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 influencing factors of repeat blood donor lapsing using a zero-inflated poisson regression model (ZIP). [Methods] The blood donation behavior of 12 498 whole blood donors from 2020 was tracked until December 31, 2023. The factors influencing the frequency of blood donations in a given year was analyzed using ZIP, and donors with 0 blood donation in that year were considered to have lapsed. The changes in relevant influencing factors associated with each blood donation were measured and modeled for analysis. [Results] The zero-inflated part of ZIP showed that the risk of lapsing of male blood donors was 2.24 times that of female blood donors (OR 95% CI:1.864-2.696, P<0.001); the risk of lapsing of the 35-44 age group and over 45 age group was respectively 40% (OR 95% CI:0.455-0.790, P<0.001) and 61%(OR 95% CI:0.268-0.578, P<0.001) lower than that of the under 25 age group; the risk of lapsing for those who have donated blood twice and ≥3 times was respectively 50% (OR 95% CI:0.405-0.609, P<0.001) and 81% (OR 95% CI:0.154-0.225, P<0.001) lower than that of first-time donors; the risk of lapsing of those with junior high or high school education was 1.2 times that of those with a college degree or higher (OR 95% CI:1.033-1.384, P<0.05); the risk of lapsing for the divorced group was 2.02 times that of the married group (OR 95% CI:1.445-2.820, P<0.001); the risk of lapsing for those with an income (Yuan) of 10 000 to 50 000, 50 000 to 100 000 and more than 100 000 was respectively 0.67 (OR 95% CI:0.552-0.818, P<0.001), 0.72 (OR 95% CI:0.591-0.884, P=0.002) and 0.67 (OR 95% CI:0.535-0.834, P<0.001) times that of those with an income (Yuan) of less than 10 000. The results of the Poisson part are consistent with the results of the zero-inflated part in terms of age and education level. [Conclusion] Blood donor lapsing is overall related to factors such as gender, age, donation frequency, education, marital status and family income. It's essential to care for those blood donors prone to lapse to retain more regular blood donors.

BACKGROUND:Bone relies on the close connection between blood vessels and bone cells to maintain its integrity.Bones are in a physiologically hypoxic environment.Therefore,the study of angiogenesis and osteogenesis in hypoxic environment is closer to the microenvironment in vivo. OBJECTIVE:To explore the influence of Wenshen Tongluo Zhitong(WSTLZT)formula on H-type bone microvascular endothelial cell/bone marrow mesenchymal stem cell co-culture system in hypoxia environment and its related mechanism. METHODS:Enzyme digestion method and flow sorting technique were used to isolate and identify H-type bone microvascular endothelial cells.Mouse bone marrow mesenchymal stem cells were isolated and obtained by bone marrow adhesion method.H-type bone microvascular endothelial cell/bone marrow mesenchymal stem cell hypoxic co-culture system was established using Transwell chamber and anoxic culture workstation.WSTLZT formula powder was used to intervene in each group at a mass concentration of 50 and 100 μg/mL.The angiogenic function of H-type bone microvascular endothelial cells in the co-culture system was evaluated by scratch migration test and tube formation test.The osteogenic differentiation ability of bone marrow mesenchymal stem cells in the co-cultured system was evaluated by alkaline phosphatase staining and alizarin red staining.The protein and mRNA expression changes of PDGF/PI3K/AKT signal axis related molecules in H-type bone microvascular endothelial cells in the co-cultured system were detected by Western Blotting and q-PCR,respectively. RESULTS AND CONCLUSION:(1)Compared with the normal oxygen group,the scratch mobility and new blood vessel length of H-type bone microvascular endothelial cells were significantly higher(P<0.05);the osteogenic differentiation capacity of bone marrow mesenchymal stem cells was higher(P<0.05);the expression of PDGF/PI3K/AKT axis-related molecular protein and mRNA increased(P<0.05)in the hypoxia group.(2)Compared with the hypoxia group,scratch mobility and new blood vessel length were significantly increased in the H-type bone microvascular endothelial cells(P<0.05);bone marrow mesenchymal stem cells had stronger osteogenic function(P<0.05);the expression of PDGF/PI3K/AKT axis-related molecular proteins and mRNA further increased(P<0.05)after treatment with different dose concentrations of WSTLZT formula.These findings conclude that H-type angiogenesis and osteogenesis under hypoxia may be related to the PDGF/PI3K/AKT signaling axis,and WSTLZT formula may promote H-type vasculo-dependent bone formation by activating the PDGF/PI3K/AKT signaling axis,thereby preventing and treating osteoporosis.

Objective:To develop and design a during-treatment records auto-review program to comply the quality assurance(QA)requirement of radiotherapy chart auditing,and thereby improve the review efficiency and accuracy.Methods:Based on the items the guideline required,the Aria Oncology Information System database backup files was analyzed by Java,Vue,and etc.languages and the corresponding review logic was formulated.A total of 530 treatment records generated at Shanghai Concord Cancer Center from January to March 2024(10 weeks)were auto-reviewed and compared with the manual results for evaluating the accuracy and efficiency of the program.Results:The auto-review program was running smoothly.Overall with the above data,the sensitivity,specificity,accuracy and the error-miss rate were 73.4%,14.3%,87.7%and 12.3%respectively.For sub-set items,the source-skin distance(SSD)error detecting rate was 100%,the wrong session reporting was 100%correlated with the plans switching and the wrong fraction reporting was 100%related to plan revision.For the other items,auto and manual reviews gave out the same accuracy.Conclusion:The none-error results from the program are all true,so the manual rechecking could limit to those auto-review error records,which can reduce the workload by 73.4%,therefore improve the effectiveness and accuracy of the radiotherapy data review.

Objective:To develop and design a during-treatment records auto-review program to comply the quality assurance(QA)requirement of radiotherapy chart auditing,and thereby improve the review efficiency and accuracy.Methods:Based on the items the guideline required,the Aria Oncology Information System database backup files was analyzed by Java,Vue,and etc.languages and the corresponding review logic was formulated.A total of 530 treatment records generated at Shanghai Concord Cancer Center from January to March 2024(10 weeks)were auto-reviewed and compared with the manual results for evaluating the accuracy and efficiency of the program.Results:The auto-review program was running smoothly.Overall with the above data,the sensitivity,specificity,accuracy and the error-miss rate were 73.4%,14.3%,87.7%and 12.3%respectively.For sub-set items,the source-skin distance(SSD)error detecting rate was 100%,the wrong session reporting was 100%correlated with the plans switching and the wrong fraction reporting was 100%related to plan revision.For the other items,auto and manual reviews gave out the same accuracy.Conclusion:The none-error results from the program are all true,so the manual rechecking could limit to those auto-review error records,which can reduce the workload by 73.4%,therefore improve the effectiveness and accuracy of the radiotherapy data review.

Objective:To evaluate the value of ultrasound radiomics combined with machine learning for early diagnosis of seronegative Hashimoto’s thyroiditis (SN-HT) .Methods:This retrospective study included 164 patients from Liaoning Provincial People’s Hospital , Lixin County People’s Hospital, Linghai Dalinghe Hospital, Fengcheng Phoenix Hospital, who underwent thyroidectomy for solitary nodules with normal thyroid function between Nov. 2016 and Jan. 2024. Postoperative pathology confirmed Hashimoto’s thyroiditis (HT) in some cases, who were further categorized into antibody-positive and antibody-negative groups based on serum antibody status. Patients without Hashimoto’s thyroiditis served as the control group. A total of 298 ultrasound images were analyzed. Radiomics features were extracted from hypoechoic non-nodular areas within 0.5 cm surrounding the tumor. Two senior pathologists and two senior ultrasound physicians independently assessed lymphocytic infiltration, eosinophilic changes of follicular epithelium, and the proportion of hypoechoic areas in pathology and ultrasound images, respectively. A machine learning model, CCH-NET, was developed using linear regression and t-distributed stochastic neighbor embedding (t-SNE) techniques. The dataset was divided into a training set (80%) and a validation set (20%) to compare the diagnostic accuracy of CCH-NET with that of senior ultrasound physicians. Results:In internal validation, CCH-NET achieved a diagnostic accuracy of 88.89% for both antibody-positive and antibody-negative groups, significantly higher than the 66.67% accuracy of senior ultrasound physicians ( P<0.01). In external validation, CCH-NET achieved 75.00% and 66.67% accuracy for the two groups, compared to 50.00% by senior ultrasound physicians. For the control group, both methods achieved 93.33% accuracy. The AUC of CCH-NET was 0.848, outperforming senior ultrasound physicians (0.681) ,demonstrating superior diagnostic performance. Conclusion:The radiomics-based CCH-NET model, using non-nodular hypoechoic areas as a specific indicator, can accurately identify early SN-HT in euthyroid patients. It significantly outperforms senior ultrasound physicians, improving diagnostic accuracy and reducing missed diagnoses.

Objective To compared the clinical efficacy of mechanical thrombectomy and standard medication therapy in the treatment of patients with acute vertebrobasilar arterial occlusive mild stroke.Methods The clinical data of a total of 105 patients with acute vertebrobasilar arterial occlusive mild stroke,who were admitted to the Tianchang Municipal People's Hospital of China from January 2019 to December 2023,were retrospectively analyzed.According to the therapeutic scheme,the patients were divided into mechanical thrombectomy group(n=37)and standard medication therapy group(n=68).Based on the age,sex,onset-to-admission interval,National Institutes of Health Stroke Scale(NIHSS)score,the propensity score matching with 1∶1 ratio was conducted for the patients.The clinical outcome was compared between the two groups.Results After propensity score matching,a total of 68 patients were enrolled in this study,with 34 patients in each group.The mean age was(68.3±11.9)years,45 patients(66.2％)were male,and the median NIHSS score was 2 points.No statistically significant differences in the baseline data existed between the two groups(all P＞0.05).In the standard medication therapy group,10 patients(29.4％)developed early deterioration of the neurological functions,and the median NIHSS score was 12 points,all the 10 patients received salvage mechanical thrombectomy,and 9 patients achieved a successful recanalization.Follow-up check at 90 days after surgery showed that modified Rankin scale(mRS)score of 0-1 point was seen in one patient,and 0-2 points was seen in 2 patients.The proportions of 90-day mRS score of 0-1 point and 0-2 points in the mechanical thrombectomy group were remarkably higher than those in the standard medical therapy group(67.6％vs.41.2％and 85.3％vs.61.8％respectively,both P＜0.05).No statistically significant differences in the hemorrhagic transformation rate(5.9％vs.8.8％)and 90-day mortality(5.9％vs.14.7％)existed between the two groups(both P＞0.05).Conclusion Compared with standard medication therapy,the mechanical thrombectomy can improve the clinical prognosis of patients with acute vertebrobasilar arterial occlusive mild stroke and will not increase the incidence of surgical complications.Therefore,acute vertebrobasilar occlusive mild stroke may be a good indication for mechanical thrombectomy treatment.

Objective To retrieve,evaluate and summarize the evidence on prevention and management of hypothermia in war trauma patients at high altitudes so as to provide data for related prevention and management.Methods Clinical decisions,practices recommended,summaries of evidence,guidelines,expert consensus,systematic reviews and randomized controlled trials concerning the prevention and management of hypothermia in patients with war trauma at high altitudes that were published between the inception and February 2024 were retrieved from related websites and Chinese and English databases.Results A total of fifteen articles were included,including 4 guidelines,1 clinical decision,3 summaries of evidence,1 recommended practice,2 systematic reviews,and 4 randomized controlled trials.Finally,29 pieces of evidence were made available,involving 6 themes:hypothermia assessment,risk factors,rewarming processes,rewarming methods,body temperature monitoring,and considerations.Conclusion This study has summed up the evidence on hypothermia management in patients with war trauma at high altitudes.It is recommended that nurses take into consideration the actual conditions and clinical scenarios and formulate personalized rewarming plans based on the results of assessment of patients when using evidence.

Objective:To investigate, for multi-sensitized allergic rhinitis (AR) patients allergic to dust mites combined with other allergens (pollen, mold, animal dander, etc.), whether the single dust mite subcutaneous immunotherapy (SCIT) can improve the specific symptoms caused by other allergens in the patients, and to analyze the relationship between the effectiveness of symptom improvement in these patients and the type, quantity and severity of the allergens.Methods:A multicenter retrospective study was conducted to collect mul-sensitized AR patients from allergy or respiratory departments of 5 hospitals who received house dust mite allergen preparation SCIT for 12 to 36 months and met other inclusion and exclusion criteria from February to July 2024. General clinical data were collected and the perennial or seasonal symptoms before and after treatment were evaluated with visual analogue scale (VAS) to assess whether there was an perennial or allergen-specific symptom improvement (VAS score decrease ≥30%), by which the patients were divided into effective group and ineffective. R software was used to analyze the differences between groups by using Fisher′s exact test and Mann-Whitney U test. Results:A total of 62 patients were enrolled, and the treatment were effective in 39 of them, with an effective rate of 62.9%. For allergen-specific symptoms, the median age of the effective group was higher than that of the ineffective group (12 years old vs. 8 years old, P=0.039), and the effective rate in dust mite specific immunoglobin E (sIgE) grade ≤5 group was higher than that in sIgE grade >5 group (81.6% vs. 45.5%, P=0.008), and the effective rate of mold sIgE grade ≤2 group was higher than that of sIgE grade >2 group (83.3% vs. 28.6%, P=0.045), and there was no statistically significant correlation between the other allergen grades and the effective rate ( P>0.05). For perennial symptoms, the effective rate in the mold grade ≤2 group was higher than that in the sIgE grade >2 group (91.3% vs. 28.6%, P=0.010), and there was no statistically significant correlation between the other allergen grades and the effective rate ( P>0.05). There was no significant correlation between the treatment effectiveness of perennial or allergen-specific symptoms and the number of combined allergens, the grade of skin test, and the difference between the grade of combined allergens and that of dust mites ( P>0.05). Conclusion:Among the patients with multi-sensitized AR allergic to dust mites included in this study, single dust mite SCIT is effective in some of them, and for allergen-specific symptoms, the effective group was elder, and dust mite sIgE grade 6 and mold sIgE grade ≥2 was related to the low effective rate of SCIT. The present results are insufficient for selecting single or multiple AIT in any type of multi-sensitized patients.