The problem of artifacts in whole slide image acquisition, prevalent in both clinical workflows and research-oriented settings, necessitates human intervention and re-scanning. Overcoming this challenge requires devel...
详细信息
High annotation costs serve as a significant hurdle in deploying modern deep learning architectures for clinically relevant medical applications, especially when dealing with the inherent heterogeneity of multimodal d...
High annotation costs serve as a significant hurdle in deploying modern deep learning architectures for clinically relevant medical applications, especially when dealing with the inherent heterogeneity of multimodal data, proving the critical need for innovative algorithms that can effectively utilize unlabeled data. In this paper, we propose a model named MCLCA, which integrates multimodal contrastive learning and cross-modal attention to diagnose Alzheimer’s Disease (AD) and identify biomarkers using both labeled and unlabeled multimodal brain imaging genetics data. Through multimodal contrastive learning, MCLCA can effectively learn representations even in the absence of sufficient labels. By utilizing cross-modal attention blocks, the model captures deep connections between different modalities, providing a more comprehensive view of diagnosis. Our proposed MCLCA model is evaluated using the ADNI database with three imaging modalities (VBM-MRI, FDG-PET, and AV45-PET) and genetic SNP data. The results demonstrate that MCLCA can identify important biomarkers with better prediction accuracy compared to the existing methods. The source code is available at https://***/MCLCA.
Few studies have used the combination of subjective and objective measures to investigate sleep problems as a primary outcome of concern in cancer patients. This study highlights the influence of sleep quality and dur...
详细信息
In distributed optimization schemes consisting of a group of agents connected to a central coordinator, the optimization algorithm often involves the agents solving private local sub-problems and exchanging data frequ...
详细信息
The Deep-in-Biopsies (DiB) platform supports diagnosing and staging diseases using patient biopsy imaging. This flexible service is designed for medical units to handle multiple organs and diseases. Specialists can tr...
The Deep-in-Biopsies (DiB) platform supports diagnosing and staging diseases using patient biopsy imaging. This flexible service is designed for medical units to handle multiple organs and diseases. Specialists can train the system on demand using their selected images and annotations, which can be generated using an embedded annotation tool. The system proposes pathology findings to the expert for validation, and validated findings can be used for further training.
Compared with the terrestrial network, the air-ground integrated network consisting of unmanned aerial ve-hicles (UAV s) and high altitude platforms (HAPs) offers the advantages of large coverage, high capacity, and s...
Compared with the terrestrial network, the air-ground integrated network consisting of unmanned aerial ve-hicles (UAV s) and high altitude platforms (HAPs) offers the advantages of large coverage, high capacity, and seamless connection, which can provide effective communication services for the Internet of remote things (IoRT). In this paper, we investigate the joint packet scheduling and UAV trajectory design problem, with the objective of minimizing the average packet queue delay of IoRT in the air-ground integrated network. Since the problem is a non-convex in nature, we reformulate it into a Markov decision process (MDP) firstly. Then, based on the multi-agent deep deterministic policy gradient (MADDPG) method, we propose a joint packet scheduling and UAV trajectory design (JPSTD) algorithm. Simulation results show that the proposed JPSTD algorithm outperforms the benchmark algorithms in terms of reducing the average packet queue delay.
Robots are more capable of achieving manipulation tasks for everyday activities than before. But the safety of manipulation skills that robots employ is still an open problem. Considering all possible failures during ...
详细信息
Digital twin technology becomes an appropriate respond to the new challenge of rapid development in science and industry. Chemical data include the information about structure and properties of materials and compounds...
详细信息
The objective of BioCreative8 Track 3 is to extract phenotypic key medical findings embedded within EHR texts and subsequently normalize these findings to their Human Phenotype Ontology (HPO) terms. However, the prese...
详细信息
With the exponentially faster computation for certain problems, quantum computing has garnered significant attention in recent years. Variational quantum algorithms are crucial methods to implement quantum computing, ...
详细信息
暂无评论