Intracranial aneurysm refers to an abnormal bulging on the wall of intracranial arteries. Its rupture is the main cause of nontraumatic subarachnoid hemorrhage, with extremely high mortality and disability rates. Traditional treatment methods include craniotomy with clipping and endovascular interventional therapy. However, these conventional approaches often face significant challenges when dealing with large, giant, fusiform, or wide-necked aneurysms. The emergence of flow diverter (FD) represents a revolutionary shift in the treatment concept of intracranial aneurysms, moving from traditional endosaccular packing to the reconstruction of the parent artery. As a type of dense-mesh stent with high metal coverage and low porosity, FD is placed at the neck of the aneurysm to significantly alter the intraluminal hemodynamics of the aneurysm, promote blood flow stasis and intraneurysmal thrombosis. Meanwhile, acting as a scaffold, FD facilitates endothelial cell migration and proliferation, ultimately achieving complete occlusion of the aneurysm and physiological reconstruction of the parent artery. This article aimed to systematically review the development process of FD devices from the proposal of the concept to clinical practice, elaborate on their mechanism of action in detail, analyze its clinical effectiveness and safety, and provide an outlook on their future development directions.

Objective To systematically evaluate the multidimensional impacts of the Beijing volume-based procurement (VBP) policy on the consumption of key high-value medical consumables, cost structure, clinical behaviors, and disciplinary development in neurosurgical interventional procedures, and to provide evidence-based basis for subsequent policy iteration. Methods All neurosurgical interventional cases using flow diverter (FD) in Xuanwu Hospital of Capital Medical University from January 2020 to March 2025 were enrolled. Taking the implementation of VBP on December 15, 2023 as the time node, the cases were divided into the pre-VBP group (1048 cases) and the post-VBP group (942 cases). Patient information, cost and consumable usage data were obtained through the hospital’s multiple data platforms to construct a three-level nested dataset of “case-surgery-consumable”, the various costs, consumable usage and the usage quantity of different types of FD before and after the implementation of VBP were compared, and explored the correlation between FD costs and other indicators. Results All research indicators did not follow a normal distribution. The number of patients undergoing FD implantation surgery increased year by year from 2020 to 2024, while the number of disease diagnosis categories showed no significant change. The average length of hospital stay of patients decreased from 6.34 d to 5.00 d after VBP (Z=6.484, P<0.05). All types of costs decreased significantly after the implementation of VBP (P<0.05), among which the consumable cost had the largest decrease (median decrease: 30860 yuan/case, Z=13.516, P<0.05), and the median total cost decreased by 33110 yuan/case (Z=13.116, P<0.05). The total consumption of consumables per case decreased by 8 units (Z=5.858, P<0.05), while the usage quantity of FD per case had no significant change (P>0.05), and the usage proportion of FD with surface treatment increased significantly (P<0.05). FD consumable cost was strongly positively correlated with consumable cost and total cost (Spearman correlation coefficients: 0.721 and 0.703, respectively, P<0.05), weakly positively correlated with the proportion of consumable cost in total cost, diagnosis and treatment cost, and surgical cost, and weakly correlated with other indicators. Conclusion Through the mechanism of “volume-based pricing+quality guarantee”, the VBP policy realized the intensive use of high-value medical consumables, optimization of cost structure and improvement of clinical efficiency in neurosurgical intervention, providing an empirical scheme for the implementation of “value-based healthcare” in the field of neurosurgery in China.

Objective To systematically analyze the perioperative safety and surgical efficiency of implanting flow diverters (FD) of different materials in patients with intracranial aneurysms, and to provide empirical evidence for the clinical selection of safer and more effective treatment methods. Methods This study was a single-center retrospective study. A total of 1316 patients who were treated with FD from January 1, 2022 to March 31, 2025 were selected as the research subjects. The enrolled patients were divided into the nickel-titanium alloy group (118 cases) and the cobalt-chromium alloy group (1198 cases) based on the implantation of FD. The following core indicators were compared and analyzed: operation time, intraoperative blood loss, postoperative hospital stay, the frequency of postoperative anti-inflammatory/anti-edema medication, and the usage quantities of FDS, support catheters, microcatheters, and microguidewires. Results The cobalt-chromium alloy group was significantly superior to the nickel-titanium alloy group in terms of postoperative hospital stay and operation time, and the differences were statistically significant (P<0.05). However, there was no statistically significant difference between the two groups in terms of intraoperative blood loss and the frequency of anti-inflammatory/anti-edema medication (P>0.05). In terms of the number of microcatheters used, the nitinol group was significantly lower than the cobalt-chromium alloy group, and the difference was statistically significant (P<0.05). Additionally, the number of microcatheters used in the cobalt-chromium alloy group increased year by year, while the proportion used in the nickel-titanium alloy group decreased significantly. Conclusion Cobalt-chromium alloy FD is superior to nickel-titanium alloy FD in terms of surgical efficiency and patient recovery speed, while there is no difference between the two in terms of surgical trauma and postoperative inflammation control. This study can provide a basis for the clinical selection of safer and more effective FD.

Objective To investigate the changes in surgical patient costs and utilization of flow diverter (FD) (designed based on different surface technologies) in the treatment of intracranial aneurysms, and to analyze the economic impact of this consumable in clinical practice. Methods The FD consumable utilization catalog of Xuanwu Hospital, Capital Medical University was established. Stents were categorized into 2 groups by whether surface treatment technology was applied (yes/no), and into 3 groups by the type of surface treatment (no surface treatment, surface coating treatment, surface polishing treatment). Statistical analysis was performed on the hospital’s consumable utilization data (from January 2020 to March 2025) by calculating parameters including the average number (amount) of consumables per case, average number of FDs per case, average proportion of FD quantity per case, and average proportion of FD amount per case. Results From 2020 to 2025, the average number of consumables per case decreased year by year, with a statistically significant difference (H=260.870, P<0.001); the average amount of consumables per case also decreased, with a statistically significant difference (H=412.300, P<0.001). Under different surface treatment technologies, the differences in the average number and amount of consumables per case were statistically significant (P<0.05); there was no significant difference in consumable quantity, but the total consumable amount for stents using coating technology increased significantly (P<0.05). Conclusion When FDs with different surface treatment technologies are used for intracranial aneurysm treatment, there are significant differences in the total quantity/amount of consumables used by patients and the unit price of stents. Studying its impact on treatment costs can provide a data basis for standardized/refined consumable management and hospital cost control.

Objective To investigate the effects of various surface treatment techniques on the perioperative bleeding risk, surgical efficiency, postoperative recovery, and antiplatelet medication burden in patients with intracranial aneurysms undergoing flow diverter (FD) stenting. The goal is to provide evidence-based guidance for clinical decision-making. Methods A retrospective analysis was conducted on 1972 patients who underwent FD implantation in Xuanwu Hospital, Capital Medical University. The patients were divided into three groups according to the surface treatment type of the stents they received: polishing group, coating group and nonsurface treatment group. A total of 1217 cases without special comorbidities were selected for focused comparison, with the operative duration, intraoperative blood loss, postoperative length of hospital stay and frequency of postoperative antiplatelet medication administration as the key outcome measures. Results No statistically significant difference was found in intraoperative blood loss among the three groups (H=3.749, P=0.153), with the median blood loss of 10 mL in each group, suggesting that different surface treatment methods did not increase the immediate bleeding risk. There was a significant difference in operative duration among the three groups (H=24.312, P<0.05), and the operative duration in both the polishing group and the coating group was significantly shorter than that in the non-surface treatment group (P<0.05). The postoperative length of hospital stay differed significantly among the three groups (H=76.899, P<0.05), the median length of hospital stay was the shortest in the polishing group (1.85 d), followed by the coating group (1.95 d) and the non-surface treatment group (2.58 d), with statistically significant differences observed in all pairwise comparisons (P<0.05). There was a significant overall difference in the frequency of antiplatelet medication use among the three groups (H=10.461, P<0.05). Multiple comparisons showed that a significant difference was detected only between the polishing group and the non-surface treatment group (P<0.05), while no statistically significant differences were identified between the other groups (P>0.05). Conclusion In the scenario of FD implantation with low bleeding risk, both polishing and coating surface treatments do not increase the incidence of immediate bleeding events, and can significantly shorten the operative duration and postoperative length of hospital stay. In addition, polishing can reduce the burden of antiplatelet medication, thus demonstrating favorable safety and advantages in clinical efficiency.

To meet the higher requirements for hospital management imposed by national policies on medical consumables, such as “zero markup” and “centralized volume-based procurement”, the supply-processing-distribution (SPD) management model for medical consumables has been gradually promoted. However, there is currently a lack of unified management standards, and the models are developing in a diversified manner. Focusing on large general hospitals, this study first sorted out and compared the operational characteristics, advantages and disadvantages of the three major mainstream SPD models, namely the hospital-owned and self-operated model, the centralized distribution service model, and the hospital-led operation outsourcing model. Secondly, it analyzed the development trend of the SPD model nationwide. Finally, it summarized the current innovation directions of the models, including the introduction of a third-party SPD pure service model with no interest ties, and the enhancement of hospitals’ management initiative through the combination and collaboration of multiple service providers. This study aims to identify the limitation that existing research on the SPD model is mostly qualitative description, and proposes that subsequent research should strengthen the quantitative comparison in dimensions such as supply chain efficiency and management quality. Meanwhile, it provides references for large general hospitals (as the core carriers of SPD model innovation), helping them analyze the advantages and disadvantages of the models based on actual data, promote the implementation of SPD management standards, and improve the management efficiency of the medical consumables supply chain.

Objective To design a multi-supplier supply-processing-distribution (SPD) management platform to solve the problem of multi-supplier management in hospital consumables management and optimize the full-process management. Methods Adopt the model of “self-built platform and connection with supplier systems”, through multi-level system architecture design, build a unified management process, data standards and assessment system, achieve seamless connection with the hospital system, cover the links of access, distribution, use and settlement, and be equipped with intelligent devices and analysis tools. Results After the system was put into use, the accuracy rate of delivery, the timeliness rate of delivery and the number of delivery orders were all improved compared with those before the system was used, and the differences were statistically significant (P<0.05). Meanwhile, the number of service personnel required by the hospital decreased from 13 to 9, and the number of orders processed per person increased, which saved labor costs and improved service efficiency. Conclusion The SPD platform enhances the refined level of consumable management, forms a management closed loop, and significantly improves the operational efficiency and medical quality of the hospital. In the future, the system will continue to be optimized, the application of radio frequency identification technology will be expanded, and the management accuracy and efficiency will be enhanced.

Objective To explore the reform of supply-processing-distribution (SPD) supply chain services for medical consumables in hospitals, facilitate the refined management of medical consumables, and strengthen the refined management practice in all circulation links of medical consumables through data integration. Methods Data interfaces were connected to link information systems such as the hospital information system (HIS), hospital resource planning (HRP), and electronic medical record (EMR). A medical consumable supply management system was constructed, covering functions including consumable coding, inventory management, clinical traceability, intelligent replenishment, and settlement connection. Through the monitoring and analysis of logistics data in various usage links, the whole-life-cycle data traceability management of medical consumable supply was realized. Results After the launch of the SPD model, more than 130000 surgical procedures were supported and over 980000 pieces of consumables were distributed in 2024. Compared with the traditional model, the SPD model shortened the preoperative preparation time, consumable retrieval time, and settlement & billing time by 84.96%, 24.00%, and 62.50% respectively (P all<0.001). In terms of cost and management optimization, zero-inventory management of consumables was achieved; 178 out of 1101 consumable specifications completed price adjustments, resulting in an annual cost saving of 3.3 million yuan. The cost of surgical consumables in 2024 was significantly lower than that in 2023 (P<0.001). Conclusion The SPD reform meets the needs of refined management of medical consumables, and realizes the full-chain closed-loop management of “clinical-logistics-finance” for the first time. It has application integration and clinical adaptability, can improve management efficiency, and holds practical significance and promotion value for the high-quality development of medical institutions.

Objective To address the current management status of non-sterile orthopedic consumables, reshape the workflow, achieve full-process supervision, effective traceability, unified usage and billing, improve billing accuracy and efficiency, and reduce sterilization costs for non-sterile consumables. Methods A digital orthopedic management platform was established using digital technology, implementing a “one-item-one-code” management model for non-sterile orthopedic consumables. An advanced image recognition system was introduced to streamline the management process, extending supervision to the hospital’s internal warehousing stage. This enabled precise acceptance of each implant, internal traceability, and accurate billing. The effectiveness of the new model was evaluated by comparing billing accuracy rates, billing duration, and sterilization costs before and after the system’s implementation. Results After the new model was implemented, billing accuracy increased from 99.95% to 100%; billing duration decreased by 62.5%; and sterilization costs decreased by 87.5% compared to pre-implementation levels, with all differences being statistically significant (P<0.05). Conclusion The implementation of the “one-item-one-code” management model for non-sterile orthopedic consumables not only enhances the efficiency and accuracy of consumable management but also reduces hospital operational costs, demonstrating significant practical value and potential for broader application.

Objective To explore the application practice and specific effects of the multi-party collaborative kit preparation and management model for surgical medical consumables based on the supply-processing-distribution (SPD) information platform in a large public Grade-III hospital. Methods Data from September 2023 (control group) and September 2024 (observation group) in the hospital were selected. The two groups were compared in terms of total daily preoperative consumable preparation time, preoperative consumable preparation error rate, intraoperative consumable acquisition time per operation, consumable billing time per operation, postoperative inventory time per operation, and surgical billing error rate. The effects before and after the application of the model were analyzed. Results After the public Grade-III hospital applied the multi-party collaborative kit preparation and management model for surgical medical consumables based on the SPD information platform, the total daily preoperative consumable preparation time, preoperative consumable preparation error rate, intraoperative consumable acquisition time per operation, consumable billing time per operation, postoperative inventory time per operation, and surgical billing error rate all decreased significantly, with statistically significant differences (P<0.05). Conclusion The collaborative preparation management mode of surgical medical apparatus based on SPD information platform is a high-efficiency, high-accuracy, and high-management effectiveness management mode of surgical medical apparatus. It can effectively improve the level of efficient and refined management of surgical medical consumables in public hospitals.

Objective To explore the risks and causes of terminal displacement control after emergency stop of medical surgical robots, design test methods, conduct actual tests, analyze the influencing factors of terminal displacement after emergency stop based on the results, and clarify the measurement conditions of this index.Methods The emergency stop signal of the robotic arm was elicited, and the spatial coordinates of the reference point at the end of the robotic arm were collected continuously and quickly by the non-contact space measurement device when the emergency stop was pressed. The influence of static or moving along different vector directions at different speeds on the end displacement after emergency stop was analyzed under different emergency stop control modes, and the influence of these parameters on the test results was compared and analyzed by one-way ANOVA.Results Through the experiments designed in this paper, it was found that the initial and final displacement of the emergency stop and the maximum displacement of the emergency stop of the robotic arm were significantly different, and the difference could be 0.49-1.61 mm in different vector directions and 0.18-0.57 mm at different speeds. For some models of robotic arms, the difference in emergency stop control modes could cause a displacement difference of more than 10 mm. At the same time, the posture during emergency stop also had a certain degree of influence on the displacement at the end of the robotic arm after emergency stop.Conclusion This paper designs a scientific and effective testing method, which can test the maximum displacement at the end and the initial and final displacement of medical surgical robots after emergency stop. When conducting tests, the influence of running speed, emergency stop control methods and the posture during emergency stops on the results should be fully considered.

Objective To discuss the influence of linear and parabolic air resistance on the capacity control of ventilator according to the new requirements of GB 9706.212-2020 standard for control accuracy, and provide guidance for optimizing the control algorithm and detection specification of ventilator.Methods In this paper, the hydrodynamic characteristics of linear and parabolic impediments and their effects on the control accuracy of tidal volume were compared by ventilator ventilation experiment.Results The experimental results showed that the control error of a ventilator was 9.6% when using parabolic resistance under R5 conditions. When linear resistance was used, the control error increased to -39.0%, and the flow value in the airway decreased significantly. Tidal volume control behaves differently with different resistance settings, especially at low resistance settings where linear resistance resistance may lead to large tidal volume errors.Conclusion Simulated pulmonary resistance can affect the accuracy of tidal volume output of ventilators. Manufacturers need to optimize the control algorithm according to the new standard to ensure that the equipment meets the requirements.

Objective To analyze the numerous challenges encountered in the classification of P300 visual evoked potential signals and to investigate novel prospective solutions.Methods This study introduced a novel KAN neural network model, augmented with a convolutional self-attention mechanism. This innovative approach proficiently captured the global feature information of P300 signals. Furthermore, the incorporation of the KAN layer enhanced its capability to manage nonlinear data effectively. To validate the efficacy of this model, experiments were executed utilizing the widely recognized brain computer interface Competition Ⅲ Challenge 2004 dataset. The performance of our proposed model was subsequently juxtaposed with modern P300 VEP classification techniques.Results The proposed model exhibited superior classification accuracy on the validation set, achieving an impressive 100.0% accuracy in P300 signal classification. This performance surpassed that of VGG-16, which achieved 98.9%, and ResNet-18, which achieved 99.0%. Furthermore, in experiments involving fast gradient sign methed attacks, the model maintained an accuracy of 82%.Conclusion This study presents a novel methodology for the classification of P300 VEP signals, which can also be applied to similar tasks in brain signal processing. This offers fresh research perspectives and contributes to the progression of brain signal research.

Objective Based on the current status of usability testing of medical surgical robots in China, to explore the common perceptions of enterprises, clinical practitioners and testing agencies regarding the usability testing of surgical robots, and to compare the differences in their needs and concerns regarding usability testing, aiming to provide reference for the formulation of usability testing standards.Methods According to the general requirements of medical device regulation on usability, questionnaires were designed based on factors such as usability testing needs and experimental processes. Data were analyzed to explore the differences and consensus revealed by common questions among different respondents regarding the usability of different types of surgical robots.Results Analysis of the 60 collected questionnaires revealed that all respondents emphasized the fundamental role of surgical instruments and the necessity of simulating the surgical environment, identifying the completion rate of surgical tasks as a key objective indicator. There were significant differences in the usability testing of different types of surgical robots in terms of auxiliary equipment needs, the role of auxiliary equipment and key objective indicators, with variations in experimental design.Conclusion This study investigated the current status of usability testing of medical surgical robots in China and revealed the common perceptions of different respondents and the specific needs of various types of surgical robots in testing through multi-level analysis. It provides scientific evidence for the formulation of unified testing standards and helps promote the application and development of surgical robots in the medical field.

ObjectiveTo analyze the global research competition and cooperation landscape in the field of brain-computer interface (BCI), to provide reference for relevant practitioners. Methods BCI-related papers were retrieved from the Web of Science Core Collection. Bibliometric analysis and social network analysis methods were employed to analyze the global research competition and cooperation dynamics in the BCI field, identifying leading countries, institutions, and researchers. Results From 1990 to 2023, a total of 9037 SCI papers in the BCI field were produced globally, with 78.9% of these papers published between 2014 and 2023, showing a ten-year compound growth rate of 13.7%. The publication volume of China and the United States placed them in the leading tier globally. Although China started later, its growth rate was faster, with its publication volume in 2023 reaching three times that of the United States. In terms of cooperation, the United States surpassed China both in the number of collaborators (64) and the number of collaborations (1507). Other notable countries included Germany (968 papers) and the United Kingdom (725 papers) with relatively high publication volumes. In terms of institution and researcher, the University of Tübingen in Germany and Graz University of Technology in Austria were early leaders in BCI research, but their paper outputs significantly declined in recent years. In contrast, researchers from institutions such as Tsinghua University (GAO Xiaorong), Chinese Academy of Sciences (WANG Yijun), Tianjin University (MING Dong), East China University of Science and Technology (JIN Jing), and South China University of Technology (LI Yuanqing) produced many papers in recent years. Conclusion The BCI field has developed rapidly over the past decade, with the United States and China leading in global competition and cooperation. The United States had an early start but has experienced slower growth in recent years, while China started later but has seen rapid growth recently. Institutions such as Chinese Academy of Sciences, Tsinghua University, and Tianjin University have shown outstanding performance.

ObjectiveTo explore the main path of technology development based on the patent citation network in the field of brain-computer interface (BCI). Methods The patent citation network in the field of brain-computer interface was constructed, and the traversal weights were calculated by the search path link count. The global main path, global key-route main path, local forward main path, local backward main path and local key-route main path were analyzed by two path search methods, global search and local search, so as to explore the technological development trajectory in this field of BCI. Results There were 15519 patent applications in the field of BCI, including 140691 patent citations, and the number of patent applications increased year by year. Technology development focused on neural stimulation and regulation technology, neural signal acquisition and processing technology, BCI hardware technology and other directions. The main path analysis showed that the technical theme was electrical stimulation as the core, emphasizing precise neural regulation and personalized treatment. The number of patents on the global main path was the largest (16), including 2 technical routes, and the global key-route main path was consistent with the global main path. There were 14 patents on the local forward main path, including 1 technical route, and the local key-route main path was consistent with the local forward main path. There were 10 patents on the local backward main path, including 1 technical route. Conclusion Through the main path analysis of patents in the field of BCI, this study identifies the technological development trajectory and provides information support for the research and development of BCI from the perspective of information science.

ObjectiveTo analyze the global registration status of brain-computer interface (BCI) clinical trials and provide data to support further clinical research of this technology. Methods Based on the registration data of BCI related clinical trials in ClinicalTrials.gov, bibliometrics approach was applied to analyze the number and annual trend of registration, national distribution and international cooperation, sponsor type and country distribution, study type and recruitment, classification of indications and their characteristics. Results As of October 2024, a total of 194 BCI related clinical trials were registered worldwide, and the number of registrations was generally on the rise. Among all the countries that carried out clinical trials, the United States ranked first with 80 trials, while no other countries surpassed 20, and there were only 3 trials involved international collaboration. Sponsors were predominantly universities and research institutions (48.7%), with the United States having the largest number of sponsors. Of all the clinical trials, 88.1% were intervention studies, 38.1% were completed, 26.3% recruited fewer than 10 patients, and almost all of the trial subjects were recruited without gender restriction and included adults. BCI related clinical trials focused on rehabilitation and treatment of neurological diseases, the main indications included stroke, spinal cord injury, amyotrophic lateral sclerosis, and quadriplegia. Conclusion The global clinical research activity in BCI technology is steadily increasing, demonstrating significant application potential.

ObjectiveTo analyze the disciplinary and knowledge composition as well as the trends in brain-computer interface (BCI) research to provide insights for team development, talent cultivation, and cross-disciplinary collaboration in the field. Methods Citation analysis from bibliometrics was applied to calculate interdisciplinary indices within the field, and trend assessments were conducted through statistical analysis. Results The disciplinary diversity in BCI research continued to grow, with varying levels of knowledge concentration across disciplines. Medicine, computer science, psychology, engineering, and physiology were the primary contributing fields. Among them, the shares of medicine and psychology decreased, while computer science significantly increased in proportion, with humanities and social sciences contributing less frequently. Conclusion Driven by advancements in areas such as artificial intelligence, BCI research has shifted from fundamental mechanistic studies to a focus on technology development and algorithmic iteration. At this stage, computer science and engineering have emerged as key pillars in the interdisciplinary knowledge framework. In the future, as technology progresses toward clinical transformation and commercialization, clinical sciences and sociology may receive greater attention.

Objective To establish a clinical application evaluation platform for multi-center medical imaging equipment by building primary and secondary information centers. Methods The clinical application data of medical imaging equipment was obtained through a secondary information center. The raw data was desensitized using methods such as data collection, data governance, and indicator design. Through data space management and data space testing verification techniques, the desensitized data was transmitted to the primary information center. The primary information center designed analysis algorithms based on the clinical application evaluation criteria for medical imaging equipment to evaluate the clinical application of the equipment. Results A number of secondary information centers transmitted standardized data listed to the primary information center through data space management technology, and was able to analyze and evaluate the reliability, availability, functional effect, equipment performance and operating efficiency of medical imaging equipment. Users could intuitively understand the clinical applicability evaluation results of various medical imaging equipment brands through the results displayed on the primary information platform. Taking PET/MR as an example, from the perspective of individual ratings, the reliability evaluation of A brand equipment was 99.69%, and the reliability evaluation of B brand equipment was 98.37%, the reliability of A brand equipment was slightly higher than that of B brand equipment. Taking MRI as an example, the advanced function application rates among the three brands in August 2024 was 61%, 47%, and 38%, respectively. According to the rank sum test, there was no significant difference in the advanced function application rates among the three brands of MRI (P=0.53). Conclusion The construction of clinical application evaluation platform for multicenter medical imaging equipment can provide systematic technical solutions and platform-wise support for medical institutions to select and verify medical imaging equipment, can provide data support for promoting the quality improvement of domestic-made medical imaging equipment, and can provide research data and basic sharing service platform-wise support for establishing and improving standards and standardized systems on the reliability, clinical application verification and maintenance of domestic-made medical imaging equipment.

Objective To analyze the reliability data of CT equipment in use, improve the theoretical framework of equipment reliability analysis and evaluation indicators, explore practical methods for equipment reliability evaluation, to provide practical reference solutions for performance comparison between different CT equipment. Methods Based on the data of the first two years of the newly commissioned CT equipment, the performance comparison framework was constructed according to the reliability theory. According to the maintenance records provided by the equipment manufacturer, the data related to the reliability, maintainability and supportability parameters in the work order of the first year, the second year and the first two years after the installation of a single CT equipment were statistically compared. And the maintenance time, fault time and maintenance interval of the first year and the second year after the installation were compared. Results The reliability parameters of the equipment in the first year, the second year, and the first two years after the installation showed excellent performance in the following order: the first year, the first two years, and second year. The maintenance time and fault time in the second year were longer than those in the first year, and the maintenance interval was shorter than that in the first year, but the difference was not statistically significant (P>0.05). The overall operation status of CT equipment was relatively stable in the first two years after installation, and decreased slightly in the second year. Conclusion The maintenance data analysis and evaluation system established in this study can provide reference information for improving equipment reliability, provide reference solutions for performance comparison between different CT equipment, which can promote the continuous progress of CT technology and industry standardization.

Objective To study and analyze the reliability level of imported and domestic high-end digital radiography (DR) equipment in clinical application using a reliability verification scheme. Methods Data such as maintenance and work logs of three DR equipment (one imported and two domestic) within a certain period were extracted. The image performance parameters, core component status, reliability parameters, maintainability parameters and supportability parameters of the equipment were analyzed, and the performance test data and reliability level of the equipment were compared. Results The image performance of the three devices was stable, and all of them could meet the standard requirements. The state parameters of the core components basically maintained stable operation.The reliability parameters of imported equipment A and domestic equipment B were good, and the availability parameters of both equipment were greater than 95%. The fault time of domestic equipment C was relatively long, and its availability was in the range of 90% to 95%. The maintenance, supportability parameters of domestic equipment performed better. The average maintenance time was 67% faster than that of imported equipment, and the maintenance support delay time was more than 70% lower. Conclusion Based on the basic method of reliability verification, analyzing the reliability level of DR equipment in the clinical application process can provide an objective evaluation of the long-term stability of equipment for medical institutions. The results indicate that domestic high-end equipment with shorter operation time does not have an average reliability performance and shows a large difference from mature imported equipment.

Objective To conduct quality control test on 327 DR equipment in Beijing, analyze the stability of the test data, and enhance the level of DR quality control management. Methods Using the multifunctional dosimeter Unfors Raysafe X2, low contrast test card, spatial resolution test card, and other detection models, the universal and specialized detection items of DR equipment were tested using the detection methods specified in the WS 76-2020 standard, and the stability of equipment performance was analyzed using Minitab 17.0 software. Results The qualification rate of 327 DR equipment quality control tests was 98.17%. The unqualified condition involved three performance indicators, namely the deviation between the light field and the four sides of the irradiation field, response uniformity, and useful beam deviation from perpendicularity. The quality control of most test items were at level Ⅰ, and the quality control of deviation (large focus) of tube voltage indication was at level Ⅲ. Conclusion Through the stability analysis of the performance quality control test data of DR equipment, it is found that the tested equipment still has room for improvement in such important indicators as the deviation between the light field and the four sides of the irradiation field, response uniformity, useful beam deviation from perpendicularity and deviation (large focus) of tube voltage indication. The important results of this study can provide scientific basis and reasonable suggestions for the quality management of DR equipment, provide a guarantee basis for the stable operation of the equipment, and ensure the image diagnosis quality and radiation safety of DR equipment.

Objective To analyze the global scientific research collaboration trends in the field of digital therapeutics through statistical analysis of digital therapeutics-related papers. Methods Digital therapeutics-related papers were retrieved from literature databases. The trends in the number of papers from different countries and institutions in this field were analyzed, a global collaboration network was constructed. The social network indicators were calculated, and the importance of each country and institution in global collaboration were evaluated using the technique for order preference by similarity to ideal solution (TOPSIS) method. Results As of the retrieval date, there were a total of 3224 scientific papers were identified in the field of digital therapies, with 3171 papers (98.36%) published from 2014 to present. The United States had the highest number of publications (1097 papers), and its collaboration partners (75) and collaboration frequency (851 times) were second only to the United Kingdom. The United States received the highest TOPSIS score due to its central position in the global collaboration network. The United Kingdom ranked second in terms of publication volume (776 papers), with slightly higher number of collaboration partners (77) and collaboration frequency (865 times) compared to the United States, but with a slightly lower TOPSIS score. Harvard University, King’s College London, University College London, University of Oxford, and the University of Melbourne had a significant number of collaboration partners and frequency, resulting in higher TOPSIS scores. Among them, Harvard University and King’s College London had a clear advantage. Conclusion The field of digital therapeutics has rapidly development in the past decade, with the United States and the United Kingdom taking an early lead in scientific research and achieving fruitful results. They occupy a central position in the global collaboration network. Other countries are relatively behind the United States and United Kingdom. Research institutions such as Harvard University and King’s College London have seen a rapid increase in output in recent years, driving the development of the field towards intelligence and personalization.

In recent years, the demand for sterilization and disinfection devices has surged, and ultraviolet sterilization, as the most commonly used sterilization and disinfection method in medical and health systems, has become more popular. The research and development of different emitting ultraviolet light source devices, optical materials and devices have also received great attention. This article investigated the clinical use of ultraviolet light equipment, and ultimately focused on the development status of different light source generation devices that emit ultraviolet C (UVC) light and the related progress in the study of photobiological effects and discussed the construction, emission principle, advantages and disadvantages, and emission wavelength of five light source generation devices that can emit UVC light, including mercury vapor ultraviolet lamps, UVC-LED ultraviolet lamps, excimer ultraviolet lamps, microwave plasma ultraviolet lamps, and laser ultraviolet emitters, and summarized the photobiological effects and analyzed their potential medical application prospect.

As a new non-antibiotic treatment method, ultraviolet in the wavelength range of 100-400 nm has been shown to have certain antibacterial properties. The sterilization effect varies from wavelength to wavelength of ultraviolet, and the biosafety also varies. In recent years, lots of research reports have emerged in this field, especially for ultraviolet C (UVC) in the wavelength range of 200-280 nm, which has become a hot research topic. This paper reviewed the new findings in this wavelength range, including the mechanism, factors and effectiveness of action of UVC inactivation of different pathogenic microorganisms, analyzed the biosafety of UVC technology, and also briefly summarized the research progress of UVC technical devices in clinical applications, aiming to provide a reference for related research.

Objective To explore the effects of 275 nm ultraviolet C (UVC) light irradiation on the survival and growth of skin cells and vascular endothelial cells, as well as the effects of long-term and short-term irradiation on the skin of nude mice, and to identify its safety. Methods Human immortalized keratinocytes cells and human umbilical vein endothelial cells were irradiated with 275 nm UVC-LED light source developed, designed and processed by the laboratory for different durations to observe the effects of different irradiation dosages on the survival and growth of the cells; the nude mice skin was irradiated for different durations to observe changes in the appearance of the skin, and tissues of irradiated areas were collected for immunohistochemical analysis. Results For low doses of direct irradiation of cells, 275 nm UVC light had certain skin-related cell safety. However, the high dose of 275 nm UVC light irradiation caused more damage to the cells. Meanwhile, neither low-dose nor high-dose irradiation of 275 nm UVC light induced DNA and epidermal damage in nude mice skin. Conclusion The 275 nm UVC light has been identified as promising for clinical applications, and is relatively safe for the skin.

As the importance of preventing the spread of pathogens in the air becomes increasingly prominent, and it is crucial to seek effective air disinfection technology. Upper space ultraviolet irradiation technology, as an efficient, economical and “human-machine coexistence” air disinfection solution, has long been used to prevent the spread of tuberculosis. With the development of solid-state ultraviolet light sources, the new upper space ultraviolet irradiation system has made many improvements compared to traditional systems in terms of light source upgrades, beam shaping, and structural design. The optimization of beam shaping allows ultraviolet rays to irradiate target areas more accurately, improving the disinfection effect. This paper reviewed the upper space ultraviolet irradiation system, discussed the technical principles, research and development status, application fields, and forecasted the future development trends.

Objective To design and manufacture an isolation laminar flow bed based on deep ultraviolet disinfection, in order to meet the clinical demand for sterile laminar flow bed. Methods The traditional laminar flow bed disinfection method was replaced by deep ultraviolet disinfection, and was magnetic opening and closing, and the connection between the bed curtain and the top of the bed was made of environmentally friendly PVC with acrylic adhesive, and the laminar flow effect inside the bed was tested using ANSYS fluid simulation to exhaust from the bottom of the overall upper air intake and exhaust from the bottom of the bed. Results The sterile laminar flow bed initially achieved the effect of isolation, the flow simulation results showed that the laminar flow effect in the bed was good. Conclusion The isolation laminar flow bed based on deep ultraviolet disinfection can achieve the laminar flow effect well under the condition of meeting the isolation conditions, and has better clinical application value and significance.

Objective To analyze the reliability level of positron emission tomography/magnetic resonance imaging (PET/MR) equipment during clinical using the basic method of reliability verification. Methods The data including maintenance and work log of two PET/MR equipment (PET/MR-1 and PET/MR-2) for a certain period of time were extracted, and the reliability parameters, maintainability parameters, supportability parameters, availability and core component status of the equipment were analyzed. The reliability level of the two equipments was compared. Results The total failure time of PET/MR-2 was 253.0 h less than that of PET/ MR-1, reducing 27.4%. The mean interfailure time of PET/MR-2 was 206.2 h longer than that of PET/MR-1 (P<0.05). There was no statistical difference in the reliability parameters of the two devices (P>0.05), and the state parameters of the core components basically maintained stable operation. However, the PET/MR-1 caused two major failures of the core components due to the failure of the auxiliary system. Conclusion Based on the basic method of reliability verification, the analysis of the reliability level of PET/ MR equipment in the clinical use process can provide an objective evaluation of the long-term stability of the equipment for medical institutions, and indicate the direction of the key points of maintenance work.

Objective To explore the human-machine interface collaboration of positron emission tomography/magnetic resonance imaging (PET/MR) equipment, and to explore the clinical usability evaluation method of this kind of equipment. Methods The usability evaluation system was established by using analytic hierarchy process, an expert group was set up and the expert authority coefficient was calculated. The usability questionnaire was designed by Delphi method, and the technicians and doctors who operated the equipment were investigated, and the questionnaire scoring table of UIH PET/MR equipment was collected nationwide. Results A total of 79 survey questionnaires were collected, involving 18 hospitals. The usability evaluation score of the physiological signal gating unit interface was the lowest, and the PET passive quality control usability evaluation score was the highest. Personnel with deputy senior or senior professional titles had slightly higher scores on indicators than those with intermediate or lower levels. The scores of personnel with different operating hours on various interfaces had the same trend, and the scores of users within 3 months were significantly lower than the other two groups, with significant differences (P<0.05). There were significant differences in the scores of six specific evaluation parameters among personnel with different operating hours (P<0.05). Conclusion The usability evaluation system established based on the analytic hierarchy process can be used for the clinical usability evaluation of PET/MR devices. The usability survey questionnaire formed based on the Delphi method provides a strong reference for exploring the usability evaluation methods of PET/MR devices.

Objective To evaluate the reliability of positron emission tomography/computed tomography (PET/CT) equipment inclinical use, to analyze the key parameters affecting the quality control pass rate, and to explore the specific practical methods for evaluating the reliability of equipment in use. Methods The maintenance records of PET/CT equipment were reviewed and analyzed, and the parameters related to the availability, maintenance and maintenance of PET/CT equipment were counted. The pass rates of active quality control LUT drift, energy drift, and TOF drift of PET detectors were calculated and analyzed by Pearson correlation with environmental factors. Results Both inherent availability (Ainh) and reachable availability (Aa) consistently exceeded 98%, and the change trend of both was consistent, and Aa was slightly lower. The mean time between failure peak occurred in the fourth year and exceeded the average level of the three years prior to commissioning. The LUT drift coefficient had a 100% pass rate and was significantly negatively correlated with system humidity (r=-0.133, P=0.025). The energy drift coefficient had a 78.32% quality control pass rate, significantly correlated with the outer ring temperature (r=0.162, P=0.006). Therefore, it was advisable for operators to monitor the temperature and humidity of the machine room closely. Conclusion In the fourth year since commissioning, the PET/CT equipment has progressed from the initial stage to the normal period stage. During PET detector quality control, it was found that the energy drift coefficient often goes out of range, which may be related to environmental factors. This study outlines practical methods for evaluating the reliability of PET/CT equipment in clinical use, offering valuable reference information for improving reliability and comparing equipment.

Objective To analyze the long-term stability of magnetic resonance imaging (MRI) equipment performance using Minitab. Methods The Minitab 17.0 software was used to analyze the long-term stability of the performance of three MRI devices (MRI-1, MRI-2 and MRI-3). The analysis of performance test items included image homogeneity, signal-to-noise ratio, slice thickness, and aspect ratio. The spatial resolution of MRI was analyzed using time-series method. Results The long-term quality control process for most performance testing items was classified as level Ⅰ, with a process performance index Pp>1.67; for MRI-1 and MRI-2 devices, the process performance of all testing items was at least level Ⅱ or higher; the image homogeneity of MRI-3 equipment was classified as level Ⅳ, indicating that the influence factors should be identified and the frequency of testing should be increased. Conclusion By using Minitab to perform process capability analysis on MRI equipment, it is possible to objectively assess the performance stability throughout the life cycle of MRI equipment, which is helpful in monitoring the operation status of the equipment and determining the frequency of quality control testing; the results of process performance analysis can be used to assess the maintenance status of the equipment and determine the replacement cycle of key components. This ensures that the performance of equipment remains in a good state, provides a reliable foundation for diagnosis and therapeutic services.

Objective To study the human-computer interaction usability of linear digital radiography (DR) equipment based on the usability concept of human-computer interaction research orientation. Methods The expert group of this project discussed and determined the project indicators of usability evaluation, and established the usability evaluation model using Delphi method. By users’ scoring five domestic DR models respectively, the usability evaluation results among user groups of different using time on different models were compared. Results The mean total score and interface score of the 5 DR models showed significant differences among the three groups (P<0.05). The score of the group with less than 3 months was significantly lower than the scores of the other groups, while there was no significant difference between the other two groups (P>0.05). In addition, the usability evaluation results of different DR models were significantly different in terms of the average total score and the group with more than 1 year. Conclusion Usability evaluation can have a comprehensive view on the safety and effectiveness of DR man-machine interfaces, which is helpful to the research and development of DR equipment and the improvement of domestic equipment.

Objective To collect the eyeground color images of diabetic patients, and to establish a computer vision model, to explore its application in the referral of diabetic retinopathy (DR). Methods The eyeground images of diabetic patients in Suzhou area were collected. The training set (n=350) and the internal validation set (n=100) were from Lixiang Eye Hospital of Soochow University, and the test set (n=200) was from Suzhou Municipal Hospital. The eyeground photos were classified into referrable DR group and the non-referrable DR group. The performance of models was evaluated based on test set and compared with two ophthalmologists. In addition, the Grad-CAM technology was used to visually interpret the computer vision model. Results The performance ofEfficientNet model was the best among all models and superior to specificity for ophthalmologists with low years of experience [accuracy (ACC) was 0.840, sensitivity (SE) was 0.793, and specificity (SP) was 0.859]. However, it was slightly lower than that of senior physicians (ACC was 0.925, SE was 0.862, SP was 0.951). With the assistance of the EfficientNet model, the ACC rose to 0.915, SE to 0.828, and SP to 0.951. Similarly, the ACC of senior physicians reached 0.965, SE was 0.931, and SP was 0.979. Conclusion Computer vision models based on deep convolutional neural networks show practicable performance in DR referral binary classification, better than junior ophthalmologist. The model can assist ophthalmologists in diagnosis and improve ACC.

Objective To explore the application of automated machine learning (AutoML) based on the texture features in distinguishing early and advanced stages of esophageal squamous cell carcinoma by magnifying endoscopy with narrow-band imaging (NBI-ME). Methods A total of 1507 NBI-ME images of esophageal squamous cell carcinoma were collected from the Endoscopy Centre of The First Affiliated Hospital of Soochow University. These images were randomly divided into a training set (n=1264) and a validation set (n=243). MATLAB software was used to extract a total of 32 texture features from the whole endoscopic images. The above variables were loaded into the H2O platform for AutoML binary classification modeling. In addition, 278 endoscopic images from the Second Affiliated Hospital of Soochow University were collected as an external test set. Moreover, one junior endoscopist and one senior endoscopist were invited to interpret the images from the external test set. Receiver operating characteristic (ROC) area under curve (AUC) and accuracy (ACC) were used to evaluate the efficiency of identification. Results The AutoML model based on random forest algorithm performed best in the external test set, with AUC of 0.975 and ACC of 0.939, which was significantly better than other models, including the generalized linear model (AUC: 0.776, ACC: 0.687) and the extreme gradient boosting model (AUC: 0.968, ACC: 0.863). It was also better than that of the junior endoscopist (AUC: 0.868, ACC: 0.871) and the senior endoscopist (AUC: 0.919, ACC: 0.921). Conclusion The AutoML model based on the texture features of endoscopic images shows excellent discriminative ability in judging the staging of esophageal carcinoma.

Objective To develop computer vision models for endoscopic bowel preparation scoring based on deep learning. Methods A total of 2394 endoscopic images from the Gastrointestinal Endoscopy Centre of the First Affiliated Hospital of Soochow University (n=600) and the HyperKvasir database (n=1794) were collected, scored by endoscopists according to the Boston bowel preparation scale (BBPS, 0-3, four categories). They were randomly divided into training sets (1439 pieces), verification sets (478 pieces) and test sets (477 pieces) according to 6∶2∶2. Three deep learning networks (DenseNet169, DenseNet121, EfficientNet B3) were used to develop the bowel preparation classification models by transfer learning. Metrics such as confusion matrices in the test set were used for model evaluation. Meanwhile, the models were compared with senior and junior endoscopists. Results The three deep learning-based bowel preparation classification models were successfully developed. The classification accuracy of all models was high, and the average classification accuracy was 0.897, which was similar to the clarification performance of junior endoscopist (0.914) and lower than that of senior endoscopist (0.941). Among them, the best performing model was the DenseNet169 model, which had the highest classification accuracy (0.914) and the highest average precision (0.892). In addition, the visual interpretation of the models’ classification results was presented in the form of heat maps by using gradient-weighted class activation mapping. Conclusion The endoscopic bowel preparation classification model developed using deep learning is feasible, and the classification and generalization ability of the model can be further improved by expanding the sample source through multicenter studies.

Objective To explore the application effects of the model constructed by using deep convolutional neural networks (DCNNs) and transfer learning in the diagnosis of osteoporosis based on knee joint X-ray images. Methods The knee joint X-ray images were collected from Jintan Hospital Affiliated to Jiangsu University, which were randomly classified into training set (n=400) and internal validation set (n=100) at a ratio of 8∶2. Four DCNNs (ResNet, Xception, NASNet and EfficientNet) were selected, which had been pre-trained in ImageNet dataset, and the parameters of their first pre-training were frozen as the model framework of the single training group. A total of 5856 images were randomly selected from the Kermany Chest X-Ray2017 dataset and performed secondary pre-training on these four DCNNs as the model framework for the secondary training group. Two sets of model frameworks were used for target-training on knee joint X-ray image, and a classification model for osteoporosis was constructed. Eighty-five images were randomly selected from the Wani-Knee X-Ray2021 dataset as anexternal test set. The performance of the classification model was evaluated according to its performance in internal validation and external test sets. Results The accuracy of the models in the secondary training group was higher than that in the single training group. Among the four DCNNs, EfficientNet model performed significantly better than the other three models. In the internal validation set, the accuracy of EfficientNet model in the secondary training group was 0.918, with F1-score 0.918, and the area under curve (AUC) of receiver operating characteristic was 0.932; in the external test set, the accuracy was 0.824, F1-score was 0.848, and AUC was 0.846. Conclusion The knee joint X-ray deep learning model established by chest X-ray secondary pre-training is superior to the single pre-training model in the application of osteoporosis classification. The use of DCNNs and transfer learning techniques can help small sample datasets perform tasks such as classification and prediction more accurately.

Objective Based on DeepLab V3+ network framework, to construct a deep learning computer vision model to achieve more accurate semantic segmentation of upper gastrointestinal ulcer endoscopic images. Methods The encoder of the DeepLab V3+ network framework initially extracted features from images using multiple parallel dilated convolutional layers with varying sampling rates, followed by global average pooling to achieve multi-scale feature extraction. In the decoder section, the deep feature layer was up-sampled four times while the shallow feature layer was stacked and adjusted in size to match that of the input image before generating predictions. The prediction results of the model were obtained. Results In the internal validation set, the accuracy (ACC) of the model was 0.963, and the mean intersection over union (mIoU) was 0.927. In the external test set, the ACC and mIoU were 0.958 and 0.915. All of them were better than the traditional algorithm U-Net (internal validation set ACC was 0.810, mIoU was 0.785; the external test set ACC was 0.779 and mIoU was 0.732). Conclusion The DeepLab V3+ network framework has high accuracy in identifying lesions, and has good clinical practice.

Cerenkov luminescence (CL) is a luminescence phenomenon with spectral peaks located near blue-violet light, which has been extensively studied in medical diagnosis and treatment since its discovery. Medical imaging technology based on Cerenkov luminescence has the characteristics of remote, non-invasive and efficient, and is expected to become a reliable dose assessment tool in clinical radiotherapy. On the basis of expounding the mechanism of CL phenomenon, this paper further reviews the related technology in tumor radiotherapy, in order to provide more reference for the monitoring, evaluation and quality assurance of tumor radiotherapy.

Pulmonary diseases have a significant impact on human health and life safety, and abnormalities in the lungs are a direct response to lung diseases. The study of lung sounds is of great significance in clinical diagnosis. With the continuous development of electronic auscultation technology, the feature extraction and classification techniques of modern lung sounds have also been further studied. At present, the research of lung sound classification is to combine signal analysis with deep learning algorithm to improve the practical application of auxiliary electronic stethoscope. This paperaims to elaborate on the concept of lung sounds, analyze the current development status of lung sound classification, and discuss the shortcomings of lung sound classification technology. The goal is to an outlook on future research directions and application development trends in the field of lung sound classification.