Objective To investigate the clinical application value of MRI Radiomics score(Radscore)combined with clinicopatho-logical features in predicting postoperative progression-free survival(PFS)of patients with endometrial cancer(EC).Methods A total of 127 patients with EC were selected.The radiomic features of the lesions were extracted from T2 WI,diffusion weighted imaging(DWI)and apparent diffusion coefficient(ADC)images.The features were screened by random forest model and Radscore was calcu-lated.Simultaneously,clinical and pathological characteristics of patients were collected and incorporated,and multivariate Cox regression analysis was used to screen the risk factors related to PFS.The MRI Radscore and clinicopathological features were mapped to the nomogram,and the performance of nomogram was evaluated by receiver operating characteristic(ROC)curve and calibration curve.Results Multivariate Cox regression analysis showed that progesterone receptor(PR),human epididymis protein 4(HE4)and MRI Radscore were independent risk factors for predicting PFS in patients with EC(P＜0.05).The area under the curve(AUC)of the predicted PFS at 1,3 and 5 years after surgery were 0.91,0.804 and 0.776,respectively.Calibration curves showed that nomogram had a good fit in predicting PFS in patients with EC 1,3 and 5 years after surgery.Conclusion The nomogram con-structed based on multi-sequence MRI Radscore and clinicopathological features has favorable accuracy and stability in predicting postoperative PFS in individuals diagnosed with EC.
Objective To evaluate the potency of applying an artificial intelligence(AI)based model for classifying vertebral fractures in lumbar X-ray images.Methods Patients who underwent lateral lumbar X-ray and MRI were retrospectively selected.Based on MRI results,the vertebrae were categorized as fresh fractures,old fractures,and normal vertebrae.A ResNet-18 classification model was constructed using delineated region of interest(ROI)on the X-ray images,and the model's performance was evaluated.Results A total of 272 patients(662 vertebrae)were included in this study.The vertebrae were randomly divided into training(n=529)and validation(n=133)sets.The model's performance in discerning normal vertebrae,fresh fractures,and old fractures revealed accuracy of 0.91,0.42,and 0.75,and the sensitivity were 0.91,0.408,and 0.72,while the specificity were 0.796,0.892,and 0.796,respectively.Conclusion The X-ray-based ResNet-18 AI model has significant accuracy for distinguishing old fractures and normal vertebrae;However,the model's accuracy needs further improvement for distinguishing fresh fractures.
目的 分析非小细胞肺癌(NSCLC)瘤体及含瘤周影像组学模型对其 Ki-67 表达状态的预测效能,进而探讨基于最优影像组学模型的联合模型列线图的预测效能.方法 回顾性分析经病理证实为 NSCLC且有Ki-67 表达结果的 301 例患者的临床及影像学资料.利用瘤体及瘤体联合瘤周 5 mm、10 mm 组织 3 种感兴趣区(ROI)提取影像组学特征.使用最大相关最小冗余(MRMR)和最小绝对收缩和选择算子(LASSO)2 种算法进行特征筛选.对 3 组 ROI所筛选特征分别使用逻辑回归(LR)及支持向量机(SVM)分类器构建影像组学模型.最终在 6 组模型中寻找最优影像组学模型并计算影像组学评分.利用二元 logistic 回归模型构建临床-影像组学联合模型列线图,使用曲线下面积(AUC)及校准曲线等评估模型效能.结果 基于瘤体联合LR分类器构建的影像组学模型表现最好[训练集 AUC 0.855,95％置信区间(CI)0.803～0.906;测试集 AUC 0.850,95％CI 0.768～0.932].联合模型在训练集中 AUC 0.905(95％CI 0.862～0.948),在测试集中 AUC 0.864(95％CI 0.789～0.939).结论 6 组模型中,瘤体联合 LR构建的影像组学模型预测 NSCLC Ki-67 表达状态的效能最高.基于联合模型列线图有助于临床鉴别 NSCLC Ki-67 表达状态.
目的 探讨CT直接征象及间接征象对肠穿孔术前诊断的临床价值.方法 回顾性分析经手术证实的 44 例肠穿孔患者临床及CT表现,将CT表现分为直接征象和间接征象.将所有患者以具备直接征象(A组)、间接征象≥3 个(B组)及间接征象＜3 个(C组)作为诊断条件分为 3 组,以术中诊断作为金标准,对各组术前诊断符合率进行对比分析,同时对好转及治愈的患者平均住院时间进行统计.结果 诊断符合率比较:A组 100％;B组 84％;C组 40％.3 组间肠穿孔诊断符合率差异有统计学意义(P＜0.05);而 A组与B组组间比较差异无明显统计学意义(P＞0.05);B组与C组组间比较差异有统计学意义(P＜0.05).平均住院时间比较:A组(13.5±6.21)d;B组(17.2±8.34)d;C组(28.1±22.9)d.A组与B组比较差异无明显统计学意义(P＞0.05);B组与 C组比较差异有统计学意义(P＜0.05).结论 CT直接征象对肠穿孔的诊断价值高,可明显减少患者平均住院时间;当直接征象不显著时,间接征象越多,诊断准确性越高,患者平均住院时间也越少.
目的 探讨高分辨率磁共振成像(HR-MRI)在发生于头面部中高危区域的皮肤基底细胞癌(BCC)的术前诊断中的价值.方法 收集13例头面部中高危区域BCC患者的临床及影像资料,对病变的大小、信号特征、边缘、强化、动态增强时间-信号强度曲线(TIC)类型、深度及累及层次进行观察和评估.结果 病灶平均最大径为(19.4±11.0)mm.T1WI均呈等或低信号,T2WI信号表现多样.9例病灶边界模糊.10例病灶均匀强化,所有病灶均呈等、低强化,TIC表现为I型或 Ⅱ型.病灶平均深度为(7.2±3.2)mm,9例累及皮下脂肪以下深层结构.结论 BCC在HR-MRI表现多样,低强化、低灌注为相对特异表现.HR-MRI可以术前无创评估病变的范围及对周围结构的浸润,对指导治疗和手术决策具有较大价值.
患者 男,70岁.因反复咳嗽、咳痰,心累、气促5年余,加重10 d人院.5年前,患者无明显诱因出现咳嗽、咳痰,无规律,间断性咳嗽,咳白色泡沫痰,伴劳动耐力下降.体温36.5℃;血常规,白细胞6.56×109/L,血红蛋白75 g/L;动脉血气分析:pH 7.40,动脉血氧分压 38 mmHg(1 mmHg=0.133 kPa),动脉二氧化碳分压50 mmHg.
Objective To investigate the clinical value of high-resolution magnetic resonance imaging(HR-MRI)combined with nomogram model in predicting the risk factors of positive circumferential resection margin(CMR)in rectal cancer surgery.Methods A retrospective analysis was conducted on preoperative data from 107 rectal cancer patients who underwent surgery and were confirmed by pathology.SPSS 2 5.0 software was used for univariate statistical analysis of potential risk factors for positive CRM,and after independent risk factors were selected,multivariate logistic regression analysis was performed to classify the risk factor categories.R software(4.2.0 version)was used to establish the nomogram model,and a curve was drawn to evaluate the model.The receiver operating characteristic(ROC)curve was used to show specificity and sensitivity,the area under the curve(AUC)was used to evaluate discriminative ability,the calibration curve was used to evaluate calibration,the decision curve analysis(DCA)was used to evaluate clinical benefit,and the model was internally validated using the Bootstrap method.Results Tumor located in the lower rectum[P=0.01,odds ratio(OR)=8.71],maximum diameter of tumor perpendicular to the intestinal tube(≥18.86 mm±5.32 mm)(P=0.01,OR=1.24),extramural vascular invasion(EMVI)(P＜0.01,OR=0.03),and mesorectal lymph node metastasis(P=0.01,OR=0.15)were independent risk factors for positive CRM in rectal cancer.The nomogram model established based on these factors had a training set AUC of 0.921(sensitivity 0.83,specificity 0.93),a validation set AUC of 0.912(sensitivity 0.87,specificity 0.82),a Bootstrap internal validation corrected AUC of 0.92,and a consistency index(C-index)of 0.92,indicating good discriminative ability,calibration,and clinical benefit value.Conclusion HR-MRI combined with nomogram model predicted that tumor located in the lower rectum,maximum diameter of tumor perpendicular to the intestinal tube ≥(18.86±5.32)mm,EMVI,and mesorectal lymph node metastasis are closely related to positive CRM in rectal cancer.
Objective To explore the apparent diffusion coefficient(ADC)value of rectal cancer lesions obtained by two measurement methods in predicting the pathological types of moderately and poorly differentiated rectal adenocarcinoma.Methods A total of 41 patients were divided into moderately differentiated group and poorly differentiated group according to degrees of differentiation.Direct measurement method and minimum ADC value method were used to measure the ADC value of the tumor,and the effectiveness of the ADC value in predicting moderately and poorly differentiated rectal adenocarcinoma was analyzed.Results The ADC values of rectal adenocarcinoma tumor obtained by two different measurement methods were different between the moderately and poorly differentiated groups(P＜0.01).The area under the curve(AUC)of the direct measurement method was 0.753,and the sensitivity and specificity were 96.67％and 63.64％,respectively.The AUC of the minimum ADC value method was 0.838,the sensitivity and specificity were 76.67％and 81.82％,respectively.The ADC value obtained by direct measurement method was positively correlated with its pathological characteristics(r=0.519,P＜0.01)and that by the minimum ADC value method was positively correlated with its pathological characteristics(r=0.391,P＜0.05).Conclusion The pathological types of moderately and poorly differentiated rectal adenocarcinoma is determined by ADC value.It is found that the direct measurement method has more greater comprehensive clinical value than the minimum ADC value method.
Objective To evaluate the performance of MRI in predicting pathological response of different breast cancer subtypes after neoadjuvant chemotherapy(NAC).Methods The MRI images and postoperative pathological results of 91 patients with breast cancer after NAC were analyzed retrospectively.The correlation between the imaging features of different molecular subtypes of breast cancer and postoperative pathological results was studied,and the diagnostic performance of MRI in predicting pathological response after NAC was evaluated,with postoperative pathological results referred as the diagnostic standard.Results Of 91 patients,27(29.7％)and 35(38.5％)cases were diagnosed as imaging complete response(iCR)and pathological complete response(pCR),respectively.The accuracy of MRI in predicting pathological response after NAC was 84.62％,with 94.64％sensitivity,68.57％specificity,and positive predictive value(PPV)and negative predictive value(NPV)of 82.81％and 88.89％,respectively.Conclusion MRI can accurately predict the pathological response of the human epidermal growth factor receptor-2(HER-2)+and triple-negative breast cancer after NAC.
Objective To develop a nomogram based on pulmonary nodules preoperative CT signs and 3D quantitative parameters for predicting mediastinal lymph node metastases in patients with clinical stage ⅠA lung adenocarcinoma.Methods The imaging data of 164 patients who underwent preoperative CT scan and systematic lymph node dissection were analyzed retrospectively.Commercially available AI software was used to extract 3D quantitative parameters of pulmonary nodules automatically,and CT signs of pulmonary nodules were analyzed.Logistic regression was used to explore the role of these parameters in predicting pathological nodal involvement.A nomogram prediction model was established,then discrimination and calibration of the model were evaluated.Results Among 164 enrolled patients,19(11.6％)were tested positive for mediastinal lymph node metastases at pathology review.The nomogram incorporated spiculation,lobulation,the largest cross-sectional area,and carcinoembryonic antigen(CEA).The model showed great discrimination and calibration,with a C-index of 0.942[95％confidence interval(CI)0.923-0.961].The predicted value of the model fitted well with the actual observed value on the calibration curve.Conclusion The nomogram prediction model based on preoperative CT signs,3D quantitative parameters,and CEA can estimate the probability of mediastinal lymph node metastases in clinical stage ⅠA lung adenocarcinoma.This model may help with clinical decision-making and individualized evaluation.
Objective To explore the application value of PDCA quality management in calcaneal weight-bearing long-axis photographic image quality management.Methods The Quality Management Team of the adopted the PDCA image quality control management model to analyze the current situation of calcaneal weight-bearing long-axis photographic examination and the factors that contributed to the image quality decline.The specific factors contributing to the degradation of the image quality were identified,and solutions were developed and implemented.Paired chi-square tests were used to compare the frequency of high-quality calcaneal weight-bearing long-axis images taken before and after PDCA management.Results The paired chi-square tests revealed statistically significant differences between the two groups(X2=9.370,P=0.002).Conclusion PDCA image quality control management could improve the image quality of calcaneal weight-bearing long-axis radiographs and provide clinicians with a more reliable imaging basis for hindfoot anatomy,force line measurement,and the diagnosis of hindfoot malalignment.
多发性骨髓瘤(multiple myeloma,MM)是一种克隆浆细胞异常增生的恶性肿瘤,仅次于白血病的第二大常见肿瘤.异常增殖浆细胞及其产物会破坏相关靶器官引起对应的临床症状,其中骨痛是最常见的症状.
Objective To explore the predictive value of artificial intelligence(AI)quantitative model for pathological subtypes of stage Ⅰ invasive lung adenocarcinoma with ground glass nodule(GGN).Methods A total of 118 cases(124 lesions)of GGN patients with stage Ⅰ invasive lung adenocarcinoma confirmed by surgery and pathology were analyzed retrospectively,and they were divided into lepidic predominant adenocarcinoma(LPA)group(46 lesions)and non-lepidic predominant adenocarcinoma(n-LPA)group(78 lesions)according to the pathological subtype results.Some relevant AI quantitative parameters were recorded,including the longest diameter,total volume,the percentage of solid volume,total mass,the percentage of solid mass,maximum CT value,minimum CT value,and average CT value.The independent predictors of n-LPA were screened by univariate and multivariate logistic regression analysis,the independent risk factors were quantified by Nomogram,and the diagnostic efficiency of the model was evaluated by using receiver operating characteristic(ROC)curve.Results Binomial logistic regression analysis showed that the percentage of solid mass[odds ratio(OR)=1.965,95％confidence interval(CI)1.515-2.549]and average CT value(OR=1.020,95％CI 1.004-1.036)were independent predictors of n-LPA(P＜0.05).The Nomogram to quantify the independent risk factors showed that the above prediction model was in good agreement with the actual results,and the C-index value was 0.872(95％CI 0.791-0.953).ROC curve analysis showed that the diagnostic performance of the combination of the above two indexes[area under the curve(AUC)=0.829]was better than that of the solid mass percentage(AUC=0.788)and the average CT value(AUC=0.765)of the single indexes,and the corresponding sensitivity and specificity were 87.2％and 84.8％,respectively,which were consistent with the pathological results(Kappa=0.667).Conclusion The percentage of solid mass and the average CT value in the AI quantitative model can effectively help predict the pathological subtypes of GGN stage Ⅰ invasive lung adenocarcinoma,and the combination of the above two indicators can improve the differential diagnosis efficiency of CT between LPA and n-LPA.
Objective To explore the value of a radiomics model based on contrast-enhanced CT at the arterial and venous phases in predicting the regional lymph node metastasis in rectal cancer.Methods The data of 303 patients with pathologically proven rectal cancer were analyzed retrospectively,and 1 648 CT radiomics features of the primary tumors were extracted from the arterial and venous phases,respectively.Four machine learning models were used including radiomics score(Rad-score),K-nearest neighbor(KNN),multi-layer perceptron(MLP)and support vector machine(SVM),and the model with the best diagnostic performance was selected.Taking the predicted probability of the best machine model in the arterial and venous phases as the input value,logistic regression algorithm was used to further construct a combined nomogram.Results The SVM model had been proved to be the most efficient among the four machine learning models,whether in arterial or venous phases.The combined nomogram based on ASVM and VSVM prediction probability further improved the ability to predict regional lymph node metastasis.The area under the curve(AUC)of the train group was 0.860,and that of the validation group was 0.801,which were higher than those of ASVM model(train group,AUC=0.822;validation group,AUC=0.777)and VSVM model(train group,AUC=0.832;validation group,AUC=0.786).Conclusion The arterial-venous phases radiomics combined nomogram model based on enhanced CT images performs well in predicting rectal cancer patients'pre-treatment lymph node metastasis.
Objective To compare the fetal imaging quality in different positions and the maternal comfort level in a super large-bore magnet,so as to estimate the application value of left lying position scanning in fetal brain MRI.Methods A total of 32 pregnant women in the third trimester who underwent fetal brain imaging in a super large-bore magnet(75cm)were collected.Each subject completed left lying position and supine position scanning and scored the comfort of the two.Radiologists scored the image subjectively,and objectively measured signal-to-noise ratio(SNR)and contrast-to-noise ratio(CNR)of fetal brain images.Results The comfort score of the left lying position(4.3±0.7)was higher than that of the supine position(3.4±0.7),and the difference was statistically significant(P＜0.001).The subjective score showed that there was no significant difference between the left lying position and the supine position in terms of brain image quality(P＞0.05),which could meet the needs of clinical diagnosis.The objective score showed that SNR of midbrain in left lying position was higher than in supine position(P＜0.05).The CNR of left occipital lobe in left lying position was higher than in supine position(P＜0.05).Conclusion In super large-bore magnet,left lying position scanning is a safe and pregnancy-friendly option that can not only improve maternal comfort level,but also improve fetal brain imaging.
Objective To quantitatively study the pathological characteristics of the joint disc in the early stage of temporomandibular joint dysfunction based on 3.0T MR T2 mapping technology.Methods There were 90 patients in the study,including 30 in the control group and 60 in the experimental group.All patients were clinically diagnosed with early temporomandibular joint dysfunction.The 90 patients underwent conventional 3.0T MR and T2 mapping scanning in open and closed mouth positions,the T2 mapping pseudo-color images of the temporomandibular joint disc was observed,and the T2 mapping values at the anterior extension,anterior,middle,posterior bands of the temporomandibular joint disc,as well as the upper and lower plates of the double plate area were quantitatively measured.The T2 mapping values of the above sites were compared between and within the two groups.Results The T2 mapping values in the anterior extension,anterior band,middle band,posterior band,upper plate and lower plate of the bilateral temporomandibular joint disc in the open and closed positions of the experimental group were significantly higher than those in the control group(P＜0.05).The T2 mapping values in the open and closed positions of the experimental group were higher than those in other areas(P＜0.05).There was no significant difference in T2 mapping values between the control group and the experimental group in the observation areas of the bilateral tem-poromandibular joint disc at the open and closed positions(P＞0.05).Conclusion The 3.0T MR T2 mapping can be used to quantitatively assess the joint disc injury of temporomandibular joint dysfunction in the early stage,and the value of T2 mapping in the middle band of the disc in the experimental group is higher than that in other observation areas.
患者 女,54岁,因头晕、左侧肢体麻木无力6 h余人院.MRI+MR血管成像示急性/亚急性期脑梗死,颅内血管多发狭窄,增强CT示左肾结石积水并左肾萎缩(图1).给予对症治疗后完善全脑血管造影,局麻下经右股动脉插管,在导丝引导导管上行过程中,导丝误入右肾动脉分支内,撤出导丝后患者未诉特殊不适,遂继续完成全脑血管造影术.
Objective To explore the utility of CT for the evaluation of abdominal aortic calcification(AAC)in chronic kidney disease(CKD).Methods Laboratory examination indexes and abdominal plain CT imaging of 132 CKD stage 3-5 patients were analyzed retrospectively.The ACC score was evaluated according to quantitative method,the risk factors related to AAC were analyzed.Results In correlation analyses,AAC score was positively correlated with creatinine(Cr)level,CO2 combining power(P＜0.05),and negatively corrected with albumin(ALB).Logistic regression analysis showed that Cr level were risk factors for AAC,and ALB were protective factors.Conclusion Based on abdominal plain CT images,Cr level,CO2 combining power and ALB level are related to AAC.Quantitative method based on CT has a high application value in evaluating AAC.
Objective To analyze the demographic data,CT characteristics,treatment methods and follow-up results of patients with spontaneous isolated dissection of celiac artery(SIDCA)and to explore the selection of treatment strategy for SIDCA.Methods Medical records of 31 patients diagnosed as SIDCA were selected.The patients were divided into 2 groups,symptomatic group and asymptotic group.Demographic data,morbidity season,CT characteristics(type,distance of the entry site from the origin of the artery,dissection length,compression rate of the true lumen,branches involvement,organ ischemia,coexisting vasculopathy),treatment methods,and follow-up results were analyzed.Results Thirty-one patients were identified by CT contrast scan,22 were symptomatic and 9 were asymptomatic.22 patients were first diagnosed in cooler season,compared with 9 patients in warmer season(22 vs 9,Fisher's exact test,P=0.029).The difference between symptomatic and asymptomatic groups about branches involvement in patients was significant(8/14 vs 0/9,Fisher's exact test,P=0.007).Treatment included observation in 24,endovascular intervention in 6 patients and surgical repair in 1 patient.No patient required bowel resection.The mean follow-up period was 13.75 months.Except for 2 patients,the condition of the remaining patients improved or stabilized during follow-up.Conclusion The difference between symptomatic and asymptomatic groups about branches involvement in patients is significant.Initial conservative treatment may be adequate for patients without end organ malperfusion or aneurysm formation or aneurysm rupture in SIDCA.Additionally,during the cooler season,visceral artery should be observed carefully on abdominal contrast CT,especially in the patient with abdominal pain,to avoid misdiagnosis.
Objective To explore the risk factors of early neurological deterioration(END)in patients with acute cerebral infarction(ACI)based on the characteristics of computed tomography angiography(CTA)images of head and neck plaques.Methods A total of 218 ACI patients were selected,and according to whether END occurred within 72 h after onset,they were divided into an END group(54 cases)and non END group(164 cases).The clinical data and imaging indicators were compared and analyzed between the two groups.Multivariate logistic regression analysis was used to analyze risk factors,and a nomograph prediction model was constructed and evaluated.Results National Institutes of Health Stroke Scale(NIHSS)score at admission ≥4 points,glycated hemoglobin(HbA1c)≥7.7％,severe or above ipsilateral carotid artery stenosis,and unstable plaque(soft plaque and mixed plaque)were the independent risk factors in END patients(P＜0.05).The result of nomograph prediction model showed that its discrimination,accuracy and effectiveness were better.Conclusion NIHSS score at admission,HbA1c level,ipsilateral carotid artery stenosis degree,and plaque type are all independent influencing factors for the occurrence of END.
