BACKGROUND:Rapid and accurate segmentation of brain tissue in medical images is of great significance for three-dimensional biomechanical modeling and diagnosis of craniocerebral injuries.Currently,artificial intelligence(AI)-based baseline models exhibit excellent generalization capabilities on large-scale datasets.However,due to the specificity and complexity of craniocerebral tissues,these models have certain limitations in their application to craniocerebral tissue segmentation.Additionally,the scarcity of craniocerebral tissue samples makes it difficult for baseline models to achieve precise segmentation results through fine-tuning.OBJECTIVE:To construct a craniocerebral tissue segmentation model based on texture feature retrieval enhancement to improve segmentation accuracy under a small number of samples.METHODS:Segment Anything in Medical Images(MedSAM)model was selected as the basic framework,and texture features were combined with deep learning to build a brain tissue segmentation model based on texture feature retrieval enhancement(DP-MedSAM).Dice Coefficient and mean intersection over union(MIoU)were selected to evaluate the efficiency of image segmentation results.In comparison with the original MedSAM model,the ablation experiment systematically evaluated the influence of key components on the model performance.The sensitivities of MedSAM,the Segment Anything Model(SAM)for medical image segmentation(SAM-Med2D)and DP-MedSAM in the mandible,left optic nerve,and left parotid gland were compared.RESULTS AND CONCLUSION:(1)By verifying the impact of the number of point prompts on segmentation results on the HaN-Seg dataset,the experimental results indicated that the optimal Dice score was achieved with the addition of three points.(2)DP-MedSAM demonstrated performance improvements compared with MedSAM and SAM-Med2D on two datasets(HaN and Public Domain Database for Computational Anatomy).Especially on the Public Domain Database for Computational Anatomy dataset,in terms of the MIoU metric,DP-MedSAM outperformed MedSAM by 6.59％and SAM-Med2D by 37.35％;in terms of the Dice metric,DP-MedSAM outperformed MedSAM and SAM-Med2D by 4.34％and 25.32％,respectively.(3)The ablation experiment results showed that removing the texture feature extraction module in the DP-MedSAM model,relying solely on original image features,led to a significant decrease in results on the test set.Furthermore,removing the vector cache database and its retrieval enhancement function from the model,which deprived the ability of the model to perform similarity retrieval using an external knowledge base,further reduced model performance.(4)Under conditions of limited data resources,the DP-MedSAM model outperformed the other two models in all evaluation metrics.The DP-MedSAM model performed excellently when processing simple and moderately difficult samples,demonstrating a clear advantage over the other two models and indicating good generalization ability.Processing the fine structures of difficult samples placed higher demands on the model's segmentation capabilities.Although the performance of the DP-MedSAM model declined slightly,it still outperformed the other two models.(5)This study proposes an innovative craniocerebral tissue segmentation model,DP-MedSAM,which improves the baseline model's performance in capturing local details and global structural information in medical images by introducing target region texture feature extraction.Through vector similarity retrieval technology,DP-MedSAM can retrieve the feature vector most similar to the current target region from a pre-constructed vector database,providing more precise guiding information for the segmentation process.

BACKGROUND:Rapid and accurate segmentation of brain tissue in medical images is of great significance for three-dimensional biomechanical modeling and diagnosis of craniocerebral injuries.Currently,artificial intelligence(AI)-based baseline models exhibit excellent generalization capabilities on large-scale datasets.However,due to the specificity and complexity of craniocerebral tissues,these models have certain limitations in their application to craniocerebral tissue segmentation.Additionally,the scarcity of craniocerebral tissue samples makes it difficult for baseline models to achieve precise segmentation results through fine-tuning.OBJECTIVE:To construct a craniocerebral tissue segmentation model based on texture feature retrieval enhancement to improve segmentation accuracy under a small number of samples.METHODS:Segment Anything in Medical Images(MedSAM)model was selected as the basic framework,and texture features were combined with deep learning to build a brain tissue segmentation model based on texture feature retrieval enhancement(DP-MedSAM).Dice Coefficient and mean intersection over union(MIoU)were selected to evaluate the efficiency of image segmentation results.In comparison with the original MedSAM model,the ablation experiment systematically evaluated the influence of key components on the model performance.The sensitivities of MedSAM,the Segment Anything Model(SAM)for medical image segmentation(SAM-Med2D)and DP-MedSAM in the mandible,left optic nerve,and left parotid gland were compared.RESULTS AND CONCLUSION:(1)By verifying the impact of the number of point prompts on segmentation results on the HaN-Seg dataset,the experimental results indicated that the optimal Dice score was achieved with the addition of three points.(2)DP-MedSAM demonstrated performance improvements compared with MedSAM and SAM-Med2D on two datasets(HaN and Public Domain Database for Computational Anatomy).Especially on the Public Domain Database for Computational Anatomy dataset,in terms of the MIoU metric,DP-MedSAM outperformed MedSAM by 6.59％and SAM-Med2D by 37.35％;in terms of the Dice metric,DP-MedSAM outperformed MedSAM and SAM-Med2D by 4.34％and 25.32％,respectively.(3)The ablation experiment results showed that removing the texture feature extraction module in the DP-MedSAM model,relying solely on original image features,led to a significant decrease in results on the test set.Furthermore,removing the vector cache database and its retrieval enhancement function from the model,which deprived the ability of the model to perform similarity retrieval using an external knowledge base,further reduced model performance.(4)Under conditions of limited data resources,the DP-MedSAM model outperformed the other two models in all evaluation metrics.The DP-MedSAM model performed excellently when processing simple and moderately difficult samples,demonstrating a clear advantage over the other two models and indicating good generalization ability.Processing the fine structures of difficult samples placed higher demands on the model's segmentation capabilities.Although the performance of the DP-MedSAM model declined slightly,it still outperformed the other two models.(5)This study proposes an innovative craniocerebral tissue segmentation model,DP-MedSAM,which improves the baseline model's performance in capturing local details and global structural information in medical images by introducing target region texture feature extraction.Through vector similarity retrieval technology,DP-MedSAM can retrieve the feature vector most similar to the current target region from a pre-constructed vector database,providing more precise guiding information for the segmentation process.

Objective To analyze the global pattern, long-term trends, and regional and population heterogeneity in the burden of 375 diseases and injuries and 88 risk factors from 1990 to 2023 using data from the Global Burden of Disease Study 2023 (GBD 2023). Methods Publicly available GBD 2023 estimates were extracted for incidence, deaths, years of life lost (YLLs), years lived with disability (YLDs), disability-adjusted life years (DALYs), and age-standardized rates at the global level and by socio-demographic index (SDI), GBD region, age group, and sex. Descriptive analysis, ranking comparison, stratified analysis, and trend analysis were performed, and 95% uncertainty interval was used to assess the stability of estimates. Results In 2023, non-communicable diseases were the leading contributor to global DALYs, accounting for 64.39% of DALYs across the three major disease and injury groups (communicable, maternal, neonatal, and nutritional diseases, non-communicable diseases, and injuries), and also had the highest age-standardized DALYs rate. From 1990 to 2023, age-standardized DALYs rates declined for three major disease and injury groups, with estimated annual percentage changes of −2.69%, −0.78%, and −1.47%, respectively. Cardiovascular diseases ranked first by both DALYs and deaths in 2023, whereas respiratory infections and tuberculosis ranked first by incidence. Non-communicable diseases were the leading cause of disease burden across all SDI strata. As the SDI level declined, the burden of communicable, maternal, neonatal, and nutritional diseases, as well as injuries, increased markedly. Behavioral and metabolic risks remained the main contributors to attributable deaths and DALYs. Conclusion From 1990 to 2023, the global burden of disease continues to shift toward non-communicable diseases; however, some regions still face a double burden of infectious, maternal and child, injury-related, and chronic diseases. Integrating deaths, DALYs, YLLs, YLDs, and risk-attributable burden helps identify priority areas for prevention and resource allocation across regions and populations.

Chronic atrophic gastritis(CAG)is a chronic gastric disorder characterized by recurrent damage to the gastric mucosal epithelium,resulting in the reduction of intrinsic glands,with or without concurrent intestinal metaplasia.The"hyperactive stomach qi"theory,derived from Huang Di Nei Jing Su Wen Ji Zhu,proposes that the core pathogenesis of CAG lies in excessive stomach qi activity,grounded in the physiological principle of"strong yang qi in earth and weak yin qi in earth".This theory synthesizes the clinical manifestations and pathological progression of CAG,asserting that its development often involves intertwined pathological factors such as stagnation,dryness-heat,phlegm-dampness and stasis-toxicity.A triphasic therapeutic framework is proposed:the spleen qi deficiency phase,marked by impaired spleen transport function and dysregulated qi-fluid distribution,requiring spleen fortification and qi-fluid regulation;the hyperactive stomach qi phase,characterized by intensified stomach qi activity coupled with dryness-damp stagnation,necessitating stagnation resolution,dampness elimination and yin nourishment;the decline and disorder of middle qi phase,characterized by the deficiency of the middle qi,with phlegm,blood stasis and toxins forming the terminal stage.Treatment should focus on reinforcing the middle and restoring balance,detoxifying and dissipating accumulation.By exploring CAG pathogenesis and treatment through the lens of"hyperactive stomach qi",this study aimed to provide novel theoretical insights and therapeutic strategies for TCM in the prevention and treatment of CAG.

Objective To develop the value of machine learning(ML)models based on contrast-transthoracic echocardiography(cTTE)and transesophageal echocardiography(TEE)combined with clinical and laboratory indicators for predicting patent foramen ovale-associated stroke(PFO-AS).Methods Totally 313 patients with PFO diagnosed with cTTE and TEE were retrospectively enrolled.Among them,65 cases were found complicated with ischemic stroke and confirmed as PFO-AS(PFO-AS group),and the rest 248 cases without ischemic stroke were classified as non-PFO-AS group.The patients were divided into training set(n=219,including 48 cases of PFO-AS and 171 cases of non-PFO-AS)and test set(n=94,including 17 cases of PFO-AS and 77 cases of non-PFO-AS)at the ratio of 7∶3.Univariable and multivariable logistic regression(LR)were used to analyze clinical and laboratory indicators as well as cTTE and TEE parameters in training set to screen independent predictive factors of PFO-AS.ML models,including LR,K-nearest neighbor(KNN),support vector machine(SVM),random forest(RF),decision tree(DT),back propagation neural network(BPNN)and gradient boosting machine(GBM)were constructed,and the predictive efficacy of the models for predicting PFO-AS was evaluated,then the optimal model was selected.Results Patient's age＞49-69 years,with smoking history,plasma albumin≥43.8 g/L,significant right-to-left shunt at rest shown on cTTE and complicated atrial septal aneurysm shown on TEE were all independent predictors of PFO-AS,which were used to construct ML models.The area under the curve(AUC)of LR,KNN,SVM,RF,DT,BPNN and GBM models in training set was 0.779-0.853,while in test set was 0.730-0.877.RF model had relatively high and comparable sensitivity,specificity and AUC in both training and test sets,also higher precision and smaller Brier score in test set,hence was regarded as the optimal ML model.Conclusion RF model based on cTTE and TEE combined with clinical and laboratory indicators could be used to effectively predict PFO-AS.

Objective To investigate the value of patient data-based real-time quality control(PBRTQC)in internal quality control(IQC)for blood cells analysis based on the data from patients.Methods The data of patients'blood cells,including white blood cell count(WBC),hemoglobin(Hb),red blood cell count(RBC),hematocrit(HCT),mean corpuscular volume(MCV),mean cor-puscular hemoglobin(MCH),mean corpuscular hemoglobin concentration(MCHC),and platelet(PLT)were collected from August 1,2023 to February 26,2024,and the extracted patient data were analyzed on the AI-based real-time quality control intelligent moni-toring platform.The corresponding IQC data for this period were reviewed,and the results of PBRTQC and IQC were compared and an-alyzed.The causes of the emerging warning or alarm prompts were checked and analyzed to explore the application value of PBRTQC in the IQC process of blood cell analysis.Results It is found that when the quality control product was unstable due to overlong opening time of the reagent or improper storage conditions,and the performance changes of the operating system during the detection process,the PBRTQC intelligent monitoring platform was able to issue risk warning or alarm prompt in advance.PBRTQC may have certain limi-tations,such as the error of red blood cell count,which need to be identified.Conclusion PBRTQC is superior to IQC in blood cell analysis and may play a complementary role in IQC.Meanwhile,it is necessary to exclude the possibility that PBRTQC is significantly influenced by the patient population in medical laboratories.

Median arcuate ligament compression syndrome (MALS) presents with atypical clinical manifestations. MALS is rarely reported in pediatric patients but is recognized as an independent risk factor for postoperative hepatic artery thrombosis in liver transplant recipients. We report a case of a pediatric liver transplant recipient with hepatolenticular degeneration, cirrhosis, and acute liver failure. Despite undergoing artificial liver support therapy, the patient showed no significant improvement in liver function and subsequently underwent liver transplantation. Intraoperatively, weak arterial pulsation and the absence of a pulsatile waveform in the hepatic artery anastomosis raised suspicion of MALS. The condition was successfully managed by releasing and transecting the median arcuate ligament, along with ligation of the splenic and left gastric arteries. To optimize transplantation outcomes, meticulous preoperative imaging assessment, particularly focusing on characteristic findings in CT angiography, is essential. Additionally, individualized surgical planning and intraoperative adjustments based on ultrasound monitoring and arterial pulsation assessments are critical for ensuring successful transplantation.

Objective:To explore clinical characteristics of multiple sclerosis(MS)patients in Suzhou,and to analyze main factors affecting their prognosis.Methods:General data,clinical symptoms,cerebrospinal fluid and imaging examinations of 51 MS patients admitted to Department of Neurology of the Second Hospital of Soochow University from July 31,2009 to July 31,2021 were retrospectively analyzed,and main factors affecting their prognosis were discussed.Results:Average age of onset of 51 MS patients was(43.3±15.6)years old,female accounted for 56.9%,male/female=1/1.3.Adult onset MS(AOMS)accounted for 62.8%,male/female=1/1.7;late onset MS(LOMS)accounted for 37.2%,male/female=1/0.9.Relapsing remitting MS(RRMS)accounted for 76.5%,and chronic onset accounted for 60.8%.Average annual recurrence rate was 8.8%.The first symptoms were numbness and weakness of limbs.Dizziness and numbness were more common in patients without recurrence after diagnosis of MS,and limb weak-ness and numbness were more common in patients with recurrence.Among lesions of MRI,62.7%(32/51)of periventricular involve-ment,52.9%(27/51)of spinal cord involvement,51.0%(26/51)of infratentorial involvement.Proportion of subtentorial and spinal cord(cervical,thoracic)involved were significantly higher in patients with recurrent MS than without recurrence.Values of albumin,IgG,IgA and IgM in cerebrospinal fluid increased with increase of recurrence times.EDSS score of male was higher than female,and LOMS score was higher than AOMS.MS patients without relapse had a low EDSS score,and median EDSS score at current follow-up was 0(0,1.00)score.MS score with relapse was relatively high,and median EDSS score at current follow-up was 2.75(0.25,7.25)score.Conclusion:MS patients with chronic onset are more common,with a high proportion of LOMS,and proportion of males increases with increasing age of onset.High EDSS score at first onset,cervical,thoracic and subtentorial lesions,increased values of cerebrospinal fluid albumin,IgG,IgA,IgM,age at first onset(50+years old),male associate with poor MS prognosis.

Chronic atrophic gastritis(CAG)is a chronic gastric disorder characterized by recurrent damage to the gastric mucosal epithelium,resulting in the reduction of intrinsic glands,with or without concurrent intestinal metaplasia.The"hyperactive stomach qi"theory,derived from Huang Di Nei Jing Su Wen Ji Zhu,proposes that the core pathogenesis of CAG lies in excessive stomach qi activity,grounded in the physiological principle of"strong yang qi in earth and weak yin qi in earth".This theory synthesizes the clinical manifestations and pathological progression of CAG,asserting that its development often involves intertwined pathological factors such as stagnation,dryness-heat,phlegm-dampness and stasis-toxicity.A triphasic therapeutic framework is proposed:the spleen qi deficiency phase,marked by impaired spleen transport function and dysregulated qi-fluid distribution,requiring spleen fortification and qi-fluid regulation;the hyperactive stomach qi phase,characterized by intensified stomach qi activity coupled with dryness-damp stagnation,necessitating stagnation resolution,dampness elimination and yin nourishment;the decline and disorder of middle qi phase,characterized by the deficiency of the middle qi,with phlegm,blood stasis and toxins forming the terminal stage.Treatment should focus on reinforcing the middle and restoring balance,detoxifying and dissipating accumulation.By exploring CAG pathogenesis and treatment through the lens of"hyperactive stomach qi",this study aimed to provide novel theoretical insights and therapeutic strategies for TCM in the prevention and treatment of CAG.

Objective To develop the value of machine learning(ML)models based on contrast-transthoracic echocardiography(cTTE)and transesophageal echocardiography(TEE)combined with clinical and laboratory indicators for predicting patent foramen ovale-associated stroke(PFO-AS).Methods Totally 313 patients with PFO diagnosed with cTTE and TEE were retrospectively enrolled.Among them,65 cases were found complicated with ischemic stroke and confirmed as PFO-AS(PFO-AS group),and the rest 248 cases without ischemic stroke were classified as non-PFO-AS group.The patients were divided into training set(n=219,including 48 cases of PFO-AS and 171 cases of non-PFO-AS)and test set(n=94,including 17 cases of PFO-AS and 77 cases of non-PFO-AS)at the ratio of 7∶3.Univariable and multivariable logistic regression(LR)were used to analyze clinical and laboratory indicators as well as cTTE and TEE parameters in training set to screen independent predictive factors of PFO-AS.ML models,including LR,K-nearest neighbor(KNN),support vector machine(SVM),random forest(RF),decision tree(DT),back propagation neural network(BPNN)and gradient boosting machine(GBM)were constructed,and the predictive efficacy of the models for predicting PFO-AS was evaluated,then the optimal model was selected.Results Patient's age＞49-69 years,with smoking history,plasma albumin≥43.8 g/L,significant right-to-left shunt at rest shown on cTTE and complicated atrial septal aneurysm shown on TEE were all independent predictors of PFO-AS,which were used to construct ML models.The area under the curve(AUC)of LR,KNN,SVM,RF,DT,BPNN and GBM models in training set was 0.779-0.853,while in test set was 0.730-0.877.RF model had relatively high and comparable sensitivity,specificity and AUC in both training and test sets,also higher precision and smaller Brier score in test set,hence was regarded as the optimal ML model.Conclusion RF model based on cTTE and TEE combined with clinical and laboratory indicators could be used to effectively predict PFO-AS.