The accurate segmentation of breast ultrasound images is an important precondition for the lesion determination. The existing segmentation approaches embrace massive parameters, sluggish inference speed, and huge memory consumption. To tackle this problem, we propose T ²KD Attention U-Net (dual-Teacher Knowledge Distillation Attention U-Net), a lightweight semantic segmentation method combined double-path joint distillation in breast ultrasound images. Primarily, we designed two teacher models to learn the fine-grained features from each class of images according to different feature representation and semantic information of benign and malignant breast lesions. Then we leveraged the joint distillation to train a lightweight student model. Finally, we constructed a novel weight balance loss to focus on the semantic feature of small objection, solving the unbalance problem of tumor and background. Specifically, the extensive experiments conducted on Dataset BUSI and Dataset B demonstrated that the T ²KD Attention U-Net outperformed various knowledge distillation counterparts. Concretely, the accuracy, recall, precision, Dice, and mIoU of proposed method were 95.26%, 86.23%, 85.09%, 83.59%and 77.78% on Dataset BUSI, respectively. And these performance indexes were 97.95%, 92.80%, 88.33%, 88.40% and 82.42% on Dataset B, respectively. Compared with other models, the performance of this model was significantly improved. Meanwhile, compared with the teacher model, the number, size, and complexity of student model were significantly reduced (2.2×10 ⁶ vs. 106.1×10 ⁶, 8.4 MB vs. 414 MB, 16.59 GFLOPs vs. 205.98 GFLOPs, respectively). Indeedy, the proposed model guarantees the performances while greatly decreasing the amount of computation, which provides a new method for the deployment of clinical medical scenarios.
Humans ; Breast Neoplasms/diagnostic imaging* ; Female ; Ultrasonography, Mammary/methods* ; Image Processing, Computer-Assisted/methods* ; Algorithms ; Neural Networks, Computer ; Breast/diagnostic imaging*

With rapid development of artificial intelligence(AI)technology,AI is reshaping new ecology in field of educa-tion.With promotion of AI,AI digital teaching materials will further clearly present knowledge context and build a new paradigm and new form of digital teaching materials.In context of general health,integration of medicine and prevention has become an important trend in medical and health field.However,at present,there is a lack of medical and preventive integration resources,and a single paper textbook cannot achieve depth and breadth of knowledge system of medical and preventive integration,which is difficult to adapt to development trend of AI enabling education.Therefore,it is urgent to use AI technology to develop and expand digital teaching materials for medical and prevention integration and realize resource penetration.Immune system is an important defense line of physi-cal diseases,and there is a close relationship between integration of medical and prevention.This paper aims to explore application path of developing digital teaching materials in medical education by AI technology.This concept of medical integration runs through every chapter and case analysis of AI textbooks,strengthening shift from"treatment-centered"to"health-centered"concept.Taking immune system as an example,this paper conducts in-depth analysis,develops and shares digital teaching materials by integrating medical knowledge,prevention strategies and AI technology,and provides new ideas and methods for medical education.

Objective:To characterize the brain structure of Chinese children with autism spectrum disorder (ASD) using artificial intelligence automatic brain segmentation technique, and to analyze the correlation between the characteristics of the brain structure and the degree of brain development.Methods:A case-control study.The data of 52 children who were diagnosed with ASD according to the diagnostic criteria for ASD in the Diagnostic and Statistical Manual of Mental Disorders-Fifth Edition of the United States at the Department of Psychology of Qingdao University Affiliated Women and Children′s Hospital from January 2023 to April 2024 were prospectively analyzed.Meanwhile, 48 gender- and age-matched typically developing (TD) children in Qingdao were also included.The three-dimensional T1 weighted imaging sequences of all patients were obtained using a GE 3.0T magnetic resonance imaging scanner.Automated brain segmentation techniques were used to obtain the standardized volumes of each brain structure (the ratio of the absolute volume of the brain structure to the whole brain volume).Two-independent-samples t and Mann-Whitney U tests were used to compare the standardized volumes of different brain regions between the 2 groups.Pearson and Spearman correlation analyses were used to depict the correlations between volume data of brain areas with significant differences and Gesell Developmental Scale scores. Results:Compared with those in the TD group, the volumes of the left grey matter[25.45%(0.70%) vs.25.16%(1.05%)], the right grey matter [(25.89±0.71)% vs.(25.51±0.73)%], the right lateral orbitofrontal cortex [(0.62±0.03)% vs.(0.59±0.05)%], the right medial orbitofrontal cortex[(0.48±0.04)% vs.(0.46±0.04)%], the right pars triangularis [(0.38±0.07)% vs.(0.35±0.05)%], the left hippocampus [0.22%(0.04%) vs.0.20%(0.02%)], the right hippocampus [0.23%(0.04%) vs.0.22%(0.02%)], the left parahippocampal gyrus [0.15%(0.03%) vs.0.14%(0.02%)], the right parahippocampal gyrus [(0.15±0.02)% vs.(0.14±0.02)%], the left fusiform gyrus [(0.82±0.08)% vs.(0.78±0.08)%], the right superior temporal gyrus [(0.96±0.10)% vs.(0.90±0.09)%], the left insular lobe [(0.54±0.03)% vs.(0.53±0.04)%], the right insular lobe [(0.55±0.03)% vs.(0.53±0.04)%], the right inferior parietal cortex [(1.40±0.16)% vs.(1.33±0.12)%], the right precuneus cortex [(0.99±0.09)% vs.(0.94±0.09)%], the right putamen [(0.37±0.04)% vs.(0.35±0.03)%], the left pallidum [(0.14±0.01)% vs.(0.13±0.01)%], the right pallidum [0.14%(0.02%) vs.0.13%(0.01%)], and the right thalamus [(0.51±0.04)% vs.(0.49±0.03)%] were significantly increased in the ASD group (all P<0.05).Nonetheless, the volumes of the left pericalcarine cortex [(0.19±0.04)% vs.(0.20±0.04)%] and the corpus callosum posterior region [0.05%(0.01%) vs.0.06%(0.01%)] in the ASD group were considerably smaller than those in the TD group (all P<0.05).Correlation analysis showed that the right thalamus volume was negatively correlated with the Gesell-adaptation development quotient in children with ASD ( r=-0.276, P=0.048).The volumes of the left fusiform gyrus and left pericalcarine cortex were negatively correlated with the Gesell-fine motor development quotient in children with ASD ( r=-0.290, P=0.037; r=-0.368, P=0.007). The right precuneus cortex volume was negatively correlated with the Gesell-personal and social competence development quotient in children with ASD ( r=-0.396, P=0.007). Conclusions:Children with ASD show abnormalities in the volumes of multiple brain regions, and some brain regions are related to the degree of brain development.Automatic brain segmentation technology based on artificial intelligence can rapidly and directly measure and display the volume of brain structures in both ASD and TD children.

With rapid development of artificial intelligence(AI)technology,AI is reshaping new ecology in field of educa-tion.With promotion of AI,AI digital teaching materials will further clearly present knowledge context and build a new paradigm and new form of digital teaching materials.In context of general health,integration of medicine and prevention has become an important trend in medical and health field.However,at present,there is a lack of medical and preventive integration resources,and a single paper textbook cannot achieve depth and breadth of knowledge system of medical and preventive integration,which is difficult to adapt to development trend of AI enabling education.Therefore,it is urgent to use AI technology to develop and expand digital teaching materials for medical and prevention integration and realize resource penetration.Immune system is an important defense line of physi-cal diseases,and there is a close relationship between integration of medical and prevention.This paper aims to explore application path of developing digital teaching materials in medical education by AI technology.This concept of medical integration runs through every chapter and case analysis of AI textbooks,strengthening shift from"treatment-centered"to"health-centered"concept.Taking immune system as an example,this paper conducts in-depth analysis,develops and shares digital teaching materials by integrating medical knowledge,prevention strategies and AI technology,and provides new ideas and methods for medical education.

Objective:To characterize the brain structure of Chinese children with autism spectrum disorder (ASD) using artificial intelligence automatic brain segmentation technique, and to analyze the correlation between the characteristics of the brain structure and the degree of brain development.Methods:A case-control study.The data of 52 children who were diagnosed with ASD according to the diagnostic criteria for ASD in the Diagnostic and Statistical Manual of Mental Disorders-Fifth Edition of the United States at the Department of Psychology of Qingdao University Affiliated Women and Children′s Hospital from January 2023 to April 2024 were prospectively analyzed.Meanwhile, 48 gender- and age-matched typically developing (TD) children in Qingdao were also included.The three-dimensional T1 weighted imaging sequences of all patients were obtained using a GE 3.0T magnetic resonance imaging scanner.Automated brain segmentation techniques were used to obtain the standardized volumes of each brain structure (the ratio of the absolute volume of the brain structure to the whole brain volume).Two-independent-samples t and Mann-Whitney U tests were used to compare the standardized volumes of different brain regions between the 2 groups.Pearson and Spearman correlation analyses were used to depict the correlations between volume data of brain areas with significant differences and Gesell Developmental Scale scores. Results:Compared with those in the TD group, the volumes of the left grey matter[25.45%(0.70%) vs.25.16%(1.05%)], the right grey matter [(25.89±0.71)% vs.(25.51±0.73)%], the right lateral orbitofrontal cortex [(0.62±0.03)% vs.(0.59±0.05)%], the right medial orbitofrontal cortex[(0.48±0.04)% vs.(0.46±0.04)%], the right pars triangularis [(0.38±0.07)% vs.(0.35±0.05)%], the left hippocampus [0.22%(0.04%) vs.0.20%(0.02%)], the right hippocampus [0.23%(0.04%) vs.0.22%(0.02%)], the left parahippocampal gyrus [0.15%(0.03%) vs.0.14%(0.02%)], the right parahippocampal gyrus [(0.15±0.02)% vs.(0.14±0.02)%], the left fusiform gyrus [(0.82±0.08)% vs.(0.78±0.08)%], the right superior temporal gyrus [(0.96±0.10)% vs.(0.90±0.09)%], the left insular lobe [(0.54±0.03)% vs.(0.53±0.04)%], the right insular lobe [(0.55±0.03)% vs.(0.53±0.04)%], the right inferior parietal cortex [(1.40±0.16)% vs.(1.33±0.12)%], the right precuneus cortex [(0.99±0.09)% vs.(0.94±0.09)%], the right putamen [(0.37±0.04)% vs.(0.35±0.03)%], the left pallidum [(0.14±0.01)% vs.(0.13±0.01)%], the right pallidum [0.14%(0.02%) vs.0.13%(0.01%)], and the right thalamus [(0.51±0.04)% vs.(0.49±0.03)%] were significantly increased in the ASD group (all P<0.05).Nonetheless, the volumes of the left pericalcarine cortex [(0.19±0.04)% vs.(0.20±0.04)%] and the corpus callosum posterior region [0.05%(0.01%) vs.0.06%(0.01%)] in the ASD group were considerably smaller than those in the TD group (all P<0.05).Correlation analysis showed that the right thalamus volume was negatively correlated with the Gesell-adaptation development quotient in children with ASD ( r=-0.276, P=0.048).The volumes of the left fusiform gyrus and left pericalcarine cortex were negatively correlated with the Gesell-fine motor development quotient in children with ASD ( r=-0.290, P=0.037; r=-0.368, P=0.007). The right precuneus cortex volume was negatively correlated with the Gesell-personal and social competence development quotient in children with ASD ( r=-0.396, P=0.007). Conclusions:Children with ASD show abnormalities in the volumes of multiple brain regions, and some brain regions are related to the degree of brain development.Automatic brain segmentation technology based on artificial intelligence can rapidly and directly measure and display the volume of brain structures in both ASD and TD children.

Endoscopic ultrasonography(EUS)can not only directly observe the morphology of the lumens of the digestive tract through the endoscope,but also perform real-time dynamic ultrasound scanning to obtain the layers of the lumen and the structure of surrounding organs.With the wide application and technical improvement of EUS,we gradually realize that it plays an important role in the diagnosis and treatment of cholangiocarcinoma.This article reviews the application of EUS in the diagnosis and treatment of cholangiocarcinoma.

An important issue facing university education is the adaptation and transformation of freshmen.In order to guide students to adapt to the needs of research-based learning,Jilin University has opened a freshman seminar for all first-year undergraduate students,emphasizing the cultivation of students'thinking patterns and learning abilities.For medical students,it is necessary to ap-ply the learned medical knowledge to clinical problem analysis.Therefore,it is particularly important to cultivate students'integrated thinking from basic to clinical at the beginning of enrollment.In recent years,organ system integration has gradually been carried out in medical schools both domestically and internationally.In the course of integrating the immune system and diseases,the morphology of the immune system is relatively abstract and difficult to observe.In the early stages of enrollment,students still lack the thinking of organ system integration,and the initial stage of immune system integration courses is difficult to carry out.Therefore,we aimed to guide students towards adaptability and academic transformation through freshman seminars,explored the integration of the immune system's organ system as a pathway,and used the BOPPPS teaching model as a means for preliminary exploration.In the freshman seminar,the main focus is on the"morphology and structure-mechanism and function-disease-diagnosis-treatment"of the immune sys-tem,helping students establish a learning mindset that integrates disease diagnosis and treatment based on basic medical courses such as immunology and anatomy.Using BOPPPS to reform the"teaching"of teachers,establishing a student-centered teaching model from six aspects:bridge,objective,pre-assessment,participatory learning,post-assessment and summary,can better guide students to actively think and participate in the entire learning process,focus on the construction of students'fragmented knowledge thinking mode,internalize medical humanities literacy,and improve the quality of medical talent cultivation in colleges and universities.

Purpose: To identify the predictors of infectious disease-specific health literacy (IDSHL), and establish an easy-to-apply nomogram to predict the IDSHL of older adults. Methods: This cross-sectional study included 380 older adults who completed the IDSHL, self-rated health, socio-demographic and other questionnaires. Logistic regression was used to identify the IDSHL predictors. Nomogram was used to construct a predictive model. Results: Up to 70.1% of older adults had limited IDSHL. Age, education, place of residence, self-rated health, and Internet access were the important influencing factors of IDSHL. The established nomogram model showed high accuracy (receiver operating characteristic curve: 0.848). Conclusions The IDSHL of Chinese older adults was significantly deficient. The constructed nomogram is an intuitive tool for IDSHL prediction that can not only contribute toward rapid screening of high-risk older adults with limited IDSHL but also provide guidance for healthcare providers to develop prevention strategies for infectious diseases.

Objective To explore the application value of rehabilitation strategies guided by cardiopulmonary exercise testing (CPET) in patients after interventional therapy for myocardial infarction. Methods A total of 96 patients after interventional therapy for myocardial infarction were selected as study subjects and divided into control group (n=48) and observation group (n=48) according to the order of admission. The control group received traditional cardiac rehabilitation strategies, while the observation group received rehabilitation strategies guided by dynamic CPET. The medication compliance rate and the incidence of adverse events were compared between the two groups. Peak oxygen uptake, anaerobic threshold, left ventricular ejection fraction (LVEF), 6-minute walk test (6MWT) distance, scores of the Self-Efficacy for Exercise (SEE) and the Chinese Quality of Life Questionnaire for Cardiovascular Patients (CQQC) were compared between the two groups before and after surgery. Results Three months after surgery, the medication compliance rate in the observation group was 87.50%, which was significantly higher than 68.75% in the control group (P < 0.05). At 1 month and 3 months after surgery, the LVEF, peak oxygen uptake, anaerobic threshold, SEE score and 6MWT distance in the observation group were significantly higher or larger than those in the control group (P < 0.05). Three months after surgery, the scores of the CQQC in the observation group were significantly higher than those in the control group (P < 0.05). Three months after surgery, the incidence of adverse events in the observation group was 4.17%, which was significantly lower than 18.75% in the control group (P < 0.05). Conclusion The rehabilitation strategies guided by dynamic CPET have high application value in patients after interventional therapy for myocardial infarction, helping them establish medication compliance behavior, improve cardiopulmonary function and quality of life, and reduce the incidence of adverse events.

Mycoplasma capricolum subsp. capripneumoniae (Mccp) is the cause of contagious caprine pleuropneumonia (CCPP) in goats. Inactivated vaccines and capsular polysaccharide (CPS) indirect hemagglutination reagents are available for prevention and serological detection, but high culture costs and complex antigen quantification have been plagued by production staff. In order to solve these problems in production practice, a sugar fermentation medium with an initial pH value of 7.8, which could improve the production of two antigens simultaneously, was screened out by changing the initial pH value based on previous Mccp metabolomics analysis. Since phenol red can be identified by UV absorption spectrum and cetyltrimethylammonium bromide (CTAB) can bind to anionic capsular polysaccharide, a UV spectrum measurement method for analyzing the culture stage reached by Mccp and a CTAB precipitation test for relative quantification of capsular polysaccharide antigen content in the fermentation broth were established. The UV spectrum observation method can guide the production of Mccp according to the growth curve of Mccp, which greatly reduces the monitoring time of the traditional CCU method and improves the accuracy of the original eye-observation method. The established CTAB precipitation test can complete the monitoring of CPS content within 5 hours, which greatly reduces the time required compared with the traditional differential technique, and its accuracy was verified by the phenol-sulfuric acid method. The optimized culture medium and the two correlation comparison methods established in this study can effectively reduce the production cost of Mccp and improve the production efficiency. The two assays have been used in the research at our laboratory, which provides experimental data for further improvement of the production process of CCPP inactivated vaccine and capsular polysaccharide as well as rapid quantification.
Humans ; Animals ; Goats ; Cetrimonium ; Mycoplasma ; Polysaccharides