1.Research progresses of the correlation between high resolution MRI characteristics of carotid atherosclerotic plaque and cardiovascular events
Rui QIN ; Chong ZHENG ; Jie LU
Chinese Journal of Medical Imaging Technology 2025;41(6):985-988
Cardiovascular disease has high disability and mortality rates,posing a serious threat to human health.Carotid atherosclerosis is closely related to cardiovascular events.Being non-invasive,high resolution(HR)-MRI has high resolution and unique advantages for evaluating characteristics of carotid plaques.The research progresses in correlations of HR-MRI characteristics of carotid atherosclerotic plaque and cardiovascular events were reviewed in this article.
2.Research progresses of ultrasound combined with artificial intelligence for evaluating hepatic fibrosis
Chinese Journal of Medical Imaging Technology 2025;41(6):997-1000
Chronic liver disease is one of the leading causes of death worldwide.Accurate early assessment of the severity of hepatic fibrosis is crucial for management of liver disease.Ultrasonography,with its non-invasive,portable and real-time imaging characteristics,has become the preferred method for imaging evaluation of chronic liver diseases.Artificial intelligence(AI)technology can extract massive quantitative features from ultrasound images for subsequent analysis,hence attracted increasing attention in evaluation of hepatic fibrosis.The research progresses of ultrasound combined with AI for evaluating hepatic fibrosis were reviewed in this article.
3.Machine learning models based on ultrasonic texture features of coronary artery for predicting incomplete Kawasaki disease in children
Yixiang LIN ; Juncheng NI ; Chi ZHANG ; Mulin SU ; Yi WU ; Qiuqin XU
Chinese Journal of Medical Imaging Technology 2025;41(7):1091-1096
Objective To explore the value of machine learning(ML)models based on ultrasonic texture features(TF)of coronary artery for predicting incomplete Kawasaki disease(IKD)in children.Methods Forty-eight children with IKD and 48 children without KD(non-KD)were enrolled with propensity score matching and divided into training set(n=67,34 cases of IKD and 33 cases of non-KD)and test set(n=29,14 of IKD and 15 of non-KD)at the ratio of 7∶3.Based on clinic-laboratory indicators(C-L)in training set and TF obtained with texture analysis of coronary artery ultrasound images,the optimal C-L-related features and TF were selected.Based on the optimal C-L correlated features,TF and their combinations,6 ML models,including random forest(RF),support vector machine(SVM),logistic regression(LR),gradient boosting decision tree(GBDT),decision tree(DT)and eXtreme gradient boosting(XGBoost)were respectively constructed for predicting IKD in children.The models were then trained in training set and validated in test set,and the best C-L ML,TF ML and C-L-TF ML models were selected.The area under the curve(AUC)of the best ML models were compared,and the clinical value of the best TF ML model was observed with decision curve analysis(DCA).Results Totally 3 optimal C-L related features and 8 optimal TF were selected.Among the constructed C-L ML,TF ML and C-L-TF ML models,C-L-LR model,TF-LR model and C-L-TF-SVM model were the optimal ones,with AUC in training set of 0.891,0.985 and 0.965,while in test set of 0.676,0.971 and 0.948,respectively.No significant difference of AUC was found between TF-LR model and C-L-TF-SVM model in both training set and test set(both P>0.05),which were both greater than those of C-L-LR model(all P<0.05).TF-LR model achieved higher clinical benefits in both training set and test set.Conclusion Ultrasound TF-LR model of coronary artery could be used to effectively predict IKD in children.
4.Correlations of chest CT quantitative parameters and readmission within one year in patients with acute exacerbation of chronic obstructive pulmonary disease
Guoqing SUN ; Lu YANG ; Liyu HE ; Min TIAN ; Bingjie ZHU ; Bahadori PARDIS ; Yanrong CHEN ; Chenwang JIN
Chinese Journal of Medical Imaging Technology 2025;41(7):1097-1102
Objective To observe the correlations of chest CT quantitative parameters and one-year readmission in patients with acute exacerbation of chronic obstructive pulmonary disease(AECOPD).Methods Totally 225 AECOPD patients were included and divided into the readmission group(group A,n=61)and non-readmission group(group B,n=164)according to whether readmitted due to acute exacerbation within one year or not.Quantitative CT parameters,including the percentage of low attenuation area volume to total lung volume(LAA%)and intrapulmonary vascular volume(IPVV)of the whole lung,left/right lung and each lobe,as well as the bronchial lumen area(LA)and percentage of bronchial wall area(WA%)of grades 3-8 were compared between groups.Spearman correlations were used to analyze the correlations between quantitative CT parameters and one-year readmission.Results LAA%of the whole lung,right lung,upper lobe of both lungs,middle lobe of the right lung,and lower lobe of the right lung in group A were higher than that in group B(all P<0.05),which were weakly positively correlated with readmissions within one year(r=0.142-0.187,all P<0.05).Significant differences of the upper lobe of right lung LAgrade 6,lower lobe of left lung LAgrade 6,lower lobe of left lung LAgrade 8,upper lobe of right lung WA%grade 5,upper lobe of right lung WA%grade 8 and lower lobe of left lung WA%grade 8 were found between groups,which were all correlated with readmissions within one year except for the lower lobe of the left lung LAgrade 6(all P<0.05).Conclusion One-year readmission in AECOPD patients might be associated with severe airway remodeling and emphysema.
5.Machine learning model based on chest non-contrast CT radiomics for diagnosing metabolic syndrome in males
Yi WEI ; Zhimin DING ; Jian ZHAI ; Xingwang WU
Chinese Journal of Medical Imaging Technology 2025;41(7):1103-1108
Objective To observe the value of machine learning(ML)model based on chest non-contrast CT(NCCT)radiomics for diagnosing metabolic syndrome(MetS)in males.Methods A total of 792 males who would undergo chest NCCT and bone density CT examination for physical check-up were prospectively enrolled and divided into training set(n=554,including 171 cases of MetS)and validation set(n=238,including 70 cases of MetS)at the ratio of 7∶3.Chest NCCT was performed,ROI of liver,intra-abdominal fat and skeletal muscle were delineated,and visceral fat area(VFA)at L2-3 intervertebral disc level was measured.Then radiomic signature(RS)of liver,intra-abdominal fat and skeletal muscle were established,and ML models were constructed using logistic regression(LR),random forests(RF)and extreme gradient boosting(XGBoost)algorithms,respectively,and their diagnostic performance were observed.Results Significant difference of age was found between MetS and non-MetS males in training set(P=0.010),while of RS scores were noticed in both training set and validation set(all P<0.001).Combined ML models were constructed with age and RS.The area under the curve(AUC)of combined LR,RF and XGBoost models for diagnosing male MetS in training set was 0.899,0.996 and 0.943,while that in validation set was 0.861,0.860 and 0.876,respectively.Combined XGBoost model had the best performance.Conclusion XGBoost model based on chest NCCT radiomics was helpful for diagnosing male MetS.Combining with age could further improve its efficacy.
6.Imaging guided percutaneous microwave ablation for unresectable pancreatic cancer:A multicenter retrospective study
Shuilian TAN ; Jie ZHOU ; Ping LIANG ; Xiaoling YU ; Xin YE ; Gang DONG ; Xiang JING ; Guanghui HUANG ; Zhen WANG ; Mengfan PENG ; Yan ZHOU ; Jie YU ; Zhiyu HAN ; Fangyi LIU ; Hongjian GAO ; Yubo ZHANG ; Zhigang CHENG
Chinese Journal of Medical Imaging Technology 2025;41(7):1109-1112
Objective To explore the feasibility and safety of ultrasound-guided percutaneous microwave ablation for unresectable pancreatic cancer.Methods Totally 84 patients who underwent ultrasound-guided percutaneous microwave ablation for unresectable pancreatic cancer were enrolled,and the technical success rate,complete ablation rate,complication rate,pain relief rate and survival time,etc.were observed.Results The median age of 84 cases was 61.5 years.Totally 86 tumors,including 44.19%(38/86)at the head/neck and 55.81%(48/86)at the body/tail of pancreas were detected,and a total of 85 ablation sessions were performed with the median ablation energy applied per tumor of 9.90(1.08,21.60)kJ and the complete ablation rate of 42.86%(36/84).The technical success rate was 100%(85/85).Thirty-nine complication events occurred in 25 cases,no ablation-related death.Among 34 patients underwent ablation mainly for pain symptoms,the pain score decreased from(6.22±1.12)points before treatment to(1.94±1.64)points after treatment(P<0.001).During 6.8(3.3,12.9)months' follow-up,the mean survival time was(8.5±6.7)months,and all 47 patients died due to tumor progression.Conclusion Ultrasound-guided percutaneous microwave ablation was safe and feasible for unresectable pancreatic cancer.
7.Research progresses of multi-organ imaging in special environments:Current status,techniques and challenges
Chinese Journal of Medical Imaging Technology 2025;41(7):1033-1036
Special environments have significant impacts on the structure and function of multiple organs in human body,thereby causing a series of physiological or pathological adaptive changes.Imaging represents a pivotal technological approach for visualizing the structural damage and functional changes in organs.The research progresses of multi-organ imaging in special environments were reviewed in this article.
8.Application status and future prospects of artificial intelligence large models in medical practice
Chinese Journal of Medical Imaging Technology 2025;41(8):1195-1199
Artificial intelligence(AI)large models are becoming the core driving force in the field of healthcare,providing new solutions to address challenges such as population aging,chronic disease burden and uneven distribution of medical resources.The AI model based on Transformer architecture significantly improved multi task generalization ability through massive data self-supervised learning and fine-tuning techniques,enabling collaborative optimization of precise diagnosis and treatment processes,acceleration of drug development cycles and innovation of medical education models.However,this technology faces bottlenecks such as lack of evaluation criteria,risks of data privacy exposure,limitations in model reliability and computing power resources.The application status of AI large models in the medical and health field,including precision diagnosis and treatment,drug development,and medical education were reviewed,and the technological challenges and future development trends were analyzed in this article.
9.Changes of brain functional connectivity in clinical frontline physicians under high occupational stress risk
Yanwei ZHANG ; Han WANG ; Zexuan WANG ; Cunzhao NIU ; Yan ZHANG ; Ge ZHANG
Chinese Journal of Medical Imaging Technology 2025;41(7):1037-1041
Objective To observe changes of brain functional connectivity in clinical frontline physicians under high occupational stress risk using gradient analysis based on MRI.Methods Totally 25 clinical frontline physicians were prospectively enrolled as observation group,while 25 non-medical staff in hospital taken as controls(control group).Brain resting-state functional MRI were acquired and preprocessed with Brainspace toolbox.Brain functional connectivity gradients were calculated,and the gradient components were compared between groups.Results In hierarchical structures of gradient components 1 and 2,significant increase of gradient components were noticed in observation group,mainly including default mode network,visual network and dorsal attention network related brain regions.Conclusion Gradient components of single modal brain regions in clinical frontline physicians significantly increased under high occupational stress risk.
10.Preoperative magnetization transfer imaging for predicting pancreatic fistula after distal pancreatectomy
Mingming YANG ; Ya LAN ; Derui HU ; Junxin LYU ; Xinyue ZHANG ; Jinggang ZHANG ; Jie CHEN ; Wei XING
Chinese Journal of Medical Imaging Technology 2025;41(7):1117-1120
Objective To observe the value of preoperative magnetization transfer imaging(MTI)for predicting postoperative pancreatic fistula(POPF)after distal pancreatectomy(DP).Methods A total of 65 patients with pancreatic tumor who underwent DP and preoperative MR scanning were retrospectively enrolled and divided into clinically relevant POPF(CR-POPF)group(n=14,with grade B or C fistula),biochemical fistula group(n=31,postoperative drain fluid amylase level exceeding 3 times the upper limit of normal)and non-fistula group(n=20,postoperative drain fluid amylase level not exceeding 3 times the upper limit of normal)based on postoperative records.Clinical data and magnetization transfer ratio(MTR)of pancreatic tissue at the surgical margin were compared among 3 groups.The predictive value of MTR for CR-POPF was evaluated according to the area under the curve(AUC)of receiver operating characteristic(ROC)curve.Results Patients' age,intraoperative blood loss and the proportion of pancreatic ductal adenocarcinoma in both CR-POPF group and biochemical fistula group were lower than those in non-fistula group(all adjusted P<0.05),while no significant difference was found between the former two groups(all adjusted P>0.05).MTR of pancreatic tissue at the surgical margin in CR-POPF group was lower than that in both biochemical fistula group and non-fistula group(both P<0.05),whereas no statistical difference was detected between the latter two groups(P>0.05).The AUC of MTR for predicting CR-POPF after DP was 0.727.Conclusion Preoperative MTI could be used to predict POPF after DP.

Result Analysis
Print
Save
E-mail