With the widespread application of robotic technology in surgery, multiple robot-assisted laparoscopic endoscopic surgical systems are currently in the R&D phase, highlighting an urgent need for establishing a scientific and standardized quality evaluation framework. In 2024, the National Medical Products Administration issued the standard YY/T 1941-2024, entitled Robotically-assisted Laparoscopic Surgical System, which specifies performance requirements and test methods for such systems. This standard provides critical guidance for the development of innovative products, technological advancement, and regulatory oversight. The standard offers clear definitions of relevant terminology and systematically standardizes the quality evaluation methods for these products across six key aspects: manual controller performance, instrument arm performance, system operational performance, safety protection functions, robotic surgical instruments, and software. This article interprets the core indicators outlined in the standard, aiming to assist clinicians in gaining a deeper understanding of the performance parameters of these systems and to support manufacturers in conducting quality assessments and risk analyses during product development, thereby promoting high-quality progress in the industry and related products.

With the deepening application of artificial intelligence (AI) technology in the field of medical devices, pre-trained models have increasingly become a crucial engine driving innovation in intelligent healthcare due to their efficiency, generalization capability, and transfer learning performance. However, potential risks associated with pre-trained models—such as issues related to source diversity and quality controllability —pose new challenges to the safety and effectiveness of AI-based medical devices. Against this background, the National Medical Products Administration (NMPA) released the sectoral standard YY/T 1833.5-2024 Artificial Intelligence Medical Devices—Quality Requirements and Evaluation-Part 5: Pre-trained Models in September 2024. This standard provides essential technical guidance and a regulatory framework for standardizing the application of pre-trained models in medical devices, marking a milestone in ensuring the safety and efficacy of AI-powered medical products. This article offers an in-depth interpretation and analysis of the standard's background, orientation, and key technical consideration. It elucidates specific requirements regarding documentation for pre-trained models, definitions of critical quality attributes, and conformity assessment pathways. Furthermore, the practical and far-reaching implications of the standard are discussed, including its role in enhancing quality assurance throughout the entire lifecycle of AI-based medical devices and guiding technological innovation and healthy development within the industry. Additionally, by dissecting the standard, this article aims to support the industry in conducting prudent evaluations during model selection phases, thereby reducing the application of low-quality and high-risk models.

ObjectiveTo explore the effect of mentality-controlling on local microcirculation of acupuncture at Hegu （LI 4）. MethodsForty healthy subjects were randomly divided into simple acupuncture group and mentality-controlling acupuncture group， with 20 cases in each group. The simple acupuncture group was acupunctured at Hegu （LI 4） on the right hand of the patient. In mentality-controlling acupuncture group， an audio clip was played to guide the participants to control their mentality when needle retention for 5 min after performing acupuncture twirling mani-pulation. The microcirculatory blood perfusion units （MBPU） was captured and compared between groups of participants using laser speckle contrast imaging （LSCI） for 5 mins in each of the four stages， state before acupuncture （R1）， after needle insertion （A1）， after needle twirling （A2） and after needle withdrawal （R2）. The participants' feelings such as relaxation， concentration， drowsiness， energy flow at acupoints were compared between groups， and the adverse reactions were recorded. ResultsThe MBPU values of R1， A1， A2 and R2 stages in mentality-controlling acupuncture group were 67.92±18.40 PU， 168.13±46.03 PU， 144.19±44.40 PU， 159.73±65.21 PU， respectively. The simple acupuncture group were 79.85 （66.80， 99.40）PU， 193.31 （173.15， 224.80）PU， 147.68 （128.12， 198.30）PU， 158.09 （126.05， 207.59）PU respectively. There was a statistically significant difference in MBPU between groups of the four stages （P＜0.01）. Among them， the MBPU in the A1 stage of both groups was significantly larger than that in the R1 stage， the MBPU in the A2 stage significantly smaller than that in the A1 stage， and the MBPU in the R2 stage was significantly larger than that in the R1 stage （P＜0.01）. MBPU in the R2 stage of the mentality-controlling acupuncture group was larger than that in the A2 stage （P＜0.05）. The difference of MBPU between A2 and A1 in mentality-controlling acupuncture group was significantly smaller than that in the simple acupuncture group， and the difference was statistically significant （t = 3.79， P＜0.01）. In mentality-controlling acupuncture group， 18 participants felt relaxed （90%）， 15 concentrated （75%）， 3 felt sleepy （15%）， and 14 felt energy flow at acupoints （70%）； while in the simple acupuncture group， 13 participants felt relaxed and sleepy （65%）， 6 concentrated and 6 felt energy flow at acupoints （30%）. The number of participants with concentration and energy flow at acupoints in mentality-controlling acupuncture group was higher than that in the simple acupuncture group， and the number felt sleepy was lower than that in the simple acupuncture group （P＜0.05）. No needle fainting， needle site infection， local haematoma and any other adverse reactions were found in both groups. ConclusionAcupuncture could affect the changes of local microcirculation， and controlling mentality during acupuncture can slow the rate of MBPU reduction and enhance the post-acupuncture effect with a favourable safety profile.

Objective In this study,we compared the masticatory efficiency,occlusal functional indexes,and satisfaction of lingual-ized occlusion(LO)vs.anatomic occlusion(AO)and discussed the difference between these occlusions.Methods Twenty edentulous patients were included in this clinical trial during 2021 April to December.They were divided into 2 groups according to their denture occlusion.Masticatory efficiency was tested by chewing the same amount of peanuts(by weight)and then the absorption of the solution was compared under a spectrophotometer after wearing dentures for 1 month.The TeeTester occlusion analysis system was used to test masticatory function,and satisfaction research was conducted for the edentulous patients at this time point as well.Finally,statistical analyses were conducted to compare the difference in masticatory efficiency between AO and LO by SPSS 20.0.Results The indexes of LO were worse than AO in masticatory efficiency,masticatory function and most of satisfaction research,while LO was better than AO in speech recovery.The difference was only statistically significant in COF-y of masticatory function(P＜0.05).Conclusion LO is similar with AO in terms of masticatory efficiency,satisfaction and denture stability;LO has good performance in denture's fit and use for patients.

Objective This study aimed to evaluate the application of digital impression and resin model technology in removable partial dentures(RPD)for Kennedy classⅠandⅡdentition defects.Methods Patients with Kennedy classⅠorⅡdental defect were selected and grouped in accordance with the following denture production processes:digital impression/resin model/cast cobalt-chromium alloy framework group(group A),digital impression/resin model/laser printed titanium framework group(group B),alginate impression/plaster model/cast cobalt-chromium alloy framework group(group C),and alginate impression/plaster model/laser printed titanium framework group(group D),with 40 cases in each group.The final RPD was examined in place in the mouth,and the evaluation indicators included the retention force of clamp ring,the tightness of connector and base,and the accuracy of occlusion.The evaluation scores of each in-dex were used for analysis on the Kruskal-Wallis rank-sum test.Results No statistically significant difference in the score of each index was found among the four groups in RPD.Conclusion The cast cobalt-chromium alloy and laser-printed titanium framework RPD using digital impression and resin model can meet the clinical restoration requirements of patients with Kennedy classⅠandⅡdentition defects.

Objective:To analyze the clinical characteristics of elderly acute pulmonary thromboembolism(APE)patients complicated with preexisting atrial fibrillation(AF)and the impact of preexisting AF on in-hospital adverse outcomes in elderly patients with APE.Methods:A retrospective analysis was performed on elderly APE patients with preexisting AF hospitalized in Beijing Anzhen Hospital, Capital Medical University between January 1, 2008 and December 31, 2021.We compared the comorbidities, symptoms, signs, laboratory test results and echocardiographic features, simplified pulmonary embolism severity index(sPESI)scores and adverse in-hospital outcomes between the preexisting AF group and the non-AF group.Logistic regression was used to analyze the risk factors of in-hospital adverse outcomes in elderly patients with APE.Results:A total of 240 patients diagnosed with APE were enrolled.There were 120 patients in the AF group and 120 patients in the non-AF group.For patients in the AF group and the non-AF group, the proportions with chronic heart failure were 38.3%(46/120)and 15.8%(19/120), the proportions with lower extremity deep vein thrombosis(DVT)were 36.7%(44/120)and 65.8%(79/120), the left ventricular ejection fractions(LVEF)were(59±10)% and(62±7)%, and hospital stays were(15±7)and(11±4)days, respectively, and the differences were statistically significant( χ2=15.381, 20.429, t=2.527, -4.710, all P<0.05). The incidences of in-hospital adverse outcomes in the AF group and the non-AF group were 4.2%(5/120)and 3.3%(4/120), respectively, with no significant difference( χ2=0.000, P=1.000). The overall incidence of in-hospital adverse outcomes was 3.8%(9/240). Multivariate Logistic regression analysis showed that elevated lactic acid was an independent risk factor for in-hospital adverse outcomes( OR=2.753, 95% CI: 1.367-5.542, P=0.005). However, AF( OR=2.880, 95% CI: 0.587-14.141, P=0.192)and sPESI score( OR=2.056, 95% CI: 0.904-4.673, P=0.086)were not associated with in-hospital adverse outcomes. Conclusions:Elderly APE patients with preexisting AF have a relatively low incidence of DVT, but a higher proportion have concurrent chronic heart failure and need a longer hospital stay.Elevated lactic acid is an independent risk factor for in-hospital adverse outcomes of elderly APE patients with preexisting AF.However, preexisting AF has no predictive value for in-hospital adverse outcomes in elderly patients with APE.

Objective: To explore the effect of ultrasound on dentin smear layer's surface and bonding strength of the universal resin adhesive under self-etching mode. Methods: Forty mandibular third molars without caries were randomly divided into two groups; one was polished with silicon carbide sandpaper; the other was polished with silicon carbide sandpaper followed by ultrasonic treatment. Scanning electron microscope (SEM) was used to observe surface of the dentin. Treated teeth were bonded with two universal resin adhesives, Clearfil Universal Bond (pH=2.3) and All-Bond Universal (pH=3.1), and the penetration of the bonding interface was observed with a confocal laser scanning microscope (CLSM) after Rhodamine B staining. Finally, the micro tensile bond strength test was conducted to test the adhesion. Results: The SEM showed that after polishing with silicon carbide sandpaper, the smear layer of the dentin surface was scratched, and dentin tubules were almost completely blocked, with no obvious dentin tubules exposed. After ultrasonic treatment, the scratches were reduced, and a large number of dentin tubules were exposed. CLSM showed that both adhesives could penetrate the dentin along the dentin tubules more deeply after ultrasound treatment. Micro tensile bond strength tests showed that ultrasonic treatment could enhance the bonding strength of two universal resin adhesives. However, there was no statistical difference in bonding strength between these two universal resin adhesives under the same treatment. . Conclusion Ultrasound can partially remove the smear layer on dentin's surface, expose dentin tubules, and increase universal resin adhesives' penetration depth and bonding strength under self-etching mode

Objective:To investigate the distribution and related factors of lipoprotein(a) [Lp(a)] level in healthy Tajik and Kazak adults in China.Methods:A cross-sectional study was conducted from May to October 2021 and March to June 2022, and blood samples were collected from 2, 637 healthy Tajik adults [1 010 men, average age: (40.08±14.74) years; 1 627 women, average age: (38.27±12.90) years] in Tashkurgan Tajik Autonomous County and 1 911 healthy Kazak adults [720 men, average age: (42.10±12.26) years; 1 191 women, average age: (38.27±12.90) years] in Fuyun County of Xinjiang. Fasting blood glucose (FBG), creatinine (Cr), total cholesterol (TC), triglycerides (TG), high-density lipoprotein cholesterol (HDL-C), and Lp(a) were measured. The distribution of Lp(a) levels in different sex and age groups was compared. The Lp(a) level of Tajik and Kazakh ethnic groups was compared by Mann-Whitney U test, and correlation factors of Lp(a) level were analyzed by multiple logistic regression. Results:The distribution of Lp(a) level in Tajik and Kazak population was skewed. Lp(a) levels of Tajik and Kazak ranged from the lowest 0.40 mg/L and 0.62 mg/L to the highest 1 229.40 mg/L and 2 108.58 mg/L, respectively, and the median Lp(a) level was 78.90 (38.60, 190.20) mg/L and 103.30 (49.57, 234.27) mg/L, respectively. Median Lp(a) level of Kazak was significantly higher than that of Tjik ( P<0.001). The median Lp(a) levels of Tajik males and females were similar: 77.45 (39.80, 187.10) mg/L and 79.90 (38.60, 192.30) mg/L ( P=0.948). The median Lp(a) levels of Kazakh males and females were also similar: 109.42 (50.49, 230.38) mg/L and 99.01 (49.11, 237.25) mg/L, respectively ( P=0.578). After pooling the data of Tajik and Kazak and adjusting for age, sex, BMI, smoking, drinking, blood pressure, blood glucose and other factors, Lp(a) level was correlated with ethnic (standard partial regression coefficient 0.066, P=0.008) and LDL-C level (standard partial regression coefficient 0.136, P<0.001). Conclusions:Lp(a) level in healthy Tajik and Kazak adults varied greatly among individuals, and Kazak residents had a higher Lp(a) level than Tajik residents. There was no significant sex difference in Lp(a) level among Tajik and Kazakh, and LDL-C and ethnicity are independent factors related to Lp(a) level.

Objective     To evaluate the early clinical effect of reimplantation in the treatment of bicuspid aortic valve (BAV) with aortic root aneurysm. Methods     The clinical data of 25 patients with BAV and aortic root aneurysm [mean diameter: 45-63 (52.68±5.55) mm] undergoing reimplantation in West China Hospital from November 2019 to May 2021 were retrospectively reviewed. There were 22 males and 3 females. The mean age was 15-65 (50.00±13.10) years and body surface area was 1.79±0.23 m2. Results    The pathological classification of BAV malformation was confirmed during the operation: Type 0 in 3 patients and Type 1 in 22 patients. There were 12 patients undergoing cusp central plication, and 2 patients were sutured with a closed fusion crest. Postoperative valve leaflet coaptation height was 0.78±0.15 cm, and effective height was 1.27±0.19 cm. In operation, maximum aortic valve flow velocity was 1.65±0.42 m/s, pressure difference was 5.46±3.05 mm Hg, and aortic valve annulus diameter was 21.32±0.95 mm. Cardiopulmonary bypass time was 225.84±35.34 min, and aortic block time was 189.60±26.51 min. In-hospital time was 11.64±3.07 d, ICU stay time was 2.64±0.99 d, and mechanical ventilation time was 1.48±0.87 d. The follow-up time was 17.20±4.70 months, and no death or major complications occurred during the follow-up in all patients. The cardiac function of the patients significantly improved postoperatively (P≤0.05). Echocardiography suggested that 12 patients had no aortic regurgitation, 10 minor aortic regurgitation, 3 mild aortic regurgitation, and no patients with moderate or more severe regurgitation. The diameter of the aortic sinus, left ventricular end-diastolic diameter and volume decreased during the follow-up, compared to preoperative ones (P≤0.05). The maximum flow velocity of the aortic valve was 1.54±0.36 m/s, and the pressure difference was 5.17±2.38 mm Hg during the follow-up. Conclusion    Reimplantation technology has a good clinical effect for highly selective BAV patients. It can effectively avoid long-term postoperative anticoagulation, but the maximum flow rate after surgery is slightly increased, which may be related to the configuration of BAV itself. While compared with valve replacement, the effect is still worthy of recognition.

Artificial intelligence is a blooming branch of medical device. Its development and quality control all rely on high quality clinical data. Since there is no established standard or guidance yet, it is important to study how to build and utilize a dataset appropriately and scientifically, especially for the decrease of clinical trial expense. With reference to the current status of premarket review and related guidance in developed countries, this paper analyzes the role and requirement of datasets in the quality control of AI medical device, providing useful information for regulation agencies and the development of public datasets for AI.
Artificial Intelligence ; Data Analysis ; Equipment and Supplies ; Quality Control