Magnetic resonance elastography (MRE), as a non-invasive biomechanical imaging technique, has demonstrated significant value in the diagnosis, staging, and prognostic evaluation of liver tumors in recent years. This article systematically reviewed the latest research progress of MRE in predicting microvascular invasion (MVI) of hepatocellular carcinoma (HCC), correlating tumor stiffness with biological behavior, and monitoring therapeutic responses. Studies indicate that MRE can effectively differentiate benign from malignant tumors and predict the risk of MVI in HCC by quantifying the shear modulus of tumors and surrounding liver tissues. Furthermore, the combination of MRE with multimodal imaging techniques (e.g., diffusion-weighted imaging, dynamic contrast-enhanced magnetic resonance imaging) enhances diagnostic performance. Although challenges such as high costs and insufficient standardization persist in clinical applications, the potential of MRE in personalized treatment decision-making is widely acknowledged. Future research should focus on technological optimization, multicenter validation, and artificial intelligence-assisted analysis.
Objective:To explore the association between magnetic resonance imaging(MRI)features and the 21-gene recurrence score(RS),and to establish RS prediction models.Methods:Clinical and imaging data of estrogen receptor(ER)+/human epidermal growth factor receptor 2(HER2)-breast cancer patients who underwent 21-gene expression assay in Fudan University Shanghai Cancer Center from April 2017 to March 2019 were collected.The patients who underwent preoperative breast MRI were selected.MRI images were evaluated according to the 2013 version of Breast Imaging Reporting and Data System lexicon.The univariate analyses were used to compare differences in MRI imaging features between the high-risk group(RS≥26)and the low-risk group(RS＜26)and multivariate logistic regression was used to construct a model.The patients were divided into the training group and the validation group in a 7∶3 ratio.Pearson correlation coefficient screening method and recursive feature elimination method were used for feature screening,the synthetic minority oversampling technique was used for balancing the training dataset,four different machine learning algorithms(linear support vector machine,random forest,decision tree and K-nearest neighbor)were used to construct the models,and the model performance was evaluated by receiver operating characteristic(ROC)curve.Results:A total of 159 patients were enrolled,with 58 in the low-risk group and 101 in the high-risk group.In clinical characteristics,progesterone receptor(PR)status showed difference(P=0.017),and the proportion of patients with positive PR expression was higher in the low-risk group than in the high-risk group.In the MRI characteristics,the distribution of tumor margins showed difference between groups(P=0.008).The tumors in the low-risk group were mostly characterized by spiculated margins(64.8%),and the tumors in the high-risk group were mostly characterized by irregular margins(54.7%).Incorporating PR status and tumor margin into the multivariate logistic regression model,PR positive patients had a lower risk of recurrence than PR negative patients,with an OR value of 0.110(P=0.038);the recurrence risk of spiculated margins was relatively lower than that of irregular margins,with an OR value of 0.343(P=0.004).The area under curve(AUC)was 0.667.The calibration curve and decision curve indicated that the model had good calibration performance and certain clinical practicability.111 patients were included in the training group(34 in low-risk group and 77 in high-risk group)and 48 patients were included in the validation group(24 in low-risk group and 24 in high-risk group).The AUC range of the four machine learning models was 0.64-0.69,and the AUCs of support vector machine and random forest models were relatively higher.Conclusion:Breast MRI features have a potential role in assessing the recurrence risk of ER+/HER-breast cancer patients.
肺癌是中国常见的恶性肿瘤,其发病率及死亡率均居癌症首位.由于早期症状不典型且缺乏有效的筛查手段,大多数肺癌患者发现时已处于晚期,预后较差.改善预后的关键在于早发现、早诊断、早治疗.随着人工智能和医疗大数据分析等前沿技术的不断进步,深度学习作为目前人工智能发展迅猛的一大分支,被认为是医学图像分析领域的宝贵工具,现已在早期肺癌的筛查、诊疗和预后评估以及晚期肺癌的随访中广泛应用,并取得诸多成果.本文阐述了近年来深度学习在肺癌18F-FDG 正电子发射体层成像(positron emission tomography,PET)/计算机体层成像(computed tomography,CT)诊疗中的应用现状及研究进展,主要介绍图像采集与重建、病灶检测与分割、诊断与鉴别诊断、基因突变状态与免疫治疗靶点及治疗反应与结局预测等,并对其发展前景及面临的挑战予以梳理.
目的:基于乳腺X线摄影钙化的形态及分布,探讨可能预测乳腺病灶浸润性的钙化特征.方法:回顾并分析乳腺X线摄影检出可疑钙化并经病理学检查证实的女性患者267例,患者平均年龄(48.1±9.7)岁.根据病理学检查结果,将浸润性癌及导管原位癌微浸润定义为浸润性病灶,将导管原位癌及各种良性病灶定义为非浸润性病灶.探讨可能预测病灶浸润性的钙化形态及分布状态:单因素分析采用χ2检验,多因素分析采用logistic回归分析以建立预测模型;绘制受试者工作特征(receiver operating characteristic,ROC)曲线并计算曲线下面积(area under curve,AUC)以评估模型预测能力;采用Hosmer-Lemeshow拟合优度检验评估模型整体拟合度.结果:单因素分析显示,伴点状/圆形钙化的可疑钙化及无定形钙化对乳腺非浸润性病灶具有预测价值(χ2=10.567,P=0.001;χ2=31.153,P＜0.001);而细线样或细分枝状钙化及段样分布的可疑钙化对乳腺浸润性病灶具有预测价值(χ2=36.275,P＜0.001;χ2=5.147,P=0.023).多因素logistic回归分析显示:无定形钙化是乳腺非浸润性病灶的独立预测因素(OR=0.273,95%CI 0.135～0.553,P＜0.001);而细线样或细分枝状钙化是乳腺浸润性病灶的独立预测因素(OR=5.211,95%CI 1.819～14.931,P=0.002).预测模型的AUC为0.759(95%CI 0.686～0.833),且Hosmer-Lemeshow检验显示该模型具有较好的拟合度(P＞0.05).结论:依据乳腺X线摄影钙化的形态及分布预测乳腺病灶的浸润性具有一定的可行性,可根据乳腺X线摄影筛查尽早地对不同性质的病灶采取相应的干预或管理措施.
Objective: To analyze the characteristics of two-dimensional ultrasound and contrast-enhanced ultrasound (CEUS) of retroperitoneal ganglioneuroma (RGN) for improving its preoperative diagnostic accuracy. Methods: The ultrasonic images of 12 patients with RGN confirmed pathologically in Zhongshan Hospital, Fudan University from January 2012 to December 2022 were collected for analysis. Eight patients underwent CEUS examination. The enhancement method, enhancement intensity and performance characteristics in different enhancement phases were observed. Results: Twelve patients had a total of 12 lesions, all of which were single. Twelve lesions were hypoechoic and hyperechoic capsules were seen around them. The internal echo of 8 lesions was uneven and 4 lesion was uniform. Seven lesions were accompanied by punctate calcification. Eight lesions were irregular in shape and 4 lesions were regular in oval shape. Four lesions had clear borders and 8 lesion had unclear border. Color Doppler flow imaging (CDFI) detected a small amount of punctate color blood flow inside 7 lesions. In the arterial phase, 5 lesions showed mild overall uniform enhancement, 1 had mild overall uneven enhancement, and 2 lesions showed mild peripheral uneven enhancement on CEUS. All 8 lesions gradually decreased and showed hypoechoic in the venous phase and the delayed phase. Conclusion: RGN is mostly manifested as an inhomogeneous low-echo tumor with a high-echo capsule on two-dimensional ultrasound. Spotty calcification has certain characteristics, with irregular morphology, and is more suggestive when it appears in a “tear-drop” shape. Although the lesion is large, there is little blood flow signal detected by CDFI. On CEUS, the lesion often shows mild enhancement, with slow regression in the venous and delayed phases.
