ObjectiveThe essence of syndrome manifestation recognition in traditional Chinese medicine (TCM) is to infer the body’s latent pathogenesis state from clinical observational information, rather than to perform simple label matching. However, previous studies have largely modeled this task as syndrome pattern classification within a fixed label space, which does not adequately reflect the cognition process of TCM syndrome differentiation centered on pathogenesis reasoning, and is also insufficient to capture the openness, semantic variability, and cross-disease reusability of syndrome manifestation expression. This study aimed to investigate whether introducing pathogenesis reasoning chain-of-thought (PR-CoT) supervision into large language models (LLMs) could improve the quality and cognitive consistency of syndrome manifestation recognition and support cross-disease transfer. MethodsSyndrome manifestation recognition was formulated as a conditional generation task under the framework of clinical observational information (X)→pathogenesis structure (Z)→syndrome pattern output (Y), where Z serves as an explicit intermediate structural variable linking the clinical evidence and syndrome judgment. Within this framework, a PR-CoT-supervised dataset for syndrome manifestation recognition was constructed based on medical case records of spleen-stomach disorders. After preprocessing, information extraction, manual proofreading, and data cleaning, the dataset comprised 4 800 training cases, 400 development cases, and 400 test cases. Each sample was annotated with a structured PR-CoT consisting of three progressive levels: clinical information summarization, comprehensive pathogenesis analysis, and syndrome pattern output. Supervised fine-tuning was conducted on open-source LLMs, with an end-to-end model serving as the baseline. Qwen3-32B was used as the primary experimental model, and Qwen3-14B as the scale comparison model. A progressive multidimensional evaluation framework was further established, comprising a structural parsing level, a semantic similarity level, and an expert blind review level. At the structural parsing level, syndrome pattern expressions were decomposed into structural elements and evaluated using Precision, Recall, F1 score, and Jaccard similarity. At the semantic similarity level, independent LLMs scored the theoretical proximity between predicted and reference syndrome patterns. At the expert blind review level, three TCM experts independently evaluated model outputs on two dimensions: syndrome differentiation consistency and terminology standardization of syndrome patterns. In addition, zero-shot cross-disease transfer evaluation was conducted on gynecological and heart-system disorder test sets. ResultsAt the structural parsing level, PR-CoT supervision did not lead to a stable improvement in the element-wise overlap of syndrome pattern structural components. Compared with the corresponding baselines, neither Qwen3-32B nor Qwen3-14B showed consistent advantages in structural matching metrics after the introduction of PR-CoT supervision. In contrast, at the semantic similarity level, PR-CoT supervision produced stable positive gains across different model scales and evaluation systems. The average semantic score of Qwen3-32B increased from 6.425 8 in the baseline model to 6.585 0 after PR-CoT supervision, and that of Qwen3-14B increased from 5.870 0 to 5.964 2. At the expert blind review level, the overall score of Qwen3-32B (PR-CoT) was 7.026 0±0.107 7, higher than 6.416 3±0.288 9 for its baseline. In zero-shot cross-disease testing, the PR-CoT model still showed advantages in semantic evaluation and expert evaluation on both gynecological and heart-system disorder test sets, indicating a certain degree of transferability. ConclusionThe benefits of PR-CoT supervision are mainly reflected in TCM semantic consistency and clinical plausibility, rather than in improved hard matching of structural elements. These findings support understanding syndrome manifestation recognition as a process of generating and expressing latent pathogenesis structures, rather than as a classification task within a traditional fixed label space. By introducing pathogenesis reasoning as an explicit intermediate structure into the modeling process and combining it with a progressive multidimensional evaluation framework, this study provides a methodological pathway for intelligent TCM syndrome differentiation that integrates theoretical alignment, interpretability, and multi-level evaluation.

ObjectiveThe essence of syndrome manifestation recognition in traditional Chinese medicine (TCM) is to infer the body’s latent pathogenesis state from clinical observational information, rather than to perform simple label matching. However, previous studies have largely modeled this task as syndrome pattern classification within a fixed label space, which does not adequately reflect the cognition process of TCM syndrome differentiation centered on pathogenesis reasoning, and is also insufficient to capture the openness, semantic variability, and cross-disease reusability of syndrome manifestation expression. This study aimed to investigate whether introducing pathogenesis reasoning chain-of-thought (PR-CoT) supervision into large language models (LLMs) could improve the quality and cognitive consistency of syndrome manifestation recognition and support cross-disease transfer. MethodsSyndrome manifestation recognition was formulated as a conditional generation task under the framework of clinical observational information (X)→pathogenesis structure (Z)→syndrome pattern output (Y), where Z serves as an explicit intermediate structural variable linking the clinical evidence and syndrome judgment. Within this framework, a PR-CoT-supervised dataset for syndrome manifestation recognition was constructed based on medical case records of spleen-stomach disorders. After preprocessing, information extraction, manual proofreading, and data cleaning, the dataset comprised 4 800 training cases, 400 development cases, and 400 test cases. Each sample was annotated with a structured PR-CoT consisting of three progressive levels: clinical information summarization, comprehensive pathogenesis analysis, and syndrome pattern output. Supervised fine-tuning was conducted on open-source LLMs, with an end-to-end model serving as the baseline. Qwen3-32B was used as the primary experimental model, and Qwen3-14B as the scale comparison model. A progressive multidimensional evaluation framework was further established, comprising a structural parsing level, a semantic similarity level, and an expert blind review level. At the structural parsing level, syndrome pattern expressions were decomposed into structural elements and evaluated using Precision, Recall, F1 score, and Jaccard similarity. At the semantic similarity level, independent LLMs scored the theoretical proximity between predicted and reference syndrome patterns. At the expert blind review level, three TCM experts independently evaluated model outputs on two dimensions: syndrome differentiation consistency and terminology standardization of syndrome patterns. In addition, zero-shot cross-disease transfer evaluation was conducted on gynecological and heart-system disorder test sets. ResultsAt the structural parsing level, PR-CoT supervision did not lead to a stable improvement in the element-wise overlap of syndrome pattern structural components. Compared with the corresponding baselines, neither Qwen3-32B nor Qwen3-14B showed consistent advantages in structural matching metrics after the introduction of PR-CoT supervision. In contrast, at the semantic similarity level, PR-CoT supervision produced stable positive gains across different model scales and evaluation systems. The average semantic score of Qwen3-32B increased from 6.425 8 in the baseline model to 6.585 0 after PR-CoT supervision, and that of Qwen3-14B increased from 5.870 0 to 5.964 2. At the expert blind review level, the overall score of Qwen3-32B (PR-CoT) was 7.026 0±0.107 7, higher than 6.416 3±0.288 9 for its baseline. In zero-shot cross-disease testing, the PR-CoT model still showed advantages in semantic evaluation and expert evaluation on both gynecological and heart-system disorder test sets, indicating a certain degree of transferability. ConclusionThe benefits of PR-CoT supervision are mainly reflected in TCM semantic consistency and clinical plausibility, rather than in improved hard matching of structural elements. These findings support understanding syndrome manifestation recognition as a process of generating and expressing latent pathogenesis structures, rather than as a classification task within a traditional fixed label space. By introducing pathogenesis reasoning as an explicit intermediate structure into the modeling process and combining it with a progressive multidimensional evaluation framework, this study provides a methodological pathway for intelligent TCM syndrome differentiation that integrates theoretical alignment, interpretability, and multi-level evaluation.

Objective: To evaluate the comprehensive intervention effect of lunch for primary and secondary school students in Minhang District, so as to provide a theoretical and practical basis for lunch intervention in school. Methods: From October to December 2023, a convenience sampling method was used to select 1 937 students from one primary and secondary school in Minhang District.A comprehensive intervention measure focusing on "reducing oil and salt" for lunch recipe optimization and nutrition education was carried out, and a questionnaire survey was conducted to evaluate the intervention effect three months later. Chi square test and Wilcoxon rank test were used to compare the data before and after the intervention. Results: After intervention, the use of cooking oil and salt, the supply of protein and fat in primary and secondary school lunches were reduced, and had no obvious impact on energy and other major nutrients. After intervention, compared to before intervention, the proportion of primary school students who felt that lunch was greasy decreased (8.9%, 6.2%, χ 2=4.35), and the proportion of primary and secondary school students who felt that lunch were delicious decreased significantly (33.2%, 23.2%; 63.9%, 53.5%, χ 2=26.39, 17.52) ( P < 0.05 ). Secondary school students also felt reduced variety of food ingredients (46.9%, 38.3%, χ 2=16.05, P <0.05). In addition, after intervention, the total surplus rate of primary school students meals decreased (7.4%, 4.4%, χ 2=5.73), mainly reflected in the decrease of the surplus rate of staple foods (7.1%, 2.4%, χ 2=17.39), while the surplus rate of vegetable dishes increased ( 16.0 %, 21.2%, χ 2=6.01) ( P <0.05). Although there was no significant change in the total surplus rate of meals for secondary school students, the surplus rate of staple foods decreased (12.9%, 5.4%, χ 2=33.52), while the surplus rates of meat and vegetable dishes increased (11.2%, 26.9%; 17.5%, 33.2%, χ 2=74.26, 61.88) ( P <0.05). After intervention, there was no statistically significant difference in the overweight and obesity rates of primary school students ( χ 2=0.11,0.43) and secondary school students ( χ 2=0.01,0.00) compared to before intervention( P >0.05). After intervention, the lung capacity of primary school students [1 564 (1 269,1 890) mL] and sitting forward flexion [11.3 (7.6, 15.2) cm] increased compared to before intervention [1 522 (1 259, 1 819 ) mL, 10.5 (6.3, 13.5) cm] ( Z =2.20, 4.68, P <0.01), but there was no statistically significant difference in lung capacity and sitting forward flexion of secondary school students before and after intervention ( Z =-0.46, -0.08, P >0.05). Conclusion The comprehensive intervention of school lunch has promoted a significant decrease in the use of oil and salt in lunch and improved the quality of recipes, and has a positive impact on the situation of leftover lunch and the health of students to a certain extent.

Objective To investigate the value of multimodal model integrating digital mammography(MG)and MRI radiomics features for preoperative prediction of lymphovascular invasion(LVI)status in breast cancer.Methods The clinical and imaging data from 336 patients with pathologically confirmed invasive breast cancer were retrospectively analyzed and randomly divided into a training group(235 cases)and a test group(101 cases)according to the ratio of 7∶3.Feature dimensionality reduction was carried out by Pearson correlation analysis followed by least absolute shrinkage and selection operator(LASSO)regression.Radiomics models were constructed based on MG craniocaudal(CC),dynamic contrast enhancement(DCE),T2 WI,and integrated MRI sequences;a multimodal model was further developed by incorporating clinical high-risk factors.The predictive efficiency of each model was evaluated by plotting receiver operating characteristic(ROC)curve.Results The ROC curve analysis showed that the multimodal model performed the best predictive efficiency,with area under the curve(AUC)of 0.989 and 0.861,accuracy of 0.949 and 0.782,sensitivity of 0.923 and 0.828,and specificity of 0.962 and 0.764 in the training group and test group respectively.Conclusion The multimodal model,integrating MG and MRI radiomics features,show optimal performance and can be served as a preoperative prediction of LVI status in breast cancer.

The main function of hemoglobin is to transport oxygen and carbon dioxide,and the level of hemoglo-bin is closely related to the adverse outcomes of infected patients.Low hemoglobin level may indicate higher infec-tion incidence,disease severity and mortality.As a simple,rapid,and economical clinical indicator,hemoglobin may become a biomarker for predicting adverse outcomes in infected patients.This paper reviews the relationship between hemoglobin levels and adverse outcomes as well as the predictive value in infected patients,with the aim of helping clinicians(especially those in outpatient departments and primary hospitals)to understand the clinical signi-ficance of monitoring hemoglobin levels,so as to accurately identify high-risk infected patients,optimize treatment and referral decisions.

Objective To investigate the prognostic value of the platelet-to-lymphocyte ratio(PLR)at different early time points in adult patients undergoing veno-arterial extracorporeal membrane oxygenation(VA-ECMO).Methods A retrospective study was conducted,selecting 55 adult patients who underwent VA-ECMO treatment at the First Hospital of Jiaxing from June 2020 to October 2022 as the study subjects.Then,the patients'gender,age,past history[including hypertension,diabetes,heart disease,chronic obstructive pulmonary disease(COPD)],and the reason for extracorporeal membrane pulmonary oxygenation(ECMO)adjuvant therapy[including severe myocarditis,acute myocardia infarction,in-hospital and out-of-hospital cardiac arrest,severe closed craniocerebral injury,severe pneumonia,pelvic fracture,other(pulmonary embolism,electrocution,traumatic hepatic rupture,post-partum hemorrhage,severe acute pancreatitis,crush syndrome)],acute physiology and chronic health evaluationⅡ(APACHEⅡ),sequential organ failure assessment(SOFA)at the time of admission,and ECMO peripheral blood tests[creatinine,alanine aminotransferase(ALT),aspartate aminotransferase(AST),blood lactate acid(Lac),white blood cell count(WBC),neutrophil count(NEU),lymphocyte count(LYM),hemoglobin(Hb),and platelet count(PLT)]and the last time prior to ECMO assistance,24 hours prior to the occurrence of acute kidney injury(AKI),and 24 hours after the occurrence of AKI.PLR levels at 24 hours ECMO,and the proportion of continuous renal replacement therapy(CRRT).The patients were divided into a death group and a survival group based on their 30-day prognosis and further categorized into a CRRT group and a non-CRRT group based on whether CRRT was administered.Clinical indicators of patients with different prognosis and the differences in PLR levels between CRRT and non-CRRT groups were compared.Logistic regression analysis was used to identify independent risk factors affecting the 30-day prognosis of VA-ECMO patients.The receiver operator characteristic(ROC curves)were plotted to evaluate the prognostic predictive value of each risk factor.Results Compared to the survival group,the death group had significantly higher APACHEⅡscores,SOFA scores,LYM and proportion receiving CRRT[APACHEⅡscore:34.00(28.50,36.00)vs.25.00(14.75,34.00),SOFA score:5.00(4.00,6.50)vs.3.00(2.00,5.25),LYM(×109/L):3.40±1.97 vs.2.24±2.11,proportion receiving CRRT:91.30%(21/23)vs.62.50%(20/32)],and a significantly lower level of the last PLR prior to ECMO adjuvant[30.00(21.06,48.17)vs.58.82(41.80,145.72)],and the differences were statistically significant(all P<0.05).Logistic regression analysis showed that the levels of the last PLR before ECMO assistance[odds ratio(OR)=0.965,95%confidence interval(95%CI)was 0.938-0.993,P=0.013],APACHEⅡscore at the time of admission(OR=1.121,95%CI was 1.018-1.234,P=0.020),and CRRT(OR=7.734,95%CI was 1.042-57.401,P=0.045)were independent risk factors affecting the prognosis of the VA-ECMO patients at 30 days after adjuvant;the ROC curve analysis showed that APACHEⅡscore,CRRT and the last PLR level before ECMO assistance had a predictive value for the prognosis of VA-ECMO patients 30 days after assistance,in which the APACHEⅡscore+the last PLR level before ECMO assistance had the greatest predictive value in predicting the prognosis of the patients,with area under the curve(AUC)of 0.846,with a sensitivity of 62.5%and a specificity of 95.7%.Higher early PLR levels were associated with better prognosis.In the CRRT group,PLR levels at 24 hours before ECMO initiation,24 hours before AKI onset,and 24 hours after AKI onset were significantly lower than those in the non-CRRT group(all P<0.05).Conclusion Early PLR levels and CRRT administration have significant predictive value for the prognosis of patients undergoing VA-ECMO therapy.

Objective:To investigate the clinical effect of indocyanine green angiography (ICGA)-assisted design and harvest of expanded flaps for scar reconstruction.Methods:This study was a retrospective observational study. From April 2019 to August 2023, 19 patients with scars (8 males, 11 females; aged 3-38 years) treated at the Plastic Surgery Hospital of Peking Union Medical College and Chinese Academy of Medical Sciences met the inclusion criteria. The scars were distributed on the head, face, trunk, and extremities. In stage Ⅰ surgery, skin soft tissue expanders were implanted in suitable areas around the scars for skin soft tissue expansion. In stage Ⅱ surgery, the scar tissue was excised, resulting in wound areas ranging from 100 to 210 cm 2, and expanded flaps were designed. ICGA was used to identify target perforators and their accompanying veins, and the flap design was adjusted to ensure the inclusion of complete arterial and venous axes. The expanded flap with an area of 120 to 240 cm2 was harvested using unilateral back-cut technique and transferred to the recipient site, and the donor site wound was sutured directly. The durations of the arterial and venous phases of ICGA during flap design were recorded. The length-to-width ratios of the back-cut flaps were calculated for different regions. After stage Ⅱ surgery, the blood perfusion and survival of the flap, the wound healing at the donor site, and the occurrence of complications were observed. During follow-up, the appearance, color, and texture of the patient's flap were observed. Results:The arterial phase of ICGA lasted 10-27 (18±5) s, and the venous phase lasted 78-116 (100±10) s. The length-to-width ratios of the back-cut flaps were 1.22±0.32, 1.63±0.12, and 1.15±0.21 for the head and neck, trunk, and limb regions, respectively. After stage Ⅱ surgery, one patient had a large area of insufficient blood perfusion in the flap. By comparing ICGA images before and after flap transfer, the sutures at the oral commissure were loosened, the blood flow of the flap was restored. The blood perfusion of the flaps in other patients was good. All flaps survived completely, with well-healed donor site wounds and no complications. During 0.5-14.0 months of follow-up, all flaps of patients demonstrated excellent appearance, with color and texture matching the surrounding skin.Conclusions:As a means of superficial blood flow visualization, ICGA can not only clearly show the microvascular distribution of the expanded flap before operation, assist in optimizing the design of the flap, but also evaluate the blood perfusion of the flap after operation, reduce the occurrence of complications, and provide a full-process navigation for the harvesting of expanded flaps, thereby improving the safety of flap transfer for scar reconstruction.

Objective:To compare the efficacy of body posture combined with pharmacotherapy and pharmacotherapy alone in the treatment of chronic subdural hematoma(CSDH).Methods:Firstly, retrospective case series study was conducted. Thirty cases of CSDH that had received body posture combined with pharmacotherapy at Department of Neurosurgery, Huashan Hospital Affiliated to Fudan University from December 2016 to October 2020 were studied retrospectively. Twenty-seven patients were male, and 3 patients were female. The age of patients ( M(IQR)) was 66(16) years (range:28 to 84). Nineteen patients had unilateral hematoma, and 11 patients had bilateral hematoma. All patients received pharmacotherapy and body posture therapy that was to raise their lower limbs 20 to 30 cm with leg lift pad and get abdominal compressed with customized abdominal belt in supine position. Patients were required to maintain the body posture as much as possible, with the maximum to 16 to 18 hours per day. Patients with unilateral hematoma should tilt the head to the affected side and avoid tilting it to the opposite side. For patients with bilateral hematoma, there was no need for head lateralization. Patient were treated with oral dexamethasone and atorvastatin simultaneously. The preliminary efficacy of body posture combined with pharmacotherapy was determined by hematoma improvement rate which was analyzed by Clopper-Pearson method. Then, the multi-center, prospective, randomized controlled trial had carried out in 9 medical centers from August 2020 to November 2021. The stratified block randomization method was adopted. Patients were randomized in a ratio of 1∶1 to either receive pharmacotherapy alone(the control group) or body posture combined with pharmacotherapy(the experiment group) for 3 months and followed up for 6 months. Effective treatment was defined as complete absorption of hematoma, or the hematoma volume decreased by more than 10 ml and Markwalder grading scale score had improved by more than 1 point compared to the baseline. The efficacy rate and surgery conversion rate at 3 months and recurrence at 6 months were observed. Comparison between groups was performed with paired sample t test, Mann-Whitney U test, χ2 test, corrected χ2 test, or Fisher exact probability method. Logistic regression was used to compare the effective rate and operation rate between the two groups. Results:In the respective study, 30 patients completed follow-up 13 to 353 days after treatment. At the last follow-up, the incidence of almost complete absorption or significantly absorption of hematoma (hematoma volume was significantly reduced accompanied by symptom improvement) was 93.3%. The 95% CI for the incidence that analyzed by the Clopper-Pearson method was 77.9% to 99.2%. One hundred and six patients were enrolled in the multicenter study. Fifty-five patients underwent body posture combined with pharmacotherapy. The age was 74(17) years (range:26 to 92). Thirty-nine patients were males and 16 were females. Fifty-one patients underwent pharmacotherapy alone. The age was 69(12) years (range:48 to 84). Thirty-seven patients were males and 14 were females. The length of body posture recorded in diary card was (15.7±2.3) hours(range:7.6 to 19.3 hours). The efficacy rate in the body posture combined with pharmacotherapy group and pharmacotherapy alone group were 83.6% (46/55) and 56.9% (29/51), respectively at 3 months. The result of the logistic regression analysis showed that the efficacy of body posture combined with pharmacotherapy group was better than that of pharmacotherapy alone group ( OR=3.88,95% CI:1.57 to 9.58, P=0.003). Surgery rate in the body posture combined with pharmacotherapy group and pharmacotherapy alone group were 5.5% (3/55) and 21.6% (11/51) respectively. The result of Logistic regression showed that the pharmacotherapy alone group was more likely to be converted to surgery ( OR=0.21,95% CI:0.05 to 0.80, P=0.023). At the 6 months, no recurrence of cases was found in the body posture combined with pharmacotherapy group. However, the recurrence rate of pharmacotherapy alone group was 6.3% (3/48), there was no significant difference between the two groups ( P>0.05). Conclusion:The effect of body posture combined with pharmacotherapy for chronic subdural hematoma is better than that of pharmacotherapy alone.

Objective To investigate the prognostic value of the platelet-to-lymphocyte ratio(PLR)at different early time points in adult patients undergoing veno-arterial extracorporeal membrane oxygenation(VA-ECMO).Methods A retrospective study was conducted,selecting 55 adult patients who underwent VA-ECMO treatment at the First Hospital of Jiaxing from June 2020 to October 2022 as the study subjects.Then,the patients'gender,age,past history[including hypertension,diabetes,heart disease,chronic obstructive pulmonary disease(COPD)],and the reason for extracorporeal membrane pulmonary oxygenation(ECMO)adjuvant therapy[including severe myocarditis,acute myocardia infarction,in-hospital and out-of-hospital cardiac arrest,severe closed craniocerebral injury,severe pneumonia,pelvic fracture,other(pulmonary embolism,electrocution,traumatic hepatic rupture,post-partum hemorrhage,severe acute pancreatitis,crush syndrome)],acute physiology and chronic health evaluationⅡ(APACHEⅡ),sequential organ failure assessment(SOFA)at the time of admission,and ECMO peripheral blood tests[creatinine,alanine aminotransferase(ALT),aspartate aminotransferase(AST),blood lactate acid(Lac),white blood cell count(WBC),neutrophil count(NEU),lymphocyte count(LYM),hemoglobin(Hb),and platelet count(PLT)]and the last time prior to ECMO assistance,24 hours prior to the occurrence of acute kidney injury(AKI),and 24 hours after the occurrence of AKI.PLR levels at 24 hours ECMO,and the proportion of continuous renal replacement therapy(CRRT).The patients were divided into a death group and a survival group based on their 30-day prognosis and further categorized into a CRRT group and a non-CRRT group based on whether CRRT was administered.Clinical indicators of patients with different prognosis and the differences in PLR levels between CRRT and non-CRRT groups were compared.Logistic regression analysis was used to identify independent risk factors affecting the 30-day prognosis of VA-ECMO patients.The receiver operator characteristic(ROC curves)were plotted to evaluate the prognostic predictive value of each risk factor.Results Compared to the survival group,the death group had significantly higher APACHEⅡscores,SOFA scores,LYM and proportion receiving CRRT[APACHEⅡscore:34.00(28.50,36.00)vs.25.00(14.75,34.00),SOFA score:5.00(4.00,6.50)vs.3.00(2.00,5.25),LYM(×109/L):3.40±1.97 vs.2.24±2.11,proportion receiving CRRT:91.30%(21/23)vs.62.50%(20/32)],and a significantly lower level of the last PLR prior to ECMO adjuvant[30.00(21.06,48.17)vs.58.82(41.80,145.72)],and the differences were statistically significant(all P<0.05).Logistic regression analysis showed that the levels of the last PLR before ECMO assistance[odds ratio(OR)=0.965,95%confidence interval(95%CI)was 0.938-0.993,P=0.013],APACHEⅡscore at the time of admission(OR=1.121,95%CI was 1.018-1.234,P=0.020),and CRRT(OR=7.734,95%CI was 1.042-57.401,P=0.045)were independent risk factors affecting the prognosis of the VA-ECMO patients at 30 days after adjuvant;the ROC curve analysis showed that APACHEⅡscore,CRRT and the last PLR level before ECMO assistance had a predictive value for the prognosis of VA-ECMO patients 30 days after assistance,in which the APACHEⅡscore+the last PLR level before ECMO assistance had the greatest predictive value in predicting the prognosis of the patients,with area under the curve(AUC)of 0.846,with a sensitivity of 62.5%and a specificity of 95.7%.Higher early PLR levels were associated with better prognosis.In the CRRT group,PLR levels at 24 hours before ECMO initiation,24 hours before AKI onset,and 24 hours after AKI onset were significantly lower than those in the non-CRRT group(all P<0.05).Conclusion Early PLR levels and CRRT administration have significant predictive value for the prognosis of patients undergoing VA-ECMO therapy.

Objective To investigate the value of multimodal model integrating digital mammography(MG)and MRI radiomics features for preoperative prediction of lymphovascular invasion(LVI)status in breast cancer.Methods The clinical and imaging data from 336 patients with pathologically confirmed invasive breast cancer were retrospectively analyzed and randomly divided into a training group(235 cases)and a test group(101 cases)according to the ratio of 7∶3.Feature dimensionality reduction was carried out by Pearson correlation analysis followed by least absolute shrinkage and selection operator(LASSO)regression.Radiomics models were constructed based on MG craniocaudal(CC),dynamic contrast enhancement(DCE),T2 WI,and integrated MRI sequences;a multimodal model was further developed by incorporating clinical high-risk factors.The predictive efficiency of each model was evaluated by plotting receiver operating characteristic(ROC)curve.Results The ROC curve analysis showed that the multimodal model performed the best predictive efficiency,with area under the curve(AUC)of 0.989 and 0.861,accuracy of 0.949 and 0.782,sensitivity of 0.923 and 0.828,and specificity of 0.962 and 0.764 in the training group and test group respectively.Conclusion The multimodal model,integrating MG and MRI radiomics features,show optimal performance and can be served as a preoperative prediction of LVI status in breast cancer.