In steganography, the cover medium takes on various types, including images, videos. Due to the advent of deep learning, neural network models are increasingly employed as the cover medium. In existing approaches, the embedding of secret data results in an obvious degradation of the model's original task performance, such as image classification. This paper proposes a low-distortion steganography method that embeds secret data into neural network models without degrading the model's original performance. The method selectively modifies parameters with minimal correlation to the loss function, ensuring that the model's performance remains unaffected. Furthermore, a suitable modification amplitude is defined to minimize the impact on task performance. Experimental results demonstrate that the proposed method enables low-distortion steganography while significantly improving embedding capacity and security. Notably, a maximum of 3.25 M bits of secret data is successfully embedded into a VGG model on the MNIST dataset, resulting in less than 0.1% accuracy degradation, a significant improvement over previous methods.

This paper introduces MedTrinity-25M, a comprehensive, large-scale multimodal dataset for medicine, covering over 25 million images across 10 modalities with multigranular annotations for more than 65 diseases. These multigranular annotations encompass both global information, such as modality and organ detection, and local information like ROI analysis, lesion texture, and region-wise correlations. Unlike the existing multimodal datasets, which are limited by the availability of image-text pairs, we have developed the first automated pipeline that scales up multimodal data by generating multigranular visual and textual annotations in the form of image-ROI-description triplets without the need for any paired text descriptions. Specifically, data from over 30 different sources have been collected, preprocessed, and grounded using domain-specific expert models to identify ROIs related to abnormal regions. We then build a comprehensive knowledge base and prompt multimodal large language models to perform retrieval-augmented generation with the identified ROIs as guidance, resulting in multigranular textual descriptions. Compared to existing datasets, MedTrinity-25M provides the most enriched annotations, supporting a comprehensive range of multimodal tasks such as captioning and report generation, as well as vision-centric tasks like classification and segmentation. We propose LLaVA-Tri by pretraining LLaVA on MedTrinity-25M, achieving state-of-the-art performance on VQA-RAD, SLAKE, and PathVQA, surpassing representative SOTA multimodal large language models. Furthermore, MedTrinity-25M can also be utilized to support large-scale pre-training of multimodal medical AI models, contributing to the development of future foundation models in the medical domain. We will make our dataset available. The dataset is publicly available at https://yunfeixie233.github.io/MedTrinity-25M/.

Construct a Knowledge Graph based on the characteristics of Dao Yin Science in Traditional Chinese Medicine to facilitate users in querying, learning, and disseminating the culture of health maintenance in Traditional Chinese Medicine, and to enhance their health literacy. Using the Bert-CRF model, with manual proofreading, we conducted named entity recognition on literature related to Dao Yin practice, established relationships between entities based on the characteristics of the subject and built a Knowledge Graph for Dao Yin Science using Neo4j. Incorporating 3152 relevant documents, the constructed knowledge graph contains 2262 entity nodes, 5108 relationship data, 7 types of entity attribute categories, and 7 types of relationships. The knowledge graph can be searched using the Cypher query language of Neo4j. This Knowledge Graph visualizes the knowledge of Dao Yin Science in Traditional Chinese Medicine and can be employed to build applications and web platforms for knowledge retrieval, intelligent question answering, and exercise recommendation. It holds practical value for developing related artificial intelligence devices and fostering innovation in the fields of computer science and automation. Moreover, it can be utilized for in-depth data mining and uncovering new research directions.

In order to protect information privacy and ensure user information security, in view of the obvious centralization of the existing identity authentication technologies such as Public Key Infrastructure(PKI) and Identity-Based Encrypted(IBE), this paper proposes an efficient authentication strategy that applies Cryptography Fundamental Logics(CFL) identity authentication technology to Mobile Crowd Sensing(MCS) system, which can complete the authentication between Task Publisher, Cluster Head and Task Participant without the participation of a third-party center. Firstly, this paper introduces to use CFL technology to solve the problem of identity authentication relying on the central server; Secondly, an algorithm combined with MCS system is proposed to solve the decentralization of authentication process; Finally, the Average System Response Time and System Throughput of the three technologies are obtained through simulation experiments, analyzed and compared. The result shows that: this strategy has obvious advantages, it can faster and more secure the identity authentication.