Fast and accurate landmark location and bounding box detection are important steps in 3D medical imaging. In this paper, we propose a novel multi-task learning framework, for real-time, simultaneous landmark location ...
详细信息
Fast and accurate landmark location and bounding box detection are important steps in 3D medical imaging. In this paper, we propose a novel multi-task learning framework, for real-time, simultaneous landmark location and bounding box detection in 3D space. Our method extends the famous single-shot multibox detector (SSD) from single-task learning to multitask learning and from 2D to 3D. Furthermore, we propose a post-processing approach to refine the network landmark output, by averaging the candidate landmarks. Owing to these settings, the proposed framework is fast and accurate. For 3D cardiac magnetic resonance (MR) images with size 224×224×64, our framework runs ~128 volumes per second (VPS) on GPU and achieves 6.75mm average point-to-point distance error for landmark location, which outperforms both state-of-the-art and baseline methods. We also show that segmenting the 3D image cropped with the bounding box results in both improved performance and efficiency.
An improved algorithm based on Quick Sort algorithm research method is proposed to deal with prevailing duplicate values in the sorting of data. The duplicate values are specially processed, which effectively reduces ...
详细信息
ISBN:
(数字)9798350376425
ISBN:
(纸本)9798350376432
An improved algorithm based on Quick Sort algorithm research method is proposed to deal with prevailing duplicate values in the sorting of data. The duplicate values are specially processed, which effectively reduces the number of division and greatly reduces the number of recursions. The experimental results show that the time efficiency is improved by more than 50% when dealing with large data than the traditional Quick Sort algorithm without reducing the space efficiency.
Deep reinforcement learning (DRL) performance is generally impacted by state-adversarial attacks, a perturbation applied to an agent’s observation. Most recent research has concentrated on robust single-agent reinfor...
详细信息
The widespread adoption of cloud storage enables users to remotely access resources through a self-service model. Utilizing pay-per-use storage services provided by cloud service providers (CSPs) requires users to com...
详细信息
ISBN:
(数字)9798350366440
ISBN:
(纸本)9798350366457
The widespread adoption of cloud storage enables users to remotely access resources through a self-service model. Utilizing pay-per-use storage services provided by cloud service providers (CSPs) requires users to commit financially to their resources. This paper introduces a Secure Cloud Storage (SCS) framework, offering a secure architecture for cloud storage using a consortium blockchain network to address trust issues. This framework substitutes the third-party auditor with peers of a consortium blockchain network, which handles the role of data storage and verification. Storage space is divided into uncommitted and committed segments. Uncommitted storage is used for storing unverified documents, while committed storage is reserved for documents that have been validated through a consensus mechanism. In contrast, committed storage is des-ignated for the storage of committed documents. Documents validated by a consensus threshold of peer nodes are moved from uncommitted to committed storage. The implementation of the SCS framework is conducted using Hyperledger Fabric, a modular blockchain platform optimized for permissioned networks. The security analysis demonstrates that SCS effectively protects cloud storage against attacks, including unauthorized access attacks, data integrity attacks, and malicious server attacks, while maintaining data integrity and auditability. The performance evaluation shows that document upload and retrieval times, block acceptance, execution times, and latency are all improved compared to state-of-the-art cloud storage techniques.
This paper provides a description of the programming language Pascal. It has been published to enable those without easy access to the official BSI ‘draft for comment’ to comment on the description.
This paper provides a description of the programming language Pascal. It has been published to enable those without easy access to the official BSI ‘draft for comment’ to comment on the description.
Recently, hyperbolic spaces have proven beneficial for service recommendation due to their exponentially growing spatial properties conforming to power-law distributed user-item networks. Among them, the combination o...
详细信息
ISBN:
(数字)9798350368550
ISBN:
(纸本)9798350368567
Recently, hyperbolic spaces have proven beneficial for service recommendation due to their exponentially growing spatial properties conforming to power-law distributed user-item networks. Among them, the combination of hyperbolic space with graph convolution has achieved great success. However, hyperbolic convolutional models still perform multi-layer convolution in tangent space (Euclidean space), leading to the still inevitable problem of over-smoothing arising from multi-layer convolution. In addition, most of these models randomly draw negative samples from items that users have not interacted with, so that some of the samples obtained may not be well suited for model optimization. To tackle the above challenges, we propose a new Hyperbolic GCN model based on Contrastive Learning and Second-order Reachable sampling for collaborative filtering (HG-CLSR), which improve high quality of representations by exploring the distribution of users and items in hyperbolic space. Specifically, We first introduce a root alignment approach to encourage embeddings to align with the tangent space, thereby reducing distortions during the embedding mapping process in space. Then, we perform contrastive learning in hyperbolic space to motivate the spatial distribution of nodes to better fit the hyperbolic space. Moreover, we sample in the user’s second-order reachable item set, which ensures that the negative sample is more similar to the positive sample, so that the negative node can provide better information for guiding model optimization. Extensive experiments on three real-world datasets demonstrate that the HG-CLSR is significantly superior compared to existing hyperbolic models.
In traditional multiple instance learning (MIL), both positive and negative bags are required to learn a prediction function. However, a high human cost is needed to know the label of each bag - positive or negative. ...
暂无评论