Compared with traditional medical devices, artificial intelligence medical devices face greater challenges in the process of clinical trials due to their related characteristics of artificial intelligence technology. This paper focused on the challenges and risks in each stage of clinical trials on artificial intelligence medical devices for assisted diagnosis, and put forward corresponding coping strategies, with the aim to provide references for the performance of high-quality clinical trials on artificial intelligence medical devices and shorten the research period in China.
In the field of artificial intelligence (AI) medical imaging, data annotation is a key factor in all AI development. In the traditional manual annotation process, there are prominent problems such as difficult data acquisition, high manual labor intensity, strong professionalism and low labeling quality. Therefore, an intelligent multimodal medical image annotation system is urgently needed to meet the requirements of labeling. Based on the image cloud, West China Hospital of Sichuan University collected the multimodal image data of hospital and allied hospitals, and designed a multi-modal image annotation system through information technology, which integrated various image processing algorithms and AI models to simplify the image data annotation. With the construction of annotation system, the efficiency of data labeling in the hospitals is improved, which provides necessary data support for the AI image research and related industry construction in the hospital, so as to promote the implementation of artificial intelligence industry related to medical images in the hospital.
In order to promote the effective development of hospital day surgery mode, a construction method of information management platform that meets the characteristics of day surgery mode is presented. By analyzing the business process of the day surgery mode, the system architecture of the information platform is given; according to the difficulty of the surgical scheduling, the two-stage surgical scheduling algorithm based on the ranking theory is given; by analyzing the day surgery data statistically, a multi-angle surgical index analysis module is provided. The information management of the day surgery mode has been realized, and the work efficiency has been improved. A reasonable day surgery information platform construction can help to optimize the daytime surgical procedure and promote the smooth development of day surgery.
Objective To explore the feasibility and effectiveness of vertebroplasty with reverse designed unilateral targeted puncture in treatment of osteoporotic vertebral compression fracture (OVCF) by comparing with curved unilateral puncture. Methods A total of 52 patients with OVCF met selection criteria and were admitted between January 2019 and June 2021 were selected as the research objects. According to the random number table method, they were divided into two groups (n=26). In trial group, the reverse designed unilateral targeted puncture was used in the percutaneous vertebroplasty (PVP); while the control group used the curved unilateral puncture. There was no significant difference in gender, age, bone mineral density (T value), cause of injury, time from injury to operation, the level of responsible vertebral body, pedicle diameter of the planned puncture vertebral body, and preoperative visual analogue scale (VAS) score, anterior vertebral height, and Cobb angle between the two groups (P>0.05). The operation time, bone cement injection volume and leakage, intraoperative radiation exposure times, and hospitalization costs in the two groups were recorded. VAS score was used to evaluate the relief degree of low back pain after operation. X-ray film was used to review the diffusion degree of bone cement in the responsible vertebral body, and Cobb angle and anterior vertebral height were measured. Results The operation was successfully completed in the two groups. Patients in the two groups were followed up 12-18 months, with an average of 13.6 months. The operation time, volume of injected bone cement, intraoperative radiation exposure times, and hospitalization costs in the trial group were significantly lower than those in the control group (P<0.05). With the prolongation of time, the low back pain of the two groups gradually relieved, and the VAS score significantly decreased (P<0.05). And there was no significant difference in VAS score between the two groups at each time point (P>0.05). There were 2 cases (7.6%) of bone cement leakage in the trial group and 3 cases (11.5%) in the control group, and no significant difference was found in the incidence of bone cement leakage and the diffusion degree of bone cement between the two groups (P>0.05). Imaging examination showed that compared with pre-operation, the anterior vertebral height of the two groups significantly increased and Cobb angle significantly decreased at 2 days and 1 year after operation (P<0.05); while compared with 2 days before operation, the anterior vertebral height of the two groups significantly decreased and Cobb angle significantly increased at 1 year after operation (P<0.05). There was no significant difference in the above indexes between the two groups at different time points after operation (P>0.05). Conclusion Compared with curved unilateral puncture, the use of reverse designed unilateral targeted puncture during PVP in the treatment of OVCF can not only achieve similar effectiveness, but also has the advantages of less radiation exposure, shorter operation time, and less hospitalization costs.
The full process information management of daytime surgery can help medical staff complete centralized patient management, improve the closed-loop quality of daytime surgery, and maximize the efficiency and management level of hospital daytime surgery operation. Since 2021, the First Hospital of Lanzhou University has integrated internal information exchange resources, big data, and artificial intelligence, created a full process management platform for daytime surgery, and explored the intelligent management of daytime surgery processes. This article shares the experience of building an intelligent daytime surgery full process management model based on interactive design information system from the aspects of platform interaction design, intelligent management mode, application effectiveness, in order to provide a reference for optimizing intelligent closed-loop management of daytime surgery.
An auxiliary dining robot is designed in this paper, which implements the humanoid feeding function with theory of inventive problem solving (TRIZ) theory and aims at the demand of special auxiliary nursing equipment. Firstly, this robot simulated the motion function of human arm by using the tandem joints of the manipulator. The end-effector used a motor-driven spoon to simulate the feeding actions of human hand. Meanwhile, the eye in hand installation style was adopted to instead the human vision to realize its automatic feeding action. Moreover, the feeding and drinking actions of the dining robot were considered comprehensively with the flexibility of spatial movement under the lowest degree of freedom (DOF) configuration. The structure of the dining robot was confirmed by analyzing its stresses and discussing the specific application scenarios under this condition. Finally, the simulation results demonstrate high-flexibility of the dining robot in the workspace with lowest DOF configuration.
The data collection form is a bridge in-between the original studies and the final systematic reviews. It’s the basis for data analyses, directly related to the results and conclusions of systematic reviews, and plays an important role in systematic reviews. There are strict requirements of data collection forms in making Cochrane systematic reviews. In this article, the authors introduce their experiences regarding to the design of data collection form.
ObjectiveTo systematically summarize recent advancements in the application of artificial intelligence (AI) in key components of radiotherapy (RT), explore the integration of technical innovations with clinical practice, and identify current limitations in real-world implementation. MethodsA comprehensive analysis of representative studies from recent years was conducted, focusing on the technical implementation and clinical effectiveness of AI in image reconstruction, automatic delineation of target volumes and organs at risk, intelligent treatment planning, and prediction of RT-related toxicities. Particular attention was given to deep learning models, multimodal data integration, and their roles in enhancing decision-making processes. ResultsAI-based low-dose image enhancement techniques had significantly improved image quality. Automated segmentation methods had increased the efficiency and consistency of contouring. Both knowledge-driven and data-driven planning systems had addressed the limitations of traditional experience-dependent approaches, contributing to higher quality and reproducibility in treatment plans. Additionally, toxicity prediction models that incorporated multimodal data enabled more accurate, personalized risk assessment, supporting safer and more effective individualized RT. ConclusionsRT is a fundamental modality in cancer treatment. However, achieving precise tumor ablation while minimizing damage to surrounding healthy tissues remains a significant challenge. AI has demonstrated considerable value across multiple technical stages of RT, enhancing precision, efficiency, and personalization. Nevertheless, challenges such as limited model generalizability, lack of data standardization, and insufficient clinical validation persist. Future work should emphasize the alignment of algorithmic development with clinical demands to facilitate the standardized, reliable, and practical application of AI in RT.