To address challenges such as diverse clinical data sources, inconsistent standards, and difficulties in sharing, this study developed a clinical data standardization system based on medical data element and Large Language Model (LLM). 1. Customizable patient and medical record templates were created using medical data element and related terminology, enabling dynamic binding for data structuring and standardization. 2. LLM was leveraged to perform intelligent information extraction, semantic normalization, and standardization of unstructured or semi-structured clinical text. 3. User permissions were configured to manage medical record data across both record repositories and institutions, supporting cross-repository and cross-institutional export of standardized data. This study offers an alternative solution for integrating multi-source, heterogeneous clinical data while providing high-quality AI-ready data for applications such as data mining and LLM training. This advancement supports research in clinical science, medical artificial intelligence, and related fields.

Human-motion video generation has been a challenging task, primarily due to the difficulty inherent in learning human body movements. While some approaches have attempted to drive human-centric video generation explicitly through pose control, these methods typically rely on poses derived from existing videos, thereby lacking flexibility. To address this, we propose HumanDreamer, a decoupled human video generation framework that first generates diverse poses from text prompts and then leverages these poses to generate human-motion videos. Specifically, we propose MotionVid, the largest dataset for human-motion pose generation. Based on the dataset, we present MotionDiT, which is trained to generate structured human-motion poses from text prompts. Besides, a novel LAMA loss is introduced, which together contribute to a significant improvement in FID by 62.4%, along with respective enhancements in R-precision for top1, top2, and top3 by 41.8%, 26.3%, and 18.3%, thereby advancing both the Text-to-Pose control accuracy and FID metrics. Our experiments across various Pose-to-Video baselines demonstrate that the poses generated by our method can produce diverse and high-quality human-motion videos. Furthermore, our model can facilitate other downstream tasks, such as pose sequence prediction and 2D-3D motion lifting.

Multi-view learning methods often focus on improving decision accuracy, while neglecting the decision uncertainty, limiting their suitability for safety-critical applications. To mitigate this, researchers propose trusted multi-view learning methods that estimate classification probabilities and uncertainty by learning the class distributions for each instance. However, these methods assume that the data from each view can effectively differentiate all categories, ignoring the semantic vagueness phenomenon in real-world multi-view data. Our findings demonstrate that this phenomenon significantly suppresses the learning of view-specific evidence in existing methods. We propose a Consistent and Complementary-aware trusted Multi-view Learning (CCML) method to solve this problem. We first construct view opinions using evidential deep neural networks, which consist of belief mass vectors and uncertainty estimates. Next, we dynamically decouple the consistent and complementary evidence. The consistent evidence is derived from the shared portions across all views, while the complementary evidence is obtained by averaging the differing portions across all views. We ensure that the opinion constructed from the consistent evidence strictly aligns with the ground-truth category. For the opinion constructed from the complementary evidence, we allow it for potential vagueness in the evidence. We compare CCML with state-of-the-art baselines on one synthetic and six real-world datasets. The results validate the effectiveness of the dynamic evidence decoupling strategy and show that CCML significantly outperforms baselines on accuracy and reliability. The code is released at https://github.com/Lihong-Liu/CCML.

Objective: To update and upgrade the semantic annotation system [1] for Traditional Chinese Medicine(TCM) literatures developed by our team in the previous period, oriented to the actual application requirements. Methods: The workflow and functional modules of the semantic annotation system are updated and upgraded to meet the functional requirements of practical application scenarios, and special functions are developed. Results: Based on the previous system, the BaiBu Knowledge Engine was upgraded and developed with new functions such as setting and visualizing the multi-level structure of entities and semantic relations at the schema layer and event annotation, and we have improved the model, and adding new functions such as semantic search of the system's front-end knowledge base and visualization of the knowledge sources for knowledge traceability, so as to improve the annotation personnel's knowledge and knowledge management skills, the new features include semantic search and visualization of knowledge sources for knowledge traceability in the front-end of the system, in order to improve the annotation efficiency of the annotators and to express the deep implicit knowledge of TCM’s literature. Conclusion: The updated and upgraded BaiBu Knowledge Engine has been verified and put into use in actual projects, which can provide powerful support for the expression and deep utilization of the knowledge of TCM literature, and realize data integration and knowledge fusion at the semantic level of TCM.

Objective: To construct a natural language processing (NLP) system focused on named entity recognition (NER) and semantic relation extraction (RE) of ancient Chinese medical books, it supports annotated corpora management and semantic knowledge retrieval. Methods: We integrate the 47 ontologies and terminologies as the terminology database. After that, we trained a preprocessing NER model using spaCy and used a hybrid approach combining automated annotation and manual review to annotate corpora of ancient Chinese medical books. Results: The semantic annotation system of Chinese ancient texts named traditional Chinese medicine - semantic annotation system (TCM-SAS), was constructed based on ontologies and terminologies. Annotations and knowledge retrieval of TCM's ancient texts were realized. Conclusion: TCM-SAS is a user-friendly semantic annotation system for ancient Chinese medical books that includes a large-scale manual annotation of TCM literature and semantic knowledge of TCM. TCM-SAS could provide users with two modes of automatic and manual NER and RE for ancient Chinese texts, as well as annotated entity and corpora management. Support the discovery of new knowledge from ancient Chinese medical texts in the future.

Objective: To construct the ontology of interaction between Chinese medicines and chemical medicine, and use ontology reasoning tools to automatically discover the results of interaction between Chinese medicines and chemical medicine. Methods: Based on the analysis of drug safety rules and the principle of drug interaction, the user-defined rules of ontology reasoning tool are designed with ontology language. Results: The ontology of interaction between Chinese medicines and chemical medicine was constructed, and the reasoning from drug interaction to use risk was realized by using ontology reasoning tools. Conclusion: The interaction mode between Chinese medicines and chemical medicine is complex. The construction and research of ontology reasoning of the interaction ontology of Chinese medicines and chemical medicine can provide ideas and methods for the research, but further research is needed.

An event-triggered attitude control algorithm is developed for quadrotor unmanned aerial vehicles (UAVs) subject to external disturbances. In this article, first an event-triggered supertwisting stabilizing control strategy for a class of second-order nonlinear systems is proposed. Then, a Lyapunov-based stability analysis is provided for the closed-loop system, and the Zeno-free execution of triggering sequence is guaranteed via rigorous analysis. Furthermore, the proposed control strategy is applied on attitude control of UAVs to reduce the computing cost without degrading the performance of the system. Finally, the efficiency of the developed method is validated by numerical simulation.

Accurate segmentation of lung cancer in pathology slides is a critical step in improving patient care. We proposed the ACDC@LungHP (Automatic Cancer Detection and Classification in Whole-slide Lung Histopathology) challenge for evaluating different computer-aided diagnosis (CADs) methods on the automatic diagnosis of lung cancer. The ACDC@LungHP 2019 focused on segmentation (pixel-wise detection) of cancer tissue in whole slide imaging (WSI), using an annotated dataset of 150 training images and 50 test images from 200 patients. This paper reviews this challenge and summarizes the top 10 submitted methods for lung cancer segmentation. All methods were evaluated using metrics using the precision, accuracy, sensitivity, specificity, and DICE coefficient (DC). The DC ranged from 0.7354 ±0.1149 to 0.8372 ±0.0858. The DC of the best method was close to the inter-observer agreement (0.8398 ±0.0890). All methods were based on deep learning and categorized into two groups: multi-model method and single model method. In general, multi-model methods were significantly better (p 0.01) than single model methods, with mean DC of 0.7966 and 0.7544, respectively. Deep learning based methods could potentially help pathologists find suspicious regions for further analysis of lung cancer in WSI.

Automatic quantification of the left ventricle (LV) from cardiac magnetic resonance (CMR) images plays an important role in making the diagnosis procedure efficient, reliable, and alleviating the laborious reading work for physicians. Considerable efforts have been devoted to LV quantification using different strategies that include segmentation-based (SG) methods and the recent direct regression (DR) methods. Although both SG and DR methods have obtained great success for the task, a systematic platform to benchmark them remains absent because of differences in label information during model learning. In this paper, we conducted an unbiased evaluation and comparison of cardiac LV quantification methods that were submitted to the Left Ventricle Quantification (LVQuan) challenge, which was held in conjunction with the Statistical Atlases and Computational Modeling of the Heart (STACOM) workshop at the MICCAI 2018. The challenge was targeted at the quantification of 1) areas of LV cavity and myocardium, 2) dimensions of the LV cavity, 3) regional wall thicknesses (RWT), and 4) the cardiac phase, from mid-ventricle short-axis CMR images. First, we constructed a public quantification dataset Cardiac-DIG with ground truth labels for both the myocardium mask and these quantification targets across the entire cardiac cycle. Then, the key techniques employed by each submission were described. Next, quantitative validation of these submissions were conducted with the constructed dataset. The evaluation results revealed that both SG and DR methods can offer good LV quantification performance, even though DR methods do not require densely labeled masks for supervision. Among the 12 submissions, the DR method LDAMT offered the best performance, with a mean estimation error of 301 mm2for the two areas, 2.15 mm for the cavity dimensions, 2.03 mm for RWTs, and a 9.5% error rate for the cardiac phase classification. Three of the SG methods also delivered comparable performances. Finally, we discussed the advantages and disadvantages of SG and DR methods, as well as the unsolved problems in automatic cardiac quantification for clinical practice applications.

Channel prediction is an appealing technique to mitigate the performance degradation due to the inevitable feedback delay of the channel state information (CSI) in modern wireless systems. We first propose a general MIMO-OFDM channel prediction framework, which exploits both the spatial and temporal correlations among antennas. Then we derive two predictors which select data for auto-regressive (AR) predictors in different ways based on the proposed framework. The first predictor chooses the data set via minimizing the mean square error (MSE) of prediction model. The second predictor chooses the data in a heuristic way, which aims to reduce the computational complexity. Our algorithms can be applied to improve the precoding performance in multi-user MIMO-OFDM systems. Simulation results show that the proposed methods can overcome the feedback delay effectively, even when the channel changes rapidly.