inference time through optimized parameters.
In this work, a video communication system to encode river video at a low bit rate is proposed. In the proposed system, a semantic segmantation is performed to organize the coding priority of each segment. Then, by se...
详细信息
ISBN:
(数字)9798350379051
ISBN:
(纸本)9798350379068
In this work, a video communication system to encode river video at a low bit rate is proposed. In the proposed system, a semantic segmantation is performed to organize the coding priority of each segment. Then, by setting an appropriate Quantization Parameter (QP) map for each region with VVenC which is the latest video codec of VVC, a low bitrate coding is achieved. We also compare the encoding performance with the latest Neural Video Codec (NVC), namely, DCVC-DC. The simulation results show that the proposed method with VVenC achieves ultra-low bit coding. Although the current version of DCVC-DC could not achieve ultra-low bit coding, the coding efficiency shows higher performance than that using VVenC in a wide range of bit rate.
In recent applications, a modern object recognition model is available together with the video encoder. In this work, an adaptive bitrate control algorithm is proposed using a You Only Look Once v8 (YOLOv8) model for ...
详细信息
ISBN:
(数字)9798350355079
ISBN:
(纸本)9798350355086
In recent applications, a modern object recognition model is available together with the video encoder. In this work, an adaptive bitrate control algorithm is proposed using a You Only Look Once v8 (YOLOv8) model for Versatile Video Coding (VVC). YOLOv8 object detection network, which is also suitable for real-time processing, is used as an algorithm to detect salient regions before encoding. The proposed algorithm estimates Rate-Distortion Cost (RD-Cost) of Coding Tree Units (CTUs) with YOLOv8 and selects the Quantization Parameter (QP) based on the object detection result. Simulation results show that the proposed algorithm achieves a maximum bitrate reduction rate of 7.76% and improves coding efficiency while minimizing image quality degradation.
multidisciplinary collaboration between public health, system engineering, and UX is able to generate a solution in healthcare problem like stunting. The principle of Agile UX gathers requirements to generate an appli...
详细信息
Unmanned aerial vehicles (UAVs) have remarkably advanced and expected to be useful in a variety of fields. However, there is a serious problem of video quality degradation due to rainy weather, regardless of whether t...
详细信息
ISBN:
(数字)9798350379051
ISBN:
(纸本)9798350379068
Unmanned aerial vehicles (UAVs) have remarkably advanced and expected to be useful in a variety of fields. However, there is a serious problem of video quality degradation due to rainy weather, regardless of whether the UAVs are flying autonomously. In this work, we propose a highly accurate method to remove the effect of rain streaks from images captured by UAVs in rainy conditions. This approach uses the multi-axis feature fusion (MFF) block including the kernel basis attention (KBA) module. The proposed method shows better performance than the previous deraining methods in PSNR and SSIM. Moreover, the proposed method is effective for varying rain streak intensity.
暂无评论