With the development of artificial intelligence technology and the growing demand for minimally invasive surgery, the intelligentization of minimally invasive surgery has become a current research hotspot. Surgical instrument segmentation is a highly promising technology that can enhance the performance of minimally invasive endoscopic imaging systems, surgical video analysis systems, and other related systems. This article summarizes the semantic and instance segmentation methods of minimally invasive surgical instruments based on deep learning, deeply analyzes the supervision methods of training algorithms, network structure improvements, and attention mechanisms, and then discusses the methods based on the Segment Anything Model. Given that deep learning methods have extremely high requirements for data, current data augmentation methods have also been explored. Finally, a summary and outlook on instrument segmentation technology are provided.
Artificial Intelligence ; Minimally Invasive Surgical Procedures/instrumentation* ; Algorithms ; Deep Learning ; Humans ; Image Processing, Computer-Assisted

Objective:To explore the hearing changes of children with different genotypes of SLC26A4 with enlarged vestibular aqueduct（EVA） using the linear mixed effect model（LMM）, providing evidence for the risk prediction of progressive hearing loss. Methods:A total of 48 children with EVA diagnosed in our hospital from January 2017 to January 2024. All subjects underwent two or more auditory tests. According to the results of deafness gene screening and sequencing, the genotypes are divided into: type A: homozygous mutation of c. 919-2A>G, type B: compound heterozygous or heterozygous mutation containing c. 919-2A>G, and type C: no mutation site of c. 919-2A>G of SLC26A4 gene. LMM was used to analyze the hearing thresholds change of 500 Hz, 1 000 Hz, 2 000 Hz, 4 000 Hz and the average in children with different genotypes with age. Results:A total of 92 ears, 314 audiograms of 48 children were included, the median number of audiograms was 3, the median age of initial diagnosis was 4 months, and the median follow-up time was 13 months. According to LMM, the standard deviation of random effects between patients and ears was large. There was no significant difference in hearing thresholds of different frequencies and the average in genotype A, genotype B, and genotype C, indicating that genotype had no effect on hearing threshold. There is an interaction between age and genotype. Taking genotype C as the reference, children with genotype B had the lowest increase in 500 Hz, 1000 Hz, and the average hearing threshold, followed by type A. Conclusion:EVA children exhibit substantial inter-individual/ear hearing threshold variability. Low-frequency thresholds progress slower than high frequencies. Genotype modulates progression rates, with wild-type（Type C） demonstrating fastest deterioration, supporting personalized auditory monitoring strategies.
Humans ; Vestibular Aqueduct/abnormalities* ; Genotype ; Sulfate Transporters ; Mutation ; Auditory Threshold ; Hearing Loss, Sensorineural/genetics* ; Male ; Female ; Child ; Child, Preschool ; Hearing Loss/genetics* ; Hearing Tests ; Linear Models ; Infant

Objective To longitudinally analyze the changes in perioperative nutritional indicators in elderly patients with spontaneous intracerebral hemorrhage(ICH)and their correlation with the score of modified Rankin scale(mRS)score.Methods A total of 295 elderly ICH patients ad-mitted to the neurosurgery intensive care unit of North China University of Science and Technolo-gy Affiliated Hospital from June 2022 to November 2024 were consecutively enrolled in this study.The levels of relevant nutritional indicators were measured before(T1),and in 1(T2),3(T3),5(T4)and 7 d(T5)after surgery.According to the mRS score in 90 d after onset of the disease,they were divided into a good prognosis group(mRS score≤ 2,114 cases)and a poor prognosis group(mRS score≥3,181 cases).The general data and levels of nutritional indicators at above time points were compared between the two groups.Binary logistic regression analysis was applied to study the relationship between the perioperative nutritional indicator levels and the prognosis in elderly patients with spontaneous ICH.Results The poor prognosis group had significantly higher SBP,larger proportions of stroke history,hemorrhage in the brainstem and midline dis-placement,and increased NIHSS score than the good prognosis group(P＜0.05,P＜0.01).Obvi-ously lower BMI(56.48±9.54 kg vs 66.62±8.12 kg),decreased serum levels of albumin[40.24(39.10,41.16)g/L vs 43.30(40.65,45.50)g/L],prealbumin[203.00(164.00,261.00)mg/L vs 229.00(180.00,282.00)mg/L]and TC[4.68(4.11,5.39)mmol/L vs 4.91(4.47,5.66)mmol/L],reduced lymphocyte count[1.58(0.95,2.03)× 109/L vs 1.77(1.99,1.91)× 109/L],and reduced GCS[7.00(5.00,10.00)vs 9.00(7.00,13.00)]and Barthel[55.00(43.00,64.00)vs 61.00(52.00,69.00)]scores at admission were observed in the poor prognosis group than the good prognosis group(all P＜0.01).At all different time points,the poor prognosis group also got notably higher Controlling Nutritional Status(CONUT)scores(reaching the highest levels at 3 d after opera-tion),and lower Prognostic Nutritional Index(PNI)score and Geriatric Nutritional Risk Index(GNRI)score(both declining the lowest levels at same time)when compared with the poor prog-nosis group(all P＜0.01).Binary logistic regression analysis showed that COUNT scores,PNI scores and GNRI scores before(T1)and at 1(T2),3(T3),5(T4)and 7 d(T5)after operation were independent risk factors for 90-day poor prognosis in elderly patients with spontaneous ICH.Conclusion The fluctuations of 3 nutritional indicators at different time points in the periopera-tive period are closely associated with the 90-day prognosis of elderly ICH patients.Monitoring the changes in the nutritional indicators can quickly and conveniently predict the prognosis of the pa-tients,which has high clinical application value.

Objective To predict the value of admission dehydration state combined with fluid accumulation index for post-operative prognosis in elderly patients with intracerebral hemorrhage(ICH).Methods A retrospective study was conducted on 320 elderly ICH patients receiving surgical treatment in Department of Neurosurgery and then admitted to its Intensive Care Unit of the Affiliated Hospital of North China University of Science and Technology from May 2023 to March 2024.According to the survival status at 30 d after onset,they were divided into a survival group(202 cases)and a death group(118 cases).Clinical data such as basic information,admission vital signs,laboratory indicators,and fluid supplementation were compared between the two groups.ROC curve was plotted.Results The death group exhibited significantly advanced age,more bleeding sites,higher 7-day fluid accumulation index,and larger proportions of ventricular rupture,midline displacement and admission dehydration status,but lower uric acid level than the the survival group(P＜0.05,P＜0.01).Binary logistic regression analysis showed that admission dehydration status and 7-day fluid accumulation index were risk factors for 30-day mortality in elderly ICH patients after surgery(P＜0.01).ROC curve analysis showed that the AUC value of admission dehydration status combined with 7-day fluid accumulation index in predicting 30-day death of elderly ICH patients after surgery was 0.774(95％CI:0.722-0.825),and that of the combination was better than that of each indicator alone(P＜0.05).Conclusion Hospital dehydration status combined with 7-day fluid accumulation index has the best effectiveness in predicting 30-day mortality in elderly ICH patients after surgical treatment.

ObjectiveTo analyze the reported cases of infectious diseases across different tiers of public medical and healthcare institutions in Minhang District, Shanghai from 2013 to 2023, to investigate the status and changes in reported infectious diseases in this district from a temporal, etiological, and demographic perspectives, so as to provide a scientific basis for the construction of a hierarchica early-warning surveillance system under the grading diagnosis and treatment model in medical institutions, as well as for optimizing sentinel surveillance at facilities of different levels. MethodsA retrospective analysis was performed using surveillance data from the China Disease Prevention and Control Information System in Minhang District from 2013 to 2023. Reported infectious diseases were categorized into three categories based on transmission routes: respiratory infectious diseases, intestinal infectious diseases, and sexually transmitted and blood borne infectious diseases. According to the implementation phase of the grading diagnosis and treatment policy, the research time was divided into four time periods: 2013‒2016, 2017‒2019, 2020‒2022, and 2023. The distribution and temporal changes of reported cases of infectious diseases were compared across community health service centers (CHCs), secondary hospitals, tertiary grade-A hospitals and tertiary grade-B hospitals. Chi-square test was used for univariate analysis of differences in the number of reported cases. Quantitative data with normal distribution were analyzed using parametric tests, otherwise, Kruskal⁃Wallis H tests were used. ResultsThe proportions of total reported cases of infectious diseases in medical institutions at all levels in Minhang District, Shanghai from 2013 to 2023 were 10.66% in CHCs, 9.10% in secondary hospitals, 64.95% in tertiary grade-B hospitals, and 15.29% in tertiary grade-A hospitals, with an overall decline and then rebound trend in the reported cases. After the implementation of grading diagnosis and treatment policy, the number of reported cases in CHCs and secondary hospitals showed a trend of first decreasing and then increasing, while that in tertiary grade-B hospitals showed a steady decreasing trend and that in tertiary grade-A hospitals showed an increasing trend. In terms of the research periods divided above, a total of 10 392 cases were reported in 2013‒2016 (70.34% from tertiary grade-B hospitals and 12.59% from CHCs), including 2 922 cases of respiratory infectious diseases, 1 241 cases of intestinal infectious diseases, and 6 229 cases of sexually transmitted and blood-borne infectious diseases. Between 2017 and 2019, a total of 6 967 cases were reported (73.49% from tertiary grade-B hospitals and 11.84% from tertiary grade-A hospitals), including 2 983 cases of respiratory infectious diseases, 279 cases of intestinal infectious diseases, and 3 705 cases of sexually transmitted and blood-borne infectious diseases. Between 2020 and 2022, a total of 4 599 cases were reported (69.92% from tertiary grade-B hospitals and 24.57% from tertiary grade-A hospitals), including 1 627 cases of respiratory infectious diseases, 123 cases of intestinal infectious diseases, and 2 849 cases of sexually transmitted and blood-borne infectious diseases. In 2023, a total of 4 648 cases were reported (35.20% from tertiary grade-B hospitals and 27.50% from tertiary grade-A hospitals), including 3 165 cases of respiratory infectious diseases, 69 cases of intestinal infectious diseases, and 1 414 cases of sexually transmitted and blood-borne infectious diseases. The proportion of reported cases from tertiary grade-B hospitals was the highest in all the four research periods, but exhibited an obvious decrease in 2023. The differences in the reported cases of infectious diseases with different transmission routes among medical institutions at all levels were statistically significant (χ²=3 225.628, P<0.05). The differences in the mean age of patients among medical institutions at all levels were statistically significant (H=1 325.927, P<0.05). ConclusionThere are significant differences in the number of reported cases of infectious disease in the medical institutions at different levels. Tertiary grade-B hospitals have historically dominated the number of reported cases, but its share has declined recently. Whereas, CHCs and tertiary grade-A hospitals have played an increasingly important role in the surveillance and early warning of respiratory and intestinal infectious diseases. Therefore, it is recommended to leverage the strengths of grading diagnosis and treatment to establish targeted sentinel sites and deploy specialized teams tailored to the epidemiological characteristics of specific disease categories.

OBJECTIVE To investigate the relationship between 23-site chip genetic screening failures and the results of newborns hearing screening,and to provide clinical reference for the diagnosis and treatment of genetic screening failures.METHODS There were 1 916 newborns born in the Beijing area from November 2022 to May 2024,who did not pass the 23-site chip genetic screening tests and underwent newborn hearing screening with definite initial screening results.Chi-square test was used to analyze the relationship between different mutation types and genotypes and the initial hearing screening results.RESULTS The overall neonatal hearing screening failure rate was 5.27%(101/1 916),with a higher failure rate of 61.54%(56/91)for homozygous and compound heterozygous mutations than the failure rate of 2.54%(45/1 772)for heterozygous mutations,0%(0/34)for digenic gene heterozygous mutations,and 0(0/19)for mtDNA 12S rRNA mutations,with a statistically significant difference(P<0.001).Among the homozygous and compound heterozygous mutations,the failure rates of homozygous and compound heterozygous for GJB2 gene and SLC26A4 gene were 59.76%(49/82)and 77.78%(7/9),respectively,with no statistically significant difference between the two groups(P=0.488).The homozygous and compound heterozygous for GJB2 gene were divided into three groups based on genotype:c.109G>A homozygous mutations,c.109G>A compound heterozygous mutations,and other homozygous and compound heterozygous mutations.The hearing screening failure rates of the three groups,from highest to lowest,were as follow:other homozygous and compound heterozygous mutations(88.89%,8/9),c.109G>A homozygous mutations(65.12%,28/43),and c.109G>A compound heterozygous mutations(43.33%,13/30),with a statistically significant difference(P=0.029).The failure rates of heterozygous for GJB2 gene,SLC26A4 gene and GJB3 gene were 2.86%(40/1 398),1.25%(4/321)and 1.89%(1/53),respectively,with no statistically significant difference among the three groups(P=0.241).The failure rate of hearing screening for individuals with GJB2 heterozygotes of different genotypes and individuals with SLC26A4 heterozygotes of different genotypes did not show statistically significant differences.CONCLUSION The failure rate of newborn hearing screening for homozygous and compound heterozygous mutation of 23-site chip genetic screening is higher than that of other mutation types,verifying the effectiveness of the newborn hearing screening program.Some newborns of homozygous and compound heterozygous mutation can pass the hearing screening,especially those with the c.109G>A homozygous and compound heterozygous mutation,who need clinical follow-up.

OBJECTIVE To explore the early detection of large vestibular aqueduct syndrome(LVAS)in children by applying several audiological indicators.METHODS Ninety-two children with hearing loss(aged 1-70 months)were enrolled and divided into an LVAS group(45 cases)and a control group(47 cases).Eleven audiological indicators were statistically analyzed:lateral of hearing loss,the degree of hearing loss,configuration of hearing loss;ABR air-conduction threshold;ABR air-bone gap;ASSR average threshold;ASSR thresholds at 0.5,1,2,and 4 kHz;and tympanogram type.Indicators showing significant two-group differences were used to construct a visualized multifactorial linear prediction model using the R language.RESULTS Nine indicators demonstrated statistically significant differences between groups(P<0.05):laterality,configuration of hearing loss,ABR air-conduction threshold,ASSR average threshold,ASSR thresholds at all frequencies(0.5,1,2,4 kHz),and tympanogram type.A prediction model was established.When the total model score ranged between 200 and 240 points,the predicted LVAS risk probability was 0.1 to 0.99.The consistency index(C-index)was 0.85,indicating good predictive ability of the model.CONCLUSION The identified nine audiological indicators are valuable for the early detection of LVAS in children.The developed model can estimate LVAS risk.After refinement,this model holds potential to support early clinical diagnosis and intervention.

Objectives:To develop modified prehospital stroke scales and to evaluate their predictive value for in-hospital acute large vessel occlusion (LVO) stroke.Methods:Patients admitted to Dongtai People's Hospital due to non-stroke-related diseases and activated the in-hospital stroke green channel due to suspected stroke symptoms during hospitalization from January 2015 to December 2022 were included retrospectively. According to the final imaging diagnosis, they were divided into LVO group and non-LVO group. The five prehospital stroke scales included Field Assessment Stroke Triage for Emergency Destination (FAST-ED), Rapid Arterial Occlusion Evaluation (RACE), Los Angeles Motor Scale (LAMS), Cincinnati Prehospital Stroke Severity Scale (CPSSS), and Prehospital Acute Stroke Severity Scale (PASS). Multivariate logistic regression analysis was used to determine independent predictive factors of LVO in patients with in-hospital stroke, and incorporating them into the prehospital stroke scale to develop modified scales. The receiver operating characteristic (ROC) curve was used to evaluate the predictive performance of the modified scales. Results:A total of 174 patients with in-hospital stroke were enrolled, including 92 males (52.9%), aged 65.7±11.9 years. Fifty-four patients (31.0%) had LVO, and 59 (33.9%) had a surgical history within 3 days before the onset of stroke, mainly cardiopulmonary surgeries. Multivariate logistic regression analysis showed that atrial fibrillation (odds ratio 2.940, 95% confidence interval 1.387-6.230; P=0.005) and recent history of cardiopulmonary surgery (odds ratio 6.861, 95% confidence interval 2.437-11.315; P<0.001) were the independent predictive factors of LVO in patients with in-hospital stroke. According to the β coefficient and ROC curve, they were assigned a score of 1 and included in the prehospital stroke scale. The area under the curve of the modified scale for predicting LVO (mRACE: 0.917; mFAST-ED: 0.865; mPASS: 0.859; mCPSSS: 0.853; mLAMS: 0.907) was significantly higher than the corresponding original scale (RACE: 0.888; FAST-ED: 0.820; PASS: 0.786; CPSSS: 0.810; LAMS: 0.859) (all P<0.05). Conclusion:The modified scales based on the prehospital stroke scales can significantly improve the predictive value of in-hospital acute LVO stroke compared to the original prehospital stroke scales.

With the rapid development of population aging in various countries around the world,the health and elderly care industry has been paid high attention.The standardization of smart health and elderly care technology and services is particularly important.This paper firstly reviewed the policies related to healthy elderly care in China.By analyzing the industrial standards and provincial standards issued,this paper focused on the policies proposed by the Shanghai Municipal Government for the standardization of smart health and elderly care,as well as the researches on the standard system and the construction of standard families.Shanghai group standards in the field of smart health and elderly care were summarized,including the guidelines for the construction of standard systems,elderly care service platforms,community elderly cafeterias,portable health monitoring terminals,indoor sports services,and home-based elderly care safety monitoring.A series of case analyses of the standardized implementation of the above aspects were also provided.Through standardization research and practice in recent years,it has been fully demonstrated that the standard research plays an important leading role in the field of smart health and elderly care.

[Objective] To explore the optimal pressure parameters for chyle plasma filtration under low-temperature conditions, and to improve the quality of chyle plasma treatment and filtration efficiency by improving experimental methods. [Methods] The filtration efficiency and filtration time of 30 severe chyle plasma samples under conventional preparation environment pressure and under preparation environment with a controlled filtration membrane pressure difference of 0.5 bar were compared. [Results] The absorbance of severe chyle plasma samples before and after filtration under two different preparation pressures was statistically significant (P<0.05), and both achieved the expected filtration effect. Under the preparation environment of controlling the pressure difference of the filtration membrane to 0.5 bar, the filtration was faster and with better effect, which was statistically significant compared to the conventional preparation environment pressure (P<0.05). [Conclusion] By selecting the optimal pressure parameters for filtering chyle plasma under low-temperature conditions, the efficiency of chyle plasma filtration under low-temperature conditions has been improved, and the practicality and reliability of low-temperature filtration technology have been enhanced.