Recent advances in large models demonstrate significant prospects for transforming the field of medical imaging. These models, including large language models, large visual models, and multimodal large models, offer unprecedented capabilities in processing and interpreting complex medical data across various imaging modalities. By leveraging self-supervised pretraining on vast unlabeled datasets, cross-modal representation learning, and domain-specific medical knowledge adaptation through fine-tuning, large models can achieve higher diagnostic accuracy and more efficient workflows for key clinical tasks. This review summarizes the concepts, methods, and progress of large models in medical imaging, highlighting their potential in precision medicine. The article first outlines the integration of multimodal data under large model technologies, approaches for training large models with medical datasets, and the need for robust evaluation metrics. It then explores how large models can revolutionize applications in critical tasks such as image segmentation, disease diagnosis, personalized treatment strategies, and real-time interactive systems, thus pushing the boundaries of traditional imaging analysis. Despite their potential, the practical implementation of large models in medical imaging faces notable challenges, including the scarcity of high-quality medical data, the need for optimized perception of imaging phenotypes, safety considerations, and seamless integration with existing clinical workflows and equipment. As research progresses, the development of more efficient, interpretable, and generalizable models will be critical to ensuring their reliable deployment across diverse clinical environments. This review aims to provide insights into the current state of the field and provide directions for future research to facilitate the broader adoption of large models in clinical practice.
Humans ; Diagnostic Imaging/methods* ; Precision Medicine/methods* ; Image Processing, Computer-Assisted/methods*

Objective:To summarized the clinical characteristics and surgical management of anterior cervical enterogenic in pediatric patients. Methods:Clinical data were retrospectively analyzed for 9 children with pathologically confirmed anterior cervical enterogenic cysts（including bronchogenic and esophagogenic subtypes） treated at the Children's Hospital of Shandong University（Jinan Children's Hospital） between January 1, 2020, and November 30, 2023. Results:Nine patients（6 males and 3 females） were involved in this study, aged 14 days to 10 years old. There were 4 cases on the left side, 4 on the right side, and 1 in the middle of the neck. All patients presented with neck masses. The patients were followed up from 3 months to 35 months after surgery and recovered well, with no recurrence or complications observed. Conclusion:①Anterior intestinal cysts in children are rare and easy to be misdiagnosed. ②Concurrent branchial cleft fistulas or associated anomalies may coexist, necessitating comprehensive evaluation. ③Preoperative diagnosis is not easy and mainly depends on pathological diagnosis. ④The treatment of anterior cervical enterogenic cysts in children is surgical resection of the lesion.
Humans ; Male ; Female ; Child ; Retrospective Studies ; Child, Preschool ; Infant ; Neck ; Cysts/surgery*

Objective: To establish a progressive research strategy for “colonic components analysis - efficacy verification and mechanism exploration - gut microbiota”, screen pharmacodynamic substances, and investigate their mechanism via gut microbiota. Methods: The pharmacodynamics of Gegen Qinlian decoction (GQD) were assessed using a mouse model of dextran sulfate sodium-induced ulcerative colitis (UC). Ultra-performance liquid chromatography-quadrupole-orbitrap mass spectrometer was used to identify the prototype and metabolic components of GQD in the colon during UC. To analyze the structure and function of characteristic genera of GQD and its active components, 16S rRNA sequencing was performed. Results: We identified 67 prototypic and 14 metabolic components of GQD in the UC colon. The primary prototype components are flavonoids and alkaloids, including puerarin (PUE), baicalin (BAI), and berberine (BER). The metabolism was predominantly sulfonation. Efficacy verification showed that the main active components, puerarin, baicalin, and berberine, had good therapeutic effects on UC. The results of 16S rRNA gene sequencing showed that GQD improved UC by regulating the structure and function of the gut microbiota. The abundance of gut microbiota involved in the metabolism of the prototype components was influenced by the corresponding components. The function prediction results showed that PUE was the most comparable to GQD, with 24 consistent pathways. BAI and BER showed comparable gut microbiota regulation pathways. Characteristic pathways of BER include glucometabolic processes. Conclusion This study focused on the key issues in the gut microbiota pathway and developed a progressive research strategy to understand the transformation mechanisms of colonic components. This research systematically analyzed the active components and metabolic transformation of GQD in the colon during the pathological state of UC, as well as changes in the structure and function of the gut microbiota, clarified the mechanism of GQD and its active components in improving UC via the gut microbiota pathway.

Objective: The aim of the present study was to develop the beverage addiction scale for college students,so as to provide an effective tool for assessing college students beverage addiction. Methods: In November 2022, a cluster sample of 8 792 college students from three colleges in Yunnan and Jiangxi were surveyed by Beverage Addication Scale for College Students. After a through literature review, 12 items were proposed, including withdrawal symptoms, health effects and addiction symptoms, with 4 items each. The ttest and correlation analysis were used to filter the items, and exploratory factor analysis and confirmatory factor analysis were used to evaluate the structural validity of the questionnaire. In May 2023, 5 279 college students from the above three universities were surveyed again to evaluate the reliability and validity of the scale and the positive demarcation value of symptoms. Results: The scale fitted into a tool for measuring symptoms of beverage dependence in a college student population. The scale was composed of 11 items in accordance with withdrawal symptoms, health effects and addiction symptoms, with a cumulative contribution rate of variance was 74.51%. Cronbach α coefficients for the overall scale and three dimensions were 0.94, 0.88, 0.90 and 0.92, respectively. The correlation coefficient between each item and the total score ranged from 0.56 to 0.79, and the correlation coefficient with the dimension ranged from 0.71 to 0.92. The confirmatory factor analysis model was well fitted, and the RMSEA, CFI, TLI and SRMR value were 0.06, 0.95, 0.93 and 0.04, respectively. The application of the scale showed that scores on the scale and each dimension ≥P85 were positive for symptoms. Conclusion The reliability and validity of Beverage Addiction Scale for College Students is good, which can be used to evaluate the beverage addiction symptom of college students.

Objective: To explore the association between processed food consumption and anxiety symptoms among college students in Yunnan Province, so as to provide a reference for the prevention and treatment of anxiety symptoms in this population. Methods: A cluster random sample of 2 515 first year students from two universities in Yunnan Province was selected to carry out a longitudinal investigation which included a baseline survey (November 2021, T1) and three follow up visits (June 2022, T2; November 2022, T3; June 2023, T4). The food frequency questionnaire was administered to assess processed food consumption, and the Depression Anxiety Stress Scale-21 (DASS-21, Chinese version) was used to evaluate anxiety symptoms. A generalized estimation equation model was used to analyze the relationship between processed food consumption and anxiety symptoms. Results: The detection rates of T1-T4 anxiety symptoms among college students in Yunnan Province were 29.70%, 36.70%, 37.69% and 38.73 %, respectively, and the corresponding anxiety symptom scores were 4(0,8), 4(0,10), 4(0,12), 2(0,14). After controlling for demographic variables and confounding factors in the generalized estimation equation model, a statistically significant association was found between consumption of carbonated beverages ( β=0.06, 95%CI =0.03-0.08), and other processed snacks ( β= 0.04 , 95%CI =0.01-0.07) ( P <0.05). The stratified analysis by gender showed that the consumption of carbonated beverages ( β=0.08, 95%CI =0.05-0.12) and fast food ( β=0.03, 95%CI =0.00-0.06) was significantly associated with anxiety symptoms in female college students ( P <0.05). There was no significant association between processed food consumption and anxiety symptoms in male college students ( P >0.05). Conclusions Processed food consumption by college students in Yunnan Province may increase the risk of anxiety symptoms, particularly among female students. There is a need to strengthen guidance in respect to processed food consumption, so as to prevent and treat anxiety symptoms.

Objective To explore the feasibility of automatic segmentation of clinical target volume(CTV)and organs at risk(OARs)for cervical cancer using AccuLearning(AL)based on geometric and dosimetric indices.Methods Seventy-five CT localization images with manual contouring data of postoperative cervical cancer were enrolled in this study.Sixty cases were randomly selected to trained to generate automatic segmentation model by AL,and the CTV and OARs of the remaining 15 cases were automatically contoured.Radiotherapy plans on the automatic segmentation contours were imported on the CT images of manual contours.The efficiency,Dice similarity coefficient(DSC),Hausdorff distance(HD)and dosimetric parameters were compared between the two methods.Results The time of automatic segmentation was significantly shorter than that of the manual contour(P<0.05).The DSC of all structures were≥0.87.The HD of bowel bag and rectum were about 10 mm,and that of the rest of OARs were less than 5 mm.CTV(D98,V90% ,V95% ,Dmean,HI),bowel bag(V50)and bladder(V50)had significant differences in dosimetric comparison(P<0.05).Conclusion The automatic segmentation model based on AL can improve the efficiency of radiotherapy.Automatic segmentation of OARs has the potential of clinical application,while that of CTV still needs to be further modified.

Objective: To study the relationship between sugar sweetened beverages consumption characteristics and insomnia of college students in Yunnan Province, so as to provide evidence for sleep quality improvement of college students. Methods: A cluster random sampling method was used to select 2 515 college students from two universities (Kunming University and Dali Nursing Vocational College) in Kunming and Dali in Yunnan Province for a longitudinal study, including baseline survey (T1, November 2021) and three follow up surveys (T2: June 2022, T3: November 2022, T4: June 2023). Sugar sweetened beverages consumption of college students was collected by Semi quantitative Food Frequency Questionnaire and insomnia was assessed by Insomnia Severity Index Scale. Sugarsweetened beverages consumption was analyzed by Kruskal-Wallis test. The Mann-Whiter U test and Kruskal-Wallis test were used to compare the detection rate of insomnia in college students with different population characteristics, and the generalized estimating equations model was established to analyze the association between sugar sweetened beverages consumption and insomnia. Results: The reported rate of insomnia among college students from T1 to T4 was 21.2%, 23.6%, 30.5 % and 36.0%, respectively. The median of sugar sweetened beverages consumption per week was 5 (1,9) bottles per person, and there were significant differences in sugar sweetened beverages (carbonated beverages, fruit beverages, tea beverages, milk beverages, energy beverages) consumption among college students with different insomnia status ( χ 2=42.91, 23.67, 29.98, 61.70, 30.82, P <0.01). The analysis of the generalized estimating equation model revealed that the consumption of carbonated beverages ( β= 0.04, 95%CI =0.00-0.08) and the consumption of milk beverages among college students ( β=0.04, 95%CI =0.00-0.09) were correlated with insomnia ( P <0.05). The stratified analysis indicated that consumption of carbonated beverages by female college students was associated with insomnia [ β(95%CI )=0.06(0.01-0.11)]; consumption of milk beverages among college students from middle income family was associated with insomnia [ β (95% CI )=0.05(0.00-0.10)], and consumption of carbonated beverages and fruit beverages from college students with high household economic status were both associated with insomnia [ β (95% CI )=0.35(0.23-0.46), 0.12(0.00-0.24)] ( P <0.05). Conclusion Sugar sweetened beverages, especially carbonated beverages, are associated with insomnia among college students in Yunnan Province.

The research focus of diabetes, a common chronic metabolic disease, has shifted from individual factors to environmental factors at the population level. Epidemiological studies suggest an association between exposure to environmental pollutants and the risk of diabetes. Major environmental pollutants include organochlorine pesticides, polychlorinated biphenyls, phthalates and their metabolites, and arsenics, which primarily enter the human body through the skin, respiratory tract, and digestive system. Long-term exposure to these pollutants can affect the pathology of diabetes through various mechanisms, such as promoting insulin resistance, causing insulin secretion deficiencies, inducing oxidative stress-induced glucose metabolism disorders, and affecting DNA methylation. However, the potential damaging mechanism of the impact of environmental pollutants on diabetes remain unclear. Limitations such as insufficient sample sizes, uncertainties regarding exposure time and dosage, and differences between single and co-exposures. In the future, it is necessary to focus on exploring and analyzing the mechanisms of environmental pollutant exposure on diabetes to develop effective prevention strategies, control and reduce the incidence and development of diabetes, and provide new insights into its diagnosis and treatment.

Objective To compare the advantages of robot assisted and DSA guided PBC for the treatment of trigeminal neuralgia.Methods A retrospective clinical analysis was conducted on 85 patients(28 in robot group and 57 in DSA group)who underwent robot assisted and DSA guided PBC surgery in a same center from September 2021 to February 2024.The single puncture success rate,improvement rate of VAS score,clinical efficacy rate,incidence of complications,average surgical time and fluoroscopy time were compared between the two groups.Results There was no statistically significant difference between the two groups in terms of single puncture success rate(96.43% vs.84.21% ),VAS improvement rate[88.9% (77.78%,100.00% )vs.88.89% (55.56%,100.00% )],clinical effective rate(92.86% vs.94.74% ),and total incidence of complications(35.09% vs.42.11% )(P>0.05).The average surgical time was significantly higher in the robot group[38.50(35.00,48.00)min]than the DSA group[19.00(15.00,25.50)min],and the average fluoroscopy time in the robot group[13.00(12.00,15.75)s]was significantly lower than the DSA group[194.00(152.50,259.50)s].The difference in average surgical and fluoroscopy time between the two groups was statistically significant(P<0.05).Conclusion DSA guided surgery has more advantages in centers with a large number of patients and a pursuit of efficiency.The robot assisted surgical puncture process is safe and controllable and patient radiation exposure time is short,thereby having high clinical application and promotion value.

Objective To analyze the incidence and disease burden of substance use disorder(SUD)in China from 1990 to 2019,to evaluate the impact of different ages,periods and birth cohorts on the disease burden of SUD,and to predict disease burden of SUD from 2020 to 2034,so as to provide strategies for the prevention of SUD.Methods Based on the Global Burden of Disease Study 2019(GBD 2019)database,the disease burden was described by incidence,years of life lost(YLLs),years lived with disability(YLDs)and disability-adjusted life years(DALYs).The Joinpoint regression model was used to analyze the trend of standardized incidence and standardized DALYs rate of SUD.Based on the age-period-cohort model,the age,period and cohort effects of SUD were discussed.The grey prediction model GM(1,1)was used to fit the trend of the incidence and standardized incidence of SUD and the trend of disease burden,and to predict the incidence and disease burden of SUD in 2020-2034.Results From 1990 to 2019,the standardized incidence of SUD of amphetamines[average annual percentage change(AAPC)=-0.9％]and cocaine(AAPC=-0.5％)in China showed a downward trend(P＜0.001),and the standardized incidence of SUD of cannabis(AAPC=0.9％)showed an increasing trend year by year(P＜0.001).The trend of standardized incidence of opioid abuse disorders was not obvious(P＞0.05).The DALYs rate caused by the 4 SUD showed a decreasing trend year by year(AAPCamphetamines=-2.2％,AAPCcocaine=-1.5％,AAPCcannabis=-1.0％,AAPCopioids=-1.0％,P＜0.001).The results of age-period-cohort effect showed that the peak incidence of amphetamine,cocaine,cannabis and opioid use disorders was in the 25-30 age group.The DALYs rate caused by cannabis SUD increased with age,while the DALYs rates of amphetamines,cocaine and opioids SUD reached the peak in the 25-29,30-34 and 35-39 age groups,respectively.The results of period effect showed that the risk of SUD in propylamines,cocaine and cannabis decreased first and then increased,while the risk of SUD in opioids increased and then decreased and increased again.The results of birth cohort effect showed that the risk of SUD of amphetamines,cocaine and opioids showed a decreasing trend as a whole except for a small fluctuation in individual birth cohorts.The risk of DALYs rate caused by SUD of amphetamines,cocaine and opioids showed a decreasing trend as a whole,while the risk of DALYs rate caused by SUD of cannabis showed an increasing trend year by year.The prediction results showed that the incidence of SUD of amphetamines,cocaine and opioids showed a downward trend from 2020 to 2034,and the incidence of SUD of cannabis showed a fluctuating upward trend.The DALYs attributed to SUD of amphetamines,cocaine,cannabis and opioids showed a decreasing trend year by year.Conclusion The disease burden of SUD in China is decreasing year by year in the future.The incidence and disease burden are affected by age effect,period effect and cohort effect to varying degrees.Early prevention and effective intervention are the key measures to control SUD.