Objective:The research group had previously developed an artificial intelligence algorithm to predict sepsis within 24 hours at the triage stage in emergency departments. This research studied the doctors’ response to algorithm-generated risk alerts and designs appropriate physician-algorithm collaboration strategies to further enhance sepsis risk identification capabilities.Methods:The research collected 40 cases of sepsis in the emergency departments from the open medical database MIMIC-IV (Medical Information Mart for Intensive Care) for a collaboration test. The cases were selected according to their typicality and classified according to the model’s confidence in its prediction. A total of 165 emergency doctors from 58 hospitals in China, stratified by professional rank, participated in the study. Four collaboration modes were designed with different information volumes and reading costs using the information offered by the algorithm. During the test, before and after the model presented its results and interpretive information according to the collaboration mode, the doctors were asked to rate the sepsis risk for each sample and record their confidence.Results:Analysis of the 4 704 valid evaluations done by 147 doctors showed that different collaboration modes caused no significant difference on doctors’ detection of sepsis risk. For cases with high model confidence, physicians’ diagnostic accuracy improved by 2.6%±0.6% ( P=0.02) post-algorithm input, with increased confidence in correct judgments. Conversely, for low-confidence model predictions, diagnostic accuracy decreased by 2.6%±1.4% ( P=0.06), accompanied by reduced clinician confidence in accurate assessments. Conclusions:The collaboration effect is mostly determined by the model’s confidence in its prediction. Different collaboration modes cause no significant difference, and doctors of different titles are influenced consistently with the same model confidence. Suggestions for collaboration design are as follow. When the model has low confidence in its own assessment of a patient’s sepsis risk, it should not directly demonstrate its assessment. When the model has high confidence, its assessment can be offered to the doctors as a reference. When predicting sepsis at the triage stage in the emergency departments, no extra interpretive information is needed.

Objective To address the limitations of conventional physics-informed neural network(PINN)in handling hemodynamic boundary constraints,an improved hard boundary-constrained PINN(HBC-PINN)framework was proposed to achieve precise prediction of blood flow fields within stenotic arteries.Methods An idealized stenosed vessel geometry model was established and computational fluid dynamic simulation was performed to obtain a validation dataset.Appropriate boundary dependent trial functions were designed according to the hard constraint method to embed the flow boundary conditions into the network output.Thus,an HBC-PINN model with the hard boundary constraint method was constructed to predict the velocity field and pressure field of stenosed blood flow.Meanwhile,an original PINN model with the soft constraint method was also built for comparison.By evaluating the accuracy of the two models on the validation dataset,the capability of the HBC-PINN model to simulate hemodynamics without using any labeled data for training was verified.Results The effectiveness of the HBC-PINN method in predicting hemodynamic parameters in stenosed blood flow tasks was validated.The relative L2 errors of the flow velocity and pressure predicted by the HBC-PINN in two different stenosis scenarios were both lower than 0.5％,representing an improvement of over 48.8％in accuracy compared to the original PINN model.Additionally,the prediction accuracy of the transverse velocity also increased by more than 35.4％.Conclusions Implementing hard constraints on boundary conditions in the PINN modeling process can effectively improve the prediction accuracy of hemodynamic parameters and the efficiency of model solving.

Objective:To understand the potential characteristics of decision-making preparation in outpatient hypertensive patients based on latent profile analysis, to identify the influencing factors of different categories of decision-making preparation levels, and to explore the performance of different decision-making preparation types in facilitation of patients involvement in treatment decision-making.Methods:Through a cross-sectional study, 350 hypertensive patients attending outpatient clinics in five different types of healthcare institutions (general hospitals, specialised hospitals and community hospitals) in Tianjin during January to May 2024 who met the inclusion and exclusion criteria were selected by the convenience sampling method as study subjects. General Information Questionnaires, Preparation for Decision Making Scale, and Facilitation of Patient Involvement Scale were used for investigation.Results:Totally 350 valid questionnaires [178 males and 172 females aged 25-89(57.24 ± 13.39)years old] were collected. The decision-making preparation score of outpatient hypertensive patients was (64.19 ± 18.69). The latent profile analysis results showed that the decision-making preparation of outpatient hypertensive patients could be divided into three potential categories: decision-making information scarcity type accounted for 20.0%(70/350), decision-making balance negotiation type accounted for 39.7%(139/350), and decision-making preparation adequacy type accounted for 40.3%(141/350). The results of multiple Logistic regression analysis showed that age, medical insurance type, occupation, and children′s condition were the influencing factors for the potential categories of decision-making preparation in outpatient hypertensive patients (all P<0.05). Age [less than 35 years old: OR(95% CI)=0.127(0.020-0.796)], occupation [on the job: OR(95% CI)=2.010 (1.034-3.906)], were the influencing factors of decision-making balance negotiation group (all P<0.05). Medical insurance type [basic medical insurance for urban employees: OR(95% CI)=0.372(0.193-0.720)], occupation [on the job: OR(95% CI)=2.500(1.270-4.920)], children′s condition[junior and senior high school: OR(95% CI)=0.391(0.190-0.802)] were the influencing factors of decision-making preparation adequacy group (all P<0.05). Conclusions:The level of promoting patient participation among outpatients with hypertension is relatively high, and there are differences in the perceived degree of promoting patient participation among patients with different types of decision preparation.It is recommended that medical staff provide decision-making related information based on the characteristics of different decision-making preparation categories of patients, encourage patients to actively participate in decision-making, and construct targeted decision support plans.

Objective To address the limitations of conventional physics-informed neural network(PINN)in handling hemodynamic boundary constraints,an improved hard boundary-constrained PINN(HBC-PINN)framework was proposed to achieve precise prediction of blood flow fields within stenotic arteries.Methods An idealized stenosed vessel geometry model was established and computational fluid dynamic simulation was performed to obtain a validation dataset.Appropriate boundary dependent trial functions were designed according to the hard constraint method to embed the flow boundary conditions into the network output.Thus,an HBC-PINN model with the hard boundary constraint method was constructed to predict the velocity field and pressure field of stenosed blood flow.Meanwhile,an original PINN model with the soft constraint method was also built for comparison.By evaluating the accuracy of the two models on the validation dataset,the capability of the HBC-PINN model to simulate hemodynamics without using any labeled data for training was verified.Results The effectiveness of the HBC-PINN method in predicting hemodynamic parameters in stenosed blood flow tasks was validated.The relative L2 errors of the flow velocity and pressure predicted by the HBC-PINN in two different stenosis scenarios were both lower than 0.5％,representing an improvement of over 48.8％in accuracy compared to the original PINN model.Additionally,the prediction accuracy of the transverse velocity also increased by more than 35.4％.Conclusions Implementing hard constraints on boundary conditions in the PINN modeling process can effectively improve the prediction accuracy of hemodynamic parameters and the efficiency of model solving.

Objective:To understand the potential characteristics of decision-making preparation in outpatient hypertensive patients based on latent profile analysis, to identify the influencing factors of different categories of decision-making preparation levels, and to explore the performance of different decision-making preparation types in facilitation of patients involvement in treatment decision-making.Methods:Through a cross-sectional study, 350 hypertensive patients attending outpatient clinics in five different types of healthcare institutions (general hospitals, specialised hospitals and community hospitals) in Tianjin during January to May 2024 who met the inclusion and exclusion criteria were selected by the convenience sampling method as study subjects. General Information Questionnaires, Preparation for Decision Making Scale, and Facilitation of Patient Involvement Scale were used for investigation.Results:Totally 350 valid questionnaires [178 males and 172 females aged 25-89(57.24 ± 13.39)years old] were collected. The decision-making preparation score of outpatient hypertensive patients was (64.19 ± 18.69). The latent profile analysis results showed that the decision-making preparation of outpatient hypertensive patients could be divided into three potential categories: decision-making information scarcity type accounted for 20.0%(70/350), decision-making balance negotiation type accounted for 39.7%(139/350), and decision-making preparation adequacy type accounted for 40.3%(141/350). The results of multiple Logistic regression analysis showed that age, medical insurance type, occupation, and children′s condition were the influencing factors for the potential categories of decision-making preparation in outpatient hypertensive patients (all P<0.05). Age [less than 35 years old: OR(95% CI)=0.127(0.020-0.796)], occupation [on the job: OR(95% CI)=2.010 (1.034-3.906)], were the influencing factors of decision-making balance negotiation group (all P<0.05). Medical insurance type [basic medical insurance for urban employees: OR(95% CI)=0.372(0.193-0.720)], occupation [on the job: OR(95% CI)=2.500(1.270-4.920)], children′s condition[junior and senior high school: OR(95% CI)=0.391(0.190-0.802)] were the influencing factors of decision-making preparation adequacy group (all P<0.05). Conclusions:The level of promoting patient participation among outpatients with hypertension is relatively high, and there are differences in the perceived degree of promoting patient participation among patients with different types of decision preparation.It is recommended that medical staff provide decision-making related information based on the characteristics of different decision-making preparation categories of patients, encourage patients to actively participate in decision-making, and construct targeted decision support plans.

BACKGROUND:Sepsis is one of the main causes of mortality in intensive care units(ICUs).Early prediction is critical for reducing injury.As approximately 36%of sepsis occur within 24 h after emergency department(ED)admission in Medical Information Mart for Intensive Care(MIMIC-IV),a prediction system for the ED triage stage would be helpful.Previous methods such as the quick Sequential Organ Failure Assessment(qSOFA)are more suitable for screening than for prediction in the ED,and we aimed to find a light-weight,convenient prediction method through machine learning. METHODS:We accessed the MIMIC-IV for sepsis patient data in the EDs.Our dataset comprised demographic information,vital signs,and synthetic features.Extreme Gradient Boosting(XGBoost)was used to predict the risk of developing sepsis within 24 h after ED admission.Additionally,SHapley Additive exPlanations(SHAP)was employed to provide a comprehensive interpretation of the model's results.Ten percent of the patients were randomly selected as the testing set,while the remaining patients were used for training with 10-fold cross-validation. RESULTS:For 10-fold cross-validation on 14,957 samples,we reached an accuracy of 84.1%±0.3%and an area under the receiver operating characteristic(ROC)curve of 0.92±0.02.The model achieved similar performance on the testing set of 1,662 patients.SHAP values showed that the five most important features were acuity,arrival transportation,age,shock index,and respiratory rate. CONCLUSION:Machine learning models such as XGBoost may be used for sepsis prediction using only a small amount of data conveniently collected in the ED triage stage.This may help reduce workload in the ED and warn medical workers against the risk of sepsis in advance.

Objective:To investigate the effects of rituximab on lymphocytes and immunoglobulin in the treatment of focal segmental glomerulosclerosis (FSGS) and minimal change disease (MCD).Methods:The subjects were FSGS and MCD patients admitted to Ruijin Hospital affiliated to Shanghai Jiaotong University on July 1, 2014 and July 1, 2019. All the enrolled patients were confirmed by clinical examination and renal biopsy, and received rituximab treatment (4 infusions of 375 mg/m 2 with the interval of 7-14 d). The levels of immunoglobulin IgA, IgG, IgM, and lymphocytes of CD19 +, CD20 +, CD3 +, CD3 +CD4 +, CD3 +CD8 + and natural killer cells (CD56 +CD16 +) were compared between baseline and the third month, the sixth month, the ninth month and the twelfth month after treatment. Results:Ninety-six patients with FSGS or MCD were enrolled in this study. The midian age was 28 years old (14-77 years old). The ratio of men to woman was 1.8∶1. There were 65 cases of MCD and 31 cases of FSGS. After rituximab treatment, the 24 h-proteinuria was significantly lower than that before treatment, and the serum albumin level was increased (both P<0.05). After rituximab treatment of 3 months, 6 months, 9 months and 12 months, CD19 + and CD20 + lymphocyte counts were significantly decreased (all P<0.01), and gradually recovered after 6 months. Compared with baseline, at 3, 6, 9, 12 months after rituximab treatment, the level of blood IgG was significantly increased ( P=0.004,<0.001,<0.001,<0.001, respectively), and the level of blood IgM was significantly decreased ( P<0.001, =0.008, =0.005,<0.001, respectively) but the median level still within the normal range (400-3 450 mg/L). The level of blood IgA was not significantly changed (all P<0.05). T lymphocytes (CD3 +, CD3 +CD4 + and CD3 +CD8 +) and natural killer cells (CD56 +CD16 +) showed no significant difference from baseline (all P>0.05). Conclusions:Rituximab can effectively eliminate CD19 + and CD20 + lymphocytes, and has little influence on peripheral blood lymphocyte count and immunoglobulin level except CD19 + and CD20 + lymphocytes. The standard administration of rituximab is safe for patients with FSGS and MCD.

Peritoneal metastasis is one of the most frequent metastatic patterns of advanced gastric cancer, but the mechanism underlying remains unclear. The 'seed and soil’ theory is now well recognized for peritoneal metastasis of gastric cancer at present. The combination of various cells, extracellular matrix, and ascites components within the abdominal cavity provide a suitable microenvironment for the plantation, infiltra-tion, growth and metastasis of gastric cancer cells. Fully under-standing of peritoneal microenvironment will help to diagnose the peritoneal metastasis of gastric cancer and tumor recurrence, and provide theoretical basis for the development of drugs targeting peritoneal microenvironment. The authors review the main cell formation, ascites and immune microenvironment involved in the formation of the peritoneal microenvironment based on relevant literatures at home and abroad, and investigate the relationship between peritoneal microenvironment and peritoneal metastasis of gastric cancer.

Objective:In combination with 3D printing technology and degradable composite materials, to discuss the preparation method of tissue engineering ossicles for middle ear hearing reconstruction.Methods:Domestic polymer (polylactic acid-glycolic acid copolymer, PLGA) and degradable ceramic material (β-tricalcium phosphate, β-TCP) were selected and prepared by low temperature deposition method according to the design ratio to Program according to the outline design code of the required scaffold to generate appropriate print files, and then the self-developed low-temperature deposition printing device was used to prepare tissue-engineered osseous scaffolds in accordance with the print files in a low-temperature environment. The scaffolds was freeze-dried and sterilized for later use after printing. Light microscopy and scanning electron microscopy were used to observe the apparent characteristics and internal structure of the scaffolds and to check its pore size, porosity and mechanical properties.Results:After printing, a degradable scaffold was obtained. Under the optical microscope, it was a small cylindrical shape with a diameter of 1.5 mm and a length of 6.0 mm, and its surface had micropores. The degradable scaffold had a horizontal and vertical interlaced warp and weft structure, the wire spacing was 1.2 mm, and the pores were connected to each other. The surface could see circular or quadrangular pores with a pore size of about 100-400 μm. The diameter of the inter-pore cross-linked channels was about 50 μm and the diameter of the surrounding circular micropores was about 10-40 μm. β-TCP particles with a size of about 700 nm were attached to the surface of the PLGA material. The average porosity of the whole scaffolds was (83.43±0.01)%, and the content of BMP-2 loaded was about 0.7 μg/mm 3. After freeze-drying, the mechanical strength of the scaffold was moderate, and there was no obvious deformation during stretching and compression, which met the mechanical requirements of tissue engineering ossicles. Conclusions:Using the low-temperature deposition printing method and strictly controlled processes and conditions, a polymer-degradable ceramic ossicle tissue engineering scaffold can be prepared for implantation experiments. The scaffold has suitable porosity and mechanical properties, and can be loaded with osteoinductive factors.

Objective: To understand the consistency of ALK Ventana-D5F3 immunohistochemistry (IHC) interpretation in Chinese lung adenocarcinoma among histopathologists from different hospitals, and to recommend solution for the problems found during the interpretation of ALK IHC in real world, with the aim of the precise selection of patients who can benefit from ALK targeted therapy. Methods: This was a multicenter and retrospective study. A total of 109 lung adenocarcinoma cases with ALK Ventana-D5F3 IHC staining were collected from 31 lung cancer centers in RATICAL research group from January to June in 2018. All cases were scanned into digital imaging with Ventana iSCANcoreo Digital Slide Scanning System and scored by 31 histopathologists from different centers according to ALK binary (positive or negative) interpretation based on its manufacturer′s protocol. The cases with high inconsistency rate were further analyzed using FISH/RT-PCR/NGS. Results: There were 49 ALK positive cases and 60 ALK negative cases, confirmed by re-evaluation by the specialist panel. Two cases (No. 2302 and No.2701) scored as positive by local hospitals were rescored as negative, and were confirmed to be negative by RT-PCR/FISH/NGS. The false interpretation rate of these two cases was 58.1% (18/31) and 48.4% (15/31), respectively. Six out of 31 (19.4%) pathologists got 100% accuracy. The minimum consistency between every two pathologists was 75.8%.At least one pathologist gave negative judgement (false negative) or positive judgement (false positive) in the 49 positive or 60 negative cases, accounted for 26.5% (13/49), 41.7% (25/60), respectively, with at least one uncertainty interpretation accounted for 31.2% (34/109). Conclusion There are certain heterogeneities and misclassifications in the real world interpretation of ALK-D5F3 IHC test, which need to be guided by the oncoming expert consensus based on the real world data.