ObjectiveTo comply with the regulatory requirements of national pharmaceutical authorities for real-time temperature and humidity monitoring of pharmaceuticals and medical consumables during usage, we developed a hospital cold chain monitoring system suitable for clinical operation.MethodsThe system architecture incorporates:Hardware: Huawei Hi3861 chip (RISC-V architecture) running HarmonyOS LiteOS-M;Sensors: SHT30 high-precision modules (±0.1°C for temperature, ±1.5%RH for humidity);Cloud platform: Huawei Cloud IoT for data storage and analytics;Applications: Multi-terminal software solutions;This HarmonyOS-based intelligent monitoring system achieves real-time precision surveillance and early warning for medical environments..Results Performance metrics demonstrated:Monitoring accuracy:Temperature error: 0.3±0.2°C (t=6.352, P<0.001),Humidity error: 2.1±0.9%RH (t=4.781, P<0.01),Response performance:Alarm latency: 4.2±0.8s (Z=12.463, P<0.001),Compatibility:Supports 6 terminal types (mobile/tablet/PC etc.),Data transmission success rate: 99.2±0.5% (Z=5.217, P<0.05). ConclusionThe system is not only able to monitor the refrigerator, cold storage and other cold chain equipment in each department of the hospital, but also can monitor the temperature and humidity environment of the large equipment room, which can meet the various needs of the hospital for temperature and humidity monitoring. The system comprehensively monitors various hospital environments including:Refrigerators and cold storage units,Large equipment rooms;While fulfilling all clinical monitoring requirements, it achieves:100% localization of critical components;52.3% cost reduction compared to imported systems;Scalability potential for blood bank and logistics applications

Uric acid test reagent is a rapid diagnostic reagent used for quantitative detection of uric acid concentration in human whole blood samples in vitro. In clinical practice, they are mainly used to monitor uric acid levels . The article focuses on the technical evaluation points of uric acid detection reagent product registration, and provides suggestions and references for product registration applicants and regulatory reviewers.

ObjectiveTo explore the solution of the electronic form system based on the PaaS mode, optimize the full-process management of low-value consumables in PIVAS and improve the level of informatization. Methods Utilizing the PaaS-based BaiShu low-code development platform, with front-end interaction, business logic, data services, and application output as the foundational architecture, and incorporating features such as user-friendly operation, automatic loading, expiration warnings, inventory verification, and real-time inventory statistics, an electronic form management system was designed to cover core processes including acceptance of incoming goods, consumable requisition, consumable return, and consumable inventory. The application effectiveness was evaluated through data comparison and questionnaire surveys. Results A three-month period before and after system implementation was selected as the control and study groups, respectively. The average account-material match rate increased from (89.03±4.89)% to (95.53±1.95)% (P<0.05). The average registration time decreased from (193.88 ± 20.91) s to (108.04 ± 10.94) s (P<0.05). ; 95.45% of the respondents recognized the system functions and were willing to continue using them. 86.36% believed that the data accuracy had significantly improved, and 81.81% thought it could effectively enhance work efficiency. However, there is still room for improvement in interface design and compatibility with mobile devices. ConclusionThe construction of an electronic form management system based on the low-code development platform of the PaaS model can effectively solve the shortcomings of traditional manual management, significantly improve the management efficiency and data reliability of low-value consumables in PIVAS, and provide medical institutions with a low-cost and easy-to-implement informatization path. In the future, we can explore the expansion of application scenarios and attach importance to data security

Based on the working principle of ultrasound equipment, this study systematically analyzes the troubleshooting and maintenance processes for three typical malfunctions of the Philips EPIQ Cvx ultrasound system (intermittent image quality degradation of multiple probes, stripe interference of single probe, and occasional error reporting). By employing a "hierarchical troubleshooting + modularized approach" combined with the equipment architecture, the study accurately identifies the source of the malfunctions and proposes solutions. Additionally, targeted preventive maintenance strategies are proposed, including regular performance testing of channel boards, increased monitoring of power supply environmental conducted noise, standardized software configuration backup management, and enhanced probe sensitivity calibration and electrical safety testing. This study provides a comprehensive solution for ultrasound equipment maintenance, encompassing "fault localization, module repair, and preventive maintenance," offering practical guidance for maintenance engineers. This helps to improve maintenance efficiency and ensure the stable operation of ultrasound equipment.

Objective Through investigating the configuration and application effect of domestic in-use medical magnetic resonance (MR) cold water machines, constructing an application effect evaluation method, and forming relevant conclusions, this study aims to assist in improving the management level of medical equipment in medical institutions and enhancing the quality of medical imaging examinations, thereby promoting the better development of domestic MR and its supporting equipment industries. Methods Through expert interviews, questionnaire surveys, and a large-sample sampling survey of MR cold water machine equipment in 123 medical institutions in seven major regions of China for the first time, the configuration of MR cold water machine equipment was comprehensively sorted out, and the application effect was analyzed. Results In terms of equipment configuration in domestic in-use MR cold water machines: The brands of MR cold water machine equipment manufacturers are concentrated in three companies, and the parameter differences of different equipment mainly lie in cooling efficiency and working mode, etc.; In terms of equipment usage effect: The maintenance cost of MR cold water machine equipment is relatively reasonable, the main maintenance mode is original factory warranty, the average annual number of faults is concentrated in 0-2 times, the maintenance response time is concentrated within 24 hours, and the overall satisfaction with equipment usage is good. ConclusionBased on the large-sample data survey, the constructed evaluation method for the application effect of MR cold water machine equipment can comprehensively and scientifically evaluate the application effect of the equipment, and has important guiding significance for medical equipment management and the application of MR equipment.

ObjectiveIn response to the prevalent clinical demand for managing upper-extremity hypertonia, this study presents a device designed to alleviate elevated muscle tone. Grounded in the principle of continuous passive motion (CPM) and developed in strict accordance with clinical guidelines, the apparatus enables controlled, repetitive mobilization of the upper limb to safely reduce spasticity. MethodsThrough the extraction and analysis of rehabilitation training movements for patients with clinically presented upper limb muscle hypertonia, the requirements for the device to achieve multi-joint coordinated functions involving the shoulder, elbow, wrist, and fingers have been clarified. Based on these requirements, the mechanical component adopts an exoskeleton-type wearable design to ensure motion trajectory accuracy and overall stability; joint motors employ the Controller Area Network (CAN) protocol for multi-degree-of-freedom coordinated control, precisely simulating shoulder and elbow rehabilitation movements; and a pneumatic glove enables finger flexion and extension training through air pressure actuation. Results Tests indicate that the device exhibits a motion cycle error of less than ±0.13 s, joint movement angle deviation within ±1°, and noise levels below 57 dB. Its electrical safety complies with the GB 9706.1–2020 standard, with both touch current and patient leakage current under normal conditions remaining below 100 μA. The structural design and functional parameters of the device strictly adhere to relevant laws and regulations. ConclusionThe device developed in this study is grounded in the CPM principle. By means of parameter-based design, it replicates therapeutic motion patterns to deliver continuous passive rehabilitation, thereby preserving joint range of motion and addressing abnormally elevated upper-limb muscle tone secondary to stroke and related conditions.

ObjectiveAddressing issues such as data silos, lack of transparency in processes, and difficulties in traceability in traditional medical device management models, this study aims to develop a blockchain-based comprehensive medical device management system to achieve end-to-end trustworthy management from departmental procurement to clinical use. Methods A consortium blockchain architecture was adopted to integrate nodes from various hospital departments. An improved PBFT consensus algorithm (transaction confirmation time <500ms) and a hybrid storage solution combining ‘on-chain hashing + off-chain IPFS’ were designed; a smart contract system encompassing five core modules, including equipment lifecycle and approval decision-making, is developed; and an innovative dynamic permission management mechanism based on attribute encryption is implemented. Results Following system implementation, key performance indicators for equipment management demonstrated significant improvement compared to pre-implementation levels: equipment turnover rate increased by 31.5%, while deployment response time decreased by 73.3%. Regarding data quality and security management, maintenance record completeness reached 100%, with 56 instances of unauthorised access successfully intercepted. In terms of system processing capacity, it consistently handled an average of 230,000 management transactions daily. All improvements in key metrics were statistically significant (P<0.05).Conclusion This study demonstrates that blockchain technology can significantly enhance the efficiency and security of medical device management. The proposed lightweight on-chain solution and dynamic permission model provide innovative solutions for medical blockchain applications, holding significant practical value for advancing hospital digital transformation.

ObjectiveTo address issues related to inaccurate monitoring, low maintenance efficiency, and poorly managed spare parts caused by multifactorial coupling faults under high pressure and frequency conditions in the expiratory flow sensor of the Hamilton G5 invasive ventilator, thereby enhancing the operational safety and cost-effectiveness of ICU critical care equipment.MethodsThis study proposes an integrated method combining multi-source signal fusion diagnosis and hierarchical collaborative maintenance. By integrating real-time high-pressure waveform features, airtightness parameters, and historical trends, a dynamic threshold adaptive diagnostic model was constructed. A three-level maintenance mechanism—"rapid calibration – component replacement – system maintenance"—was designed, and a Weibull-based life prediction model was used to optimize spare part inventory for high-pressure ventilation circuits.Results The model achieved a fault identification accuracy of 92.8% and reduced the false alarm rate to 6.2%. Maintenance efficiency increased by 57.1%, spare part turnover improved by 133.3%, and the stock-out risk was controlled within 1.5%.ConclusionThe study develops an end-to-end intelligent operation and maintenance solution for invasive ventilator sensors. Through interdisciplinary integration and engineering validation, it provides theoretical and practical support for data-driven management of critical medical equipment and promotes a shift in clinical operation and maintenance towards proactive prevention.

ObjectiveTo conduct electromagnetic compatibility design for an intelligent cardiopulmonary resuscitation machine to improve its pass rate in electromagnetic compatibility tests and shorten the registration time. MethodsKey components required for the intelligent cardiopulmonary resuscitation machine were studied and analyzed. Based on electromagnetic compatibility standards such as YY 9706.102-2021 and GB 4824-2019, conducted emission tests were performed for different installation positions of the power filter; radiation emission tests were conducted for the presence or absence of shielding in the switching power supply, whether cables were separately bundled, grounding of the servo motor and PCB board, etc. The data before and after the tests were compared and analyzed to explore their impact on conducted disturbance and radiated disturbance.Results Installing the power filter near the power inlet can effectively reduce conducted disturbance quasi-peak by more than 25dB (μV) and conducted disturbance average by more than 21dB (μV); selecting a switching power supply with transformer shielding can effectively reduce radiated disturbance quasi-peak by more than 6dB (μV/m); grounding the servo motor housing can effectively reduce radiated disturbance quasi-peak by more than 7dB (μV/m); separating and bundling cables during wiring can effectively reduce radiated disturbance quasi-peak by more than 15dB (μV/m); grounding the PCB board and grounding near external connectors can effectively reduce radiated disturbance quasi-peak by more than 15dB (μV/m).ConclusionThe intelligent cardiopulmonary resuscitation machine, through designs such as shielding, filtering, and grounding, can effectively enhance its electromagnetic compatibility, providing reference value for subsequent research on electromagnetic compatibility of intelligent cardiopulmonary resuscitation machines.

Objective To establish a clinical application evaluation framework for medical equipment, guiding procurement planning and project approval of medical equipment in hospital.Methods Issues across the full lifecycle of medical equipment were systematically analyzed. An initial clinical application evaluation framework was constructed. The clinical application evaluation framework was improved through expert consultation. The weights of each level of indicators were acquired by Analytic Hierarchy Process and Entropy Weight Method. Clinical evaluations of infusion pumps from three mainstream brands were conducted using institutional data. Results A comprehensive evaluation framework was developed, comprising 4 first-level, 4 second-level, and 24 third-level indicators. The Analytic Hierarchy Process determined that clinical usage evaluation carried the highest weight among first-level indicators, while safety held the highest weight among second-level indicators, indicating that clinical application and safety performance were the most valued factors by experts. Through the application of this clinical evaluation framework to three brands of infusion pumps in the hospital, Brand C's infusion pump exhibited optimal performance in practical applications.Conclusion This lifecycle-oriented evaluation system, characterized by structural integrity and high reliability, provides critical reference value for clinical evaluation, selection, and deployment of medical equipment.

ObjectiveTo address the issues of high construction costs, large data analysis volumes, and low efficiency in existing medical record management systems, an intelligent medical record information management system is designed that considers both cost and operational risks, in order to achieve automation in medical record information management. MethodsBased on the cloud platform, optimize the medical record information management system from three aspects: data analysis, storage access, and functional design of the medical record system. This includes building a MySQL database structure, designing a data interaction framework and automatic query mode, and constructing a Naive Bayes disease diagnosis auxiliary model. The designed intelligent case information management system is examined and analyzed with the help of accuracy, recall, evaluation metrics, throughput, response time, and cost. Results Plain Bayes outperforms other comparative methods in terms of accuracy and recall over 0.85 for disease classification with inference delay values less than 2ms. The throughput response time exhibited by the research-designed cloud platform server is short, and the batch statistics time for data access of case information is 1.8s, and the risk of data loss is only 0.001%, which is a good application. ConclusionThe intelligent medical record information management system designed for research can effectively meet the needs of medical information management, improve application efficiency, and reduce operation and maintenance costs and risks.

ObjectiveBased on the HM-J-16-I accelerator, ultra-high dose rate electron FLASH (with a dose rate exceeding 40 Gy/s) was achieved through modification and precise dose control was realized. The electron beam energy and dose distribution under the electron FLASH state were verified to meet the requirements of GB15213 for conventional electron beams.MethodsBased on the 6MeV and 10MeV states, the dose rate was increased by significantly boosting the injected beam current, removing the ionization chamber, reducing the thickness of the primary and secondary scattering foils to minimize dose rate attenuation, and lowering the SSD to increase the dose rate to reach the electron FLASH threshold. The dose was precisely controlled by the pulse control unit through the control of the number of pulses.Results The highest dose rate of the 6MeV electron FLASH reached 240 Gy/s, and that of the 10MeV electron FLASH reached 415 Gy/s. The dose output stability was better than ±2%. The determination coefficients R²for 10MeV FLASH dose linearity were 0.99997 (SSD705 mm) and 0.99997 (SSD1000mm), both exceeding 0.999. The dose rate output stability was better than ±2%, with penetration index R50 at 41.4mm(SSD705mm)41.1mm (SSD1000mm).The energy variation was less than ±0.5 MeV, The flatness of the 10MeV electron FLASH at the standard depth was 5.5 mm, and the symmetry was 101.5%, and central offset CAX.Dev=0.55mm. ConclusionThe tests and analyses of the dose rate, dose, percentage depth dose curve, and off-axis ratio curve of the electron FLASH of the HM-J-16-I accelerator demonstrated that the dose rate fully met the requirements of electron FLASH and far exceeded the threshold of 40 Gy/s. Moreover, the stability of the output dose, energy, symmetry, and homogeneity all met the requirements of GB15213 for conventional electron beams. This provides important dosimetric parameters and beam characteristics for the further research and application of FLASH-RT, accelerating the development of electron FLASH in medical applications.

Acute altitude sickness is a frequent and common disease in the plateau population, and in severe cases, it will rapidly develop into high-altitude pulmonary edema and high-altitude cerebral edema, which is an important factor threatening the life safety of people. A large number of studies have confirmed the effectiveness of various objective physiological indicators such as blood pressure, blood oxygen saturation, heart rate and sleep in predicting acute altitude sickness. At present, the development of portable devices is becoming more and more mature, with the characteristics of real-time, dynamic and continuous monitoring of physiological indicators, and plays an increasingly important role in the fields of physiological monitoring and disease prevention. The application of portable devices to physiological monitoring of the plateau population has certain advantages in the prevention of acute altitude sickness. This paper aims to systematically review the application of portable monitoring devices in acute altitude sickness and highlight their application value. This paper reviews the current research status of portable devices at home and abroad from the monitoring of physiological parameters such as blood pressure, blood oxygen saturation, heart rate, and sleep, analyzes the characteristics and application effects of different types of portable devices, and discusses the intelligent prediction methods of acute altitude sickness based on data mining and artificial intelligence, as well as the challenges and countermeasures faced by system integration.

Objective To evaluate the effectiveness and correlation of corneal wavefront aberrations and double-pass optical quality analysis (OQAS) in the assessment of visual quality after small-incision lenticule extraction (SMILE). Methods This prospective study included 47 patients (93 eyes) who underwent SMILE surgery at the Affiliated Hospital of Yunnan University. Visual acuity, refraction, corneal wavefront aberrations, and OQAS-based visual quality assessments were conducted preoperatively and at 6 months postoperatively. Corneal wavefront aberrations were measured using Pentacam, with parameters including trefoil 0° [Z(3,3)], coma 0° [Z(3,1)], coma 90° [Z(3,-1)], trefoil 30° [Z(3,-3)], tetrafoil 0° [Z(4,4)], fourth-order astigmatism 0° [Z(4,2)], spherical aberration [Z(4,0)], fourth-order astigmatism 45° [Z(4,-2)], and tetrafoil 22.5° [Z(4,-4)]. OQAS parameters included modulation transfer function cutoff frequency (MTF cutoff), objective scatter index (OSI), and Strehl ratio (SR). Statistical analysis was performed using SPSS 19.0 to compare pre- and postoperative indices and analyze correlations. Results At 6 months postoperatively, uncorrected visual acuity (UCVA) improved to 1.11±0.15, and refractive error was -0.12±0.05 D, indicating satisfactory correction. Root mean square (RMS) of higher-order aberrations increased significantly from 0.333±0.081 to 0.702±0.207 (p < 0.05), with notable increases in Z(3,-1) and Z(4,0) by 0.368±0.211 and 0.165±0.114, respectively. No significant differences were observed in OSI, MTF cutoff, or SR between pre- and postoperative measurements (p > 0.05). Correlation analysis showed only weak (R<0.3) or statistically insignificant associations between corneal wavefront aberrations and OQAS parameters, both pre- and postoperatively as well as in their changes. ConclusionCorneal wavefront aberrations and the double-pass optical quality analysis system each have distinct advantages in evaluating visual quality after SMILE. The combined application of both methods contributes to a more comprehensive assessment of postoperative visual quality, providing valuable reference for the design of personalized surgical plans and postoperative management.

General Medical Artificial Intelligence (GMAI) is reshaping medical practice with its technological commonality and ability to emerge across scenarios of intelligence. This article first outlines the concept and technical characteristics of GMAI, pointing out its potential in clinical decision-making, imaging diagnosis and other applications, but there are also risks of incomplete, unpredictable decision-making and data dependence; Secondly, the difficulties of legal regulation are analyzed. The special challenges are embodied in the dimensions of data dependence, decision-making black box characteristics, model bias transmission and risk overflow, which are difficult to be covered by the existing medical device regulatory framework. For this, the EU implements strict supervision with the Artificial Intelligence Bill, and the United States is guided by soft law. There are significant differences in the regulatory threshold and the applicable rules of medical AI between the two; Finally, it proposes ways to improve, incorporate generic model providers into the scope of responsibility subjects, build a normative system of “basic security+scenario adaptation”, give full play to the functions of anti-monopoly law, prevent data monopoly and tying, strengthen ethical constraints, establish data governance and dynamic review mechanisms, innovate regulatory paradigms to deal with derivative risks, and achieve the effect of promoting development and escort safety.

Objective This study aims to develop an interpretable deep learning model based on multi-feature fusion for bowel sounds frequency classification and evaluate its application in determining the timing of feeding after general anesthesia. Methods A multi-center prospective study design was adopted. Bowel sounds from the right lower abdominal quadrant were collected using a digital electronic stethoscope, and audio data (60 seconds) were gathered from four hospitals. After preprocessing and segmenting the audio into 10-second clips, three features—Chroma, filter bank energy (FBank), and Mel-frequency cepstral coefficients (MFCC)—were extracted and converted into RGB images as input for the ResNet50-V2 model. Subsequently, a long short-term memory (LSTM) network was employed to fuse and classify the temporal features. Evaluation metrics included accuracy, Matthew’s correlation coefficient, and Cohen’s Kappa coefficient. Additionally, the Local Interpretable Model-agnostic Explanations (LIME) method was introduced to enhance model transparency. Results The LSTM model achieved an accuracy of 0.855, a Matthew’s correlation coefficient of 0.799, and a Cohen’s Kappa coefficient of 0.780, outperforming two physicians (a junior physician: 0.821, 0.726, and 0.730; a senior physician: 0.848, 0.770, and 0.770). In clinical applications, the model demonstrated excellent performance in determining postoperative feeding timing for patients under general anesthesia, achieving an accuracy of 92.7%. Conclusion The ResNet-LSTM-based multi-feature fusion model shows high applicability in bowel sounds classification and postoperative dietary management.

ObjectiveTo evaluate the long-term stability of spatial resolution of a Siemens CT-simulator, by analyzing the modulation transfer function (MTF) data from monthly constancy tests of the CT-simulator. MethodsThe MTF data of spatial resolution (high contrast resolution) for a Siemens Sensation Open CT-simulator were obtained from the monthly QA tests, by using the phantom set and constancy test program. Statistical analysis and comparative study were conducted on the MTF data for a total of 72 months from January 2019 to December 2024. Results Three sets of MTF data (50% MTF, 10% MTF, and 2% MTF) were obtained for three groups (BS1, BS2, and H). The 50% MTF, 10% MTF, and 2% MTF of Group BS1 are (unit: lp/cm): 4.78 ± 0.13, 9.93 ± 0.12, 12.68 ± 0.15, respectively; three sets of MTF values of the Group BS2 are (unit: lp/cm): 4.78 ± 0.16, 9.93 ± 0.12, 12.66 ± 0.15, respectively; three sets of MTF values of Group H are (unit: lp/cm): 4.96 ± 0.10, 10.16 ± 0.09, and 12.77 ± 0.12, respectively. The paired t-test results showed that for each set of MTF data, there were statistically significant differences (P<0.05) between H and BS1, and between H and the BS2, while there was no statistically significant difference (P>0.05) between BS1 and BS2. Conclusion From data analysis, it can be seen that the spatial resolution of the CT-simulator was relatively stable for the 6 years, meeting the requirements for clinical use.

Uncontrollable physiological activities such as breathing and heartbeat can cause complex dynamic deformations in soft tissues, which far exceed the correction range of rigid registration and result in significant spatial displacement of anatomical structures in images at different time phases, leading to unsatisfactory artifact correction effects. Therefore, a magnetic resonance imaging artifact correction method based on improved ORB algorithm is proposed. By optimizing the feature point extraction strategy and descriptor construction method, combined with local adaptive gamma correction to enhance the response of strong and weak texture regions, the feature matching accuracy is significantly improved. Based on grid motion statistics and RANSAC algorithm, matching point pairs are selected to achieve spatial pose calculation and rigid body transformation registration. Based on the registration results, the snake algorithm is used to extract the contour of the region of interest, and the motion artifact offset is estimated through Fourier transform. The artifact correction is completed through phase recovery and reverse offset. The experimental results show that the average peak signal-to-noise ratio of the proposed method reaches 32.78 dB, and the multi-scale gradient phase consistency index remains stable in the high-level range of 0.92-0.97 in 10 to 100 image tests, and still maintains MSPC values of 0.87 to 0.93 in 216 multi parameter disturbance combination tests. Therefore, it indicates that the proposed method effectively solves the problem of spatial misalignment caused by physiological movement, providing a more accurate and stable solution for MRI image artifact correction.

Currently, CT serves as the primary imaging modality for pulmonary nodule screening. However, its inherent ionizing radiation risk has prompted clinical exploration of safer alternative imaging technologies. MRI, leveraging its absence of ionizing radiation, superior soft-tissue contrast, and multi-parametric imaging capabilities, demonstrates significant clinical potential in the detection and characterization of pulmonary nodules. Advances in rapid imaging sequences, such as ultrashort echo time (UTE) and zero echo time (ZTE), coupled with hardware upgrades and the integration of deep learning techniques, have substantially improved the sensitivity of thoracic MRI in nodule detection. This article systematically reviews recent progress in MRI applications for pulmonary nodule screening, with a focus on critical scientific issues including current diagnostic efficacy, clinical utility, advancements in artificial intelligence, and challenges in clinical implementation. This review aims to promote a comprehensive understanding of the capabilities and limitations of MRI in pulmonary nodule screening, with the goal of providing references for optimizing MRI screening protocols, enhancing diagnostic efficacy, and guiding future standardized and rational application.

To deepen cross-strait economic cooperation and implement the State Council's strategic plan for establishing a demonstration zone for the integrated development across the Taiwan Strait, the National Medical Products Administration (NMPA) has introduced a specialized work plan supporting Fujian's pioneering initiatives in pharmaceutical and medical device regulation. Establishing a regulatory reliance mechanism is crucial for promoting integrated development in the cross-strait medical device industry and ensuring public safety. This article employs literature review and comparative analysis to examine the similarities and differences between China’s mainland and Taiwan region concerning medical device regulatory structures, legal frameworks, definitions, classification systems, and market approval requirements. The study finds significant similarities in the medical device regulatory systems and comparable industrial development levels across the strait, providing a foundation for regulatory reliance. By analyzing global models of cross-regional regulatory cooperation within international pharmaceutical and medical device organizations, it recommends exploring a distinctive cross-strait regulatory reliance pathway in Fujian. Specific proposals include: Establishing a bilateral，dialogue platform，Building a digital collaboration platform，Developing a technical review quality system. These measures aim to provide innovative approaches for the integrated development of the cross-strait health industry.

With the improvement of people's pursuit of quality of life, dental implant restoration has become the main treatment choice for patients with dentition defects, and its precision and safety have always been the research hotspots. In recent years, the introduction of Mixed Reality (MR) Technology has provided new design and navigation solutions for dental implant surgery. Mixed reality technology offers doctors intuitive and accurate surgical navigation through the integration of the virtual and the real. At present, it is believed that Mixed Reality Technology can significantly improve the precision of implant placement, reduce operation time and complications, and improve patient satisfaction. This technology shows substantial potential in dental implant navigation and is expected to become the mainstream navigation method in the future. This review aims to review the current applications, technical features, clinical effects, and future development trends of mixed reality technology in dental implant surgery.

ObjectiveTo explore the evaluation index system that combines the expert experience with the structured analysis framework to accurately reflect the technical performance of different brands of flow diverter. Methods The Delphi method was used to conduct two rounds of expert consultations on 17 experts from clinical care, production research and development, consumables management, and selection and procurement, and then the analytic hierarchy process was used to determine the weight coefficient of the evaluation index, and based on the obtained evaluation index system to guide the procurement and selection of such consumables in hospitals. Results A systematic, multi-dimensional and high-standard quantitative evaluation index system for flow diverter was constructed. The weight distribution of each three-level index generally showed the characteristics of “safety-oriented, clinical effectiveness and economy, and relative weakening of technical characteristics of instruments”, which can effectively enhance the objectivity of flow diverter stent evaluation and the flexibility of decision-making. ConclusionThe establishment of the evaluation index system of flow diverter can provide multi-dimensional decision-making basis for the selection of stents by procurement selection experts and the optimization of clinicians’ surgical plans, and also provide data support for the standardized evaluation, technical specification formulation and procurement selection of such high-value consumables.

​Objective To develop and implement an intelligent medical equipment management system based on Internet of Things (IoT) technology, aiming to address the inefficiencies and tracking difficulties in hospital equipment center management. The system enables precise and efficient management of hospital equipment, enhances medical equipment utilization and maintenance quality, and provides critical infrastructure support for building smart hospitals.Methods The project leverages core IoT architecture and mobile internet technology. It integrates Radio Frequency Identification (RFID) technology for automatic equipment identification and binding. Smart cabinet terminals (equipped with touchscreens, RFID readers, and lock control mechanisms) are deployed as user interaction and equipment access points. A cloud-based backend management system is established to unify data management for equipment information, user permissions, borrowing workflows, and maintenance schedules.​Results Newly deployed system increased average monthly loan frequency by 27.5% and reduced average loan processing time by 38.4%.​Conclusion Practical application demonstrates that this platform significantly improves equipment management efficiency and utilization rate, optimizes resource allocation, and reduces operational maintenance costs and management risks.

The rapid development of mobile intelligent technologies has catalyzed significant progress in the home-based rehabilitation of stroke patients with limb motor dysfunction. Currently, technologies such as virtual reality, exoskeleton robots, wearable devices, smart home IoT systems, rehabilitation management apps, and remote video guidance offer significant benefits in improving rehabilitation adherence, reducing costs, and enhancing patients’ self-management. Nevertheless, they face challenges including complex operation, data security and privacy concerns, high costs, and strong reliance on network connectivity. This review synthesizes current applications, outcomes, and limitations of mobile intelligent technologies in the home rehabilitation of stroke patients with motor dysfunction, aiming to provide insights for advancing the field of home rehabilitation for stroke survivors.

In recent years, the biomechanics of eye especially corneal biomechanical properties and its relationship between eye diseases become a research hotspot. Because of the deformation of cornea under external force can be affected by many factors, and the interplay between these factors is complex, the measurement of cornea biomechanical and corneal Stiffness in vivo has always been a problem in the field of ophthalmology. Under different stress conditions, cornea will produce different deformation responses, so it is difficult to evaluate the effect of a single factor on corneal biomechanics. The Corneal Visualization Scheimpflug Technology (Corvis-ST) updated a new index that reflecting corneal material Stiffness --SSI (stress-strain index) in 2019, which can intuitively reflect the stress strain relationship of corneal,It is believed that SSI is not affected by CCT,IOP and other factors, making it possible to measure corneal Stiffness in vivo.Based on the introduction of SSI, this paper will review its clinical application in the field of keratoconus, refractive surgery and glaucoma. The existing problems and future research directions are put forward to provide reference for their application and development.

Prediabetes is a high-risk stage for type 2 diabetes, and timely intervention is particularly important for patients with prediabetes. Traditional self-management approaches are ineffective in managing the condition. In recent years, digital health technology has gradually emerged as an efficient means for prediabetes patients to conduct self-management. It has effectively improved patients' self-management capabilities and reduced the incidence of type 2 diabetes. This article conducts a scoping review on the types of technologies, intervention contents, and outcome indicators of digital health technology applications in prediabetes, analyzes its application effects, shortcomings, and challenges, and provides references for the promotion of digital health technology in the self-management of prediabetes.

ObjectiveTo address image evaluation challenges arising from vendor-specific parameter variations, lack of standardized phantoms, and inconsistent resolution of custom-made phantoms in OCT-IVUS dual-modality imaging systems, this study developed a high-resolution, dynamically adjustable test phantom to meet cross-platform imaging compatibility requirements.​Methods​​ The phantom adopted a ​​modular dual-structure design​​, comprising a ​​3D positioning module​​ and a ​​tungsten wire spacing adjustment module​​. The main structure was constructed from ​​imported low-temperature chrome-plated stainless steel​​, enabling ​​micrometer-level precision adjustment​​ of catheter spatial position and tungsten wire spacing via ​​micrometer screw knobs​​. The accuracy of the phantom’s ​​3D displacement​​ and tungsten wire spacing adjustment modules was systematically verified using an ​​optical three-coordinate measuring machine​​ (MPE: ±(0.25 + L/900) μm). Results For the ​​three-axis (X/Y/Z) adjustment module​​, the ​​absolute displacement deviation​​ was <0.1 mm, with a ​​repeatability standard deviation​​ ≤0.00362 mm. The tungsten wire spacing adjustment module achieved an ​​adjustment range​​ of up to 10 mm, with an ​​absolute adjustment deviation​​ <0.025 mm, a ​​coefficient of variation (CV)​​ <2.5%, and a stable ​​systematic error​​ (mean: -0.014 mm). ConclusionThis phantom achieves micrometer-level precision and wide-range high-resolution adjustment, providing a standardized, highly reliable performance evaluation tool for OCT-IVUS devices, which is significant for promoting multi-center result mutual recognition and industry standard unification.

ObjectiveTaking a public tertiary general hospital as an empirical setting, this study aims to design and implement a fully digitalized intelligent examination appointment system based on the deep integration of large language models (LLMs) and deep learning-based optical character recognition (OCR). The goal is to reconstruct the medical resource scheduling paradigm through intelligent technologies, achieving optimal allocation of medical examination resources and significant improvement in patient experience.MethodsAn intelligent system integrating LLMs and OCR technologies was developed. The system consists of a three-layer architecture including data interaction, intelligent engine, and business application layers, with performance compared between traditional and intelligent modes through controlled experiments.Results.After the system was launched, the number of examinations increased by 32.84%, the appointment rate rose from 51.35% to 69.54%, the rate of rescheduling decreased by 3.1%, the efficiency of processing paper documents improved by 77.66%, the appointment time was significantly reduced, patient satisfaction was notably enhanced, and the return on investment was as high as 15 times.ConclusionThis study validates the synergistic effect of LLMs and OCR technologies in medical appointment systems, providing an effective technical solution for medical resource scheduling challenges. The system achieves optimal resource allocation and process reengineering through intelligent methods, offering a replicable practical model for smart hospital construction.

ObjectiveTo evaluate the effect of the Left Ventricular Assist Device (LVAD) on platelet activation by detecting platelet activation markers. MethodsBlood from the five volunteers was circulated for 6 hours, with one in vitro circuit and pump per volunteer. A total of seven blood samples were collected: one baseline sample (T0) and one sample per hour thereafter (T1–T6). The control group consisted of blood maintained under the same conditions but not circulated through the pump. Each sample was analyzed for platelet count and the platelet activation markers PAC-1, CD62P, vWF: Ag, and D-Dimer. Results The platelet count in the experimental group (ranging from 111.1429 10⁹/L to 167.1429 10⁹/L) was lower than that in the control group (ranging from 122.5714 10⁹/L to 183.7143 10⁹/L). Although this difference was statistically significant, all values remained within clinically acceptable limits. No significant differences were observed in the platelet activation markers (PAC-1 <5%, CD62P <5%, D-Dimer <0.5 mg/L, vWF: Ag 40%–160%) between the experimental and control groups, except for Volunteer #4, where significant differences were found in D-Dimer (p = 0.036) and vWF:Ag (p = 0.048). ConclusionEvaluating the effect of the blood pump on platelet activation by measuring platelet activation markers in an in vitro circulatory circuit is crucial for ensuring the efficacy and safety of the pump. This approach also provides a basis for developing dynamic, personalized antithrombotic therapy strategies for patients following LVAD implantation.

Parkinson's disease (PD) is a progressive neurodegenerative illness with both motor and nonmotor symptoms. Deep brain stimulation (DBS) is an established safe neurosurgical symptomatic therapy for eligible patients with advanced disease in whom medical treatment fails to provide adequate symptom control and good quality of life, or in whom dopaminergic medications induce severe side effects. Advances in engineering and imaging along with an improved understanding of brain disorders are poised to reshape how DBS is viewed and delivered to patients. This review summarizes the recent technological advances in deep brain stimulator.

Stenosis in autogenous arteriovenous fistulas (AVF) and arteriovenous grafts (AVG) represents a frequent clinical challenge in patients with end-stage renal disease undergoing hemodialysis. Covered stents have emerged as a significant treatment alternative following failed percutaneous transluminal angioplasty or in cases of complex lesions; however, targeted technical review guidelines for such devices remain underdeveloped within China’s regulatory framework. The U.S. Food and Drug Administration (FDA) has established considerable expertise through the evaluation and approval of multiple such covered stents, providing a valuable reference for informing regulatory policy in China. This study comprehensively reviews recent FDA regulatory approvals of such covered stents. For non-clinical evaluation methodologies, critical bench testing parameters and animal model design considerations are detailed. In terms of clinical assessment, pivotal trial design elements such as clinical design style, inclusion/exclusion criteria, follow-up duration and clinical endpoints are analyzed. Furthermore, the authors present their evaluation perspectives on these covered stents, with the aim of providing insights for product design verification/validation, clinical evaluation and technical assessment of such devices.

ObjectiveWith the rapid development of medical imaging technology, early detection and diagnosis of brain tumors are of great significance in clinical medicine. Brain magnetic resonance imaging (MRI) provides essential lesion information for physicians. However, traditional manual diagnosis is not only time-consuming but also easily influenced by subjective factors, highlighting the need for advancing intelligent healthcare systems.MethodsThis study employed the BraTS2021 public brain tumor MRI dataset, which was preprocessed through standardization and multimodal fusion to construct the training, validation, and test sets. Based on the U-Net architecture, the proposed model incorporates multi-scale feature fusion and an attention gate mechanism, combined with an improved upsampling strategy and a composite loss function, to enhance the recognition of complex boundaries and small tumor regions.Results On the BraTS2021 test set, the proposed model achieved a Dice score of 0.914, a Jaccard index of 0.856, a Hausdorff Distance of 6.011, and an ASD of 1.066. Compared with U-Net, U-Net++, and Attention Gate U-Net, the Dice score improved by approximately 5.54%, 1.67%, and 0.99%, respectively. Ablation experiments demonstrated that both multi-scale feature fusion and the attention gate mechanism significantly improve segmentation performance.ConclusionThe proposed multi-scale attention gate-controlled U-Net outperforms existing methods in brain tumor segmentation and shows strong potential for clinical application. Future work will focus on further optimizing the network structure and computational efficiency to enable real-time deployment in clinical settings.

Objective To address the issues of low space utilization, redundant operation lines, insufficient sterilization efficiency, and high risk of cross contamination in the traditional equipment layout of the Disinfection Supply Center (CSSD), a data-driven optimization method is constructed to systematically improve the operational efficiency and infection control level of CSSD. Methods Based on multi-source heterogeneous sensor networks and NeRF 3D reconstruction technology, a high fidelity and real-time mapping CSSD digital twin of physical entity operation status is constructed. The real-time and reliability of data are guaranteed through spatiotemporal calibration and adaptive weighted fusion algorithms. Results The multi center validation results showed that the scheme has good generalization ability, with an average spatial gain of 15.85% for general hospitals and 13.7% for specialized hospitals, an overall efficiency improvement of about 26.4%, and a cross contamination risk reduction of about 61.3%. Conclusion Research has confirmed that the optimization method driven by the synergy of digital twins and intelligent algorithms can effectively solve the complex optimization problem of CSSD device layout, significantly improve space utilization, operational efficiency, and infection control level, and provide a scalable dynamic decision support system and key technical means for the fine and intelligent management of medical institutions.

ObjectiveMulti-dimensional analysis and evaluation are conducted on the using data of high-value medical consumables in the whole hospital based on DRG diseases, providing data support for the establishment of the use specifications of high-value medical consumables. MethodsThrough multi-system linkage, the entire process data of high-value consumables has been integrated. Statistical analysis was conducted on the usage data of high-value medical consumables at various levels and dimensions, such as hospital area, department, doctor group, the surgical procedure, the consumable variety, etc. Based on the average cost and quantity of consumables per patient under the same disease type and surgical procedure, establish relevant usage norms and early warning mechanisms for high-value consumables, and carry out internal control optimization and adjustments for the using abnormal consumables. Results Take thoracoscopic lobectomy for a malignant tumor in the right upper lobe of the lung as an example, the overuse rates of key consumables have all decreased, and the difference is statistically significant (P < 0.05). The early warning control has gradually shown its effectiveness. ConclusionBased on the multi-dimensional usage data analysis of high-value consumables in DRG disease types, it provides data references for evaluating the rationality of consumable usage, promotes the standardized use of consumables in clinical practice, and can provide decision-making basis and effective management tools for hospital DRG cost control.

With the acceleration of the global population ageing process, the disease burden of chronic Non-communicable disease is increasing, and the traditional chronic disease management model is facing unprecedented challenges. The rapid development of large language models based on artificial intelligence technology has brought great opportunities for the intelligent transformation and service of chronic disease management. This paper systematically discusses the overview of the large language model, focuses on its application status in the management of major chronic diseases such as diabetes and cancer, and deeply analyzes the key issues such as technology and safety, the aim is to provide a theoretical basis for the construction of a new intelligent and precise chronic disease management system.

ObjectiveIn order to study the targeted maintenance scheme of medical ultrasound, the fault performance and high incidence fault parts of medical ultrasound are counted. Methods The statistical analysis of ultrasound maintenance data from 2021 to 2024 was conducted, with a total of 307 cases. The ultrasound fault manifestations and fault components were classified, and the percentage of components corresponding to different fault manifestations was counted. The high-frequency fault components of different equipment types were compared analyzed. Results The higher frequency of medical ultrasound faults is abnormal status, probe abnormality, panel abnormality, startup abnormality and display abnormality accounting for about 23.45%, 21.83%, 15.96%, 15.64% and 14.33% in turn. The higher frequency of fault components probe, control panel, software, power components and hardware circuit, accounting for 30.29% and 20.85%, 17.26%, 12.05% and 8.14% in turn. The higher frequency of fault components of desktop ultrasound equipment is probe, control panel and software, accounting for 30.60%, 21.35and 17.79% in turn, and the higher frequency of fault components of portable ultrasound equipment is probe, power components and control panel, accounting for 26.92%, 19.23% and 15.38% in turn. The higher frequency of imported ultrasound equipment is probe, control panel and software, accounting for 28.41%, 21.21% and 18.18% in turn, while the higher frequency of domestic equipment is probe, control panel and software, accounting 41.86%, 18.60% and 11.63% in turn. ConclusionThrough statistical analysis of maintenance data and combined with the practical maintenance experience of engineers, the causes of medical ultrasound failures can be divided into environmental factors, human factors equipment aging and natural wear and tear. The high incidence fault components of medical ultrasound are transducers, control panels, software, power components and hardware circuit boards, and there are in the high incidence fault components of desktop ultrasound and portable ultrasound equipment. When carrying out preventive maintenance, these components need to be focused on.

ObjectiveConstruct an evaluation index system for emergency medical quality from the perspective of public hospitals, with the aim of carrying out precise and objective quality control on emergency medical treatment and reducing medical safety risks. MethodsUse the Delphi method to screen the evaluation indicators for emergency medical quality, and adopt the analytic hierarchy process and factor analysis method to determine the coefficients and weights of the evaluation indicators. Results After two rounds of expert consultations, the final evaluation index system for emergency medical quality was determined. It mainly includes 4 first-level indicators, 14 second-level indicators, and 43 third-level indicators. The recovery rates of the first and second rounds of opinion consultations in the expert research were both 100%. The Kendall’s W coefficients for the importance scores of the first-level indicators, second-level indicators, and third-level indicators were 0.250, 0.239, and 0.161 respectively. The authority degrees of the 13 experts were all relatively high (the Cr values were all > 0.7), and the coordination of the experts' opinions was relatively good (the coefficient P < 0.05). Among the first-level indicators, the quality control indicators for specific diseases (0.4375) are higher than the process indicators (0.2708), which in turn are higher than the structural indicators and outcome indicators (both at 0.1458). Among these, the quality control indicators for specific diseases focus more on critical maternal/neonatal centers and trauma centers, while the process indicators focus on the standardization of diagnosis and treatment. ConclusionThe constructed emergency medical quality system is scientific and reliable. It can be adjusted according to the actual situations of various medical institutions to conduct a comprehensive evaluation of emergency medical quality.

With the surge in global End-Stage Renal Disease (ESRD) patient populations and an increasing proportion of patients with suboptimal vascular conditions, Arteriovenous Grafts (AVG) have become a critical alternative solution for vascular access, highlighting their growing clinical significance. Given significant disparities in risk classification and evidence requirements among global regulatory agencies for such devices, this study focuses on the U.S. Food and Drug Administration (FDA) regulatory framework and practices for AVGs. We conduct an in-depth analysis of the FDA's regulatory pathways, compare Sino-U.S. regulatory models, and derive actionable insights for optimizing China's medical device registration and review processes. Furthermore, we propose strategic recommendations for formulating specialized guidelines and advancing regulatory science for emerging technologies, aiming to theoretically and practically support the enhancement of China's regulatory strategies and the high-quality development of innovative medical devices.

Deep Brain Stimulation（DBS） provides a breakthrough therapeutic approach for refractory neurological disorders such as Parkinson's disease, dystonia, and epilepsy by precisely modulating targeted neural circuits, significantly improving patients' quality of life. With advancements in brain science and engineering technology, the indications for DBS continue to expand, and its technical architecture undergoes iterative upgrades. This article systematically reviews the evolution of DBS from its early technological prototypes to current intelligent systems featuring directional electrodes and adaptive closed-loop stimulation. It focuses on comparing the regulatory frameworks, clinical guidelines, and review requirements across major regions including the European Union, the United States, and China, analyzes the technical characteristics and clinical application advantages of domestically produced versus imported DBS products, and addresses existing regulatory challenges. Finally, it proposes a scientific regulatory and policy pathway tailored to China's context, offering theoretical references and practical strategies to promote the standardized and international development of the DBS field in China.

This study systematically explores the classification and supervision of medical robots and their core components. Currently, robot technology has been widely applied in medical related fields such as surgery, logistics, rehabilitation, diagnosis, and teaching research. Regulatory authorities urgently need to conduct research on product classification and definition. Research and propose core judgment principles for defining the regulatory attributes of medical devices; In the classification of management categories, emphasis is placed on implementing classified supervision based on product risk levels. For core components, the study proposes that the division of registration units should consider factors such as functional independence, structural integrity, and usage forms; The determination of its management category is based on risk correlation and other factors. Research suggests that it is urgent to adjust the classification catalog and develop specialized technical guidance principles to clarify the classification and definition of medical robots and core components, in order to assist in the scientific supervision of such products.