BackgroundRecurrent pregnancy loss (RPL) frequently links to a prolonged endometrial receptivity (ER) window, leading to the implantation of non-viable embryos. Existing ER assessment methods face challenges in reliability and invasiveness. Radiomics in medical imaging offers a non-invasive solution for ER analysis, but complex, non-linear radiomic-ER relationships in RPL require advanced analysis. Machine learning (ML) provides precision for interpreting these datasets, although research in integrating radiomics with ML for ER evaluation in RPL is limited.ObjectiveTo develop and validate an ML model that employs radiomic features derived from multimodal transvaginal ultrasound images, focusing on improving ER evaluation in RPL.MethodsThis retrospective, controlled study analyzed data from 346 unexplained RPL patients and 369 controls. The participants were divided into training and testing cohorts for model development and accuracy validation, respectively. Radiomic features derived from grayscale (GS) and shear wave elastography (SWE) images, obtained during the window of implantation, underwent a comprehensive five-step selection process. Five ML classifiers, each trained on either radiomic, clinical, or combined datasets, were trained for RPL risk stratification. The model demonstrating the highest performance in identifying RPL patients was selected for further validation using the testing cohort. The interpretability of this optimal model was augmented by applying Shapley additive explanations (SHAP) analysis.ResultsAnalysis of the training cohort (242 RPL, 258 controls) identified nine key radiomic features associated with RPL risk. The extreme gradient boosting (XGBoost) model, combining radiomic and clinical data, demonstrated superior discriminatory ability. This was evidenced by its area under the curve (AUC) score of 0.871, outperforming other ML classifiers. Validation in the testing cohort of 215 subjects (104 RPL, 111 controls) confirmed its accuracy (AUC: 0.844) and consistency. SHAP analysis identified four endometrial SWE features and two GS features, along with clinical variables like age, SAPI, and VI, as key determinants in RPL risk stratification.ConclusionIntegrating ML with radiomics from multimodal endometrial ultrasound during the WOI effectively identifies RPL patients. The XGBoost model, merging radiomic and clinical data, offers a non-invasive, accurate method for RPL management, significantly enhancing diagnosis and treatment.

BackgroundRecurrent pregnancy loss (RPL) poses significant challenges in clinical management due to an unclear etiology in over half the cases. Traditional screening methods, including ultrasonographic evaluation of endometrial receptivity (ER), have been debated for their efficacy in identifying high-risk individuals. Despite the potential of artificial intelligence, notably deep learning (DL), to enhance medical imaging analysis, its application in ER assessment for RPL risk stratification remains underexplored.ObjectiveThis study aims to leverage DL techniques in the analysis of routine clinical and ultrasound examination data to refine ER assessment within RPL management.MethodsEmploying a retrospective, controlled design, this study included 346 individuals with unexplained RPL and 369 controls to assess ER. Participants were allocated into training (n = 485) and testing (n = 230) datasets for model construction and performance evaluation, respectively. DL techniques were applied to analyze conventional grayscale ultrasound images and clinical data, utilizing a pre-trained ResNet-50 model for imaging analysis and TabNet for tabular data interpretation. The model outputs were calibrated to generate probabilistic scores, representing the risk of RPL. Both comparative analyses and ablation studies were performed using ResNet-50, TabNet, and a combined fusion model. These were evaluated against other state-of-the-art DL and machine learning (ML) models, with the results validated against the testing dataset.ResultsThe comparative analysis demonstrated that the ResNet-50 model outperformed other DL architectures, achieving the highest accuracy and the lowest Brier score. Similarly, the TabNet model exceeded the performance of traditional ML models. Ablation studies demonstrated that the fusion model, which integrates both data modalities and is presented through a nomogram, provided the most accurate predictions, with an area under the curve of 0.853. The radiological DL model made a more significant contribution to the overall performance of the fusion model, underscoring its superior predictive capability.ConclusionThis investigation demonstrates the superiority of a DL-enhanced fusion model that integrates routine ultrasound and clinical data for accurate stratification of RPL risk, offering significant advancements over traditional methods.

Fetal limb deformity is a common congenital malformation in clinic, and prenatal ultrasound is one of the most commonly used methods in clinical diagnosis of fetal limb deformities. At present, few studies have used target detection to find deformity in ultrasound images of fetal limbs. In this study, based on deep learning technology, we propose a framework that integrates Swin-Transformer and YOLOv5 to detect limb deformities in fetal ultrasound images. Swin-Transformer makes up for CNN’s incapability of capturing global features, effectively extracts the features of the ultrasound images, which is conducive to the detection of fetal limb deformities. Experiments show that the YOLOv5 model with Swin-Transformer as the backbone has an average detection accuracy of 92.10

1 病例简介 女,28岁,孕1产0.主诉:停经23+3周,外院产前超声提示胎儿姿势异常.外院产检,唐氏筛查示:21-三体综合征低风险;无创DNA示:低风险;早孕期胎儿颈项透明层厚度正常.超声检查提示:胎儿小于孕周,双足、双手姿势异常;小下颌.孕妇选择终止妊娠,引产后临床表现与产前超声检查一致.

本文考察Finkler超声评分联合狐猴酪氨酸激酶2(LMTK2)基因在上皮性卵巢癌(EOC)诊断中的价值.本研究对112例EOC患者(EOC组)和76例卵巢良性上皮性肿瘤患者(良性组)进行超声检查及Finkler超声评分,并采用RT-PCR法检测血清LMTK2水平.结果显示,EOC组的Finkler超声评分和血清LMTK2 mRNA相对表达量显著高于良性组(P＜0.001).Finkler超声评分和血清LMTK2 mRNA相对表达量显著正相关(r=0.563,P＜0.001).当Finkler超声评分＞5时,诊断EOC的曲线下面积(AUC)为0.964.当血清LMTK2 mRNA相对表达量＞0.415时,诊断EOC的AUC为0.907.Finkler超声评分联合血清LMTK2诊断EOC的AUC(0.986)和敏感性(94.64％)高于单独诊断.Ⅲ～Ⅳ期组的Finkler超声评分和血清LMTK2 mRNA相对表达量显著高于Ⅰ～Ⅱ期组(P＜0.05).Finkler超声评分、血清LMTK2及二者联合诊断Ⅲ～Ⅳ期的AUC分别为0.643、0.850、0.851.本研究表明,在EOC的早期诊断中,与单独诊断相比,Finkler超声评分联合血清LMTK2可有效提高诊断灵敏度,并且血清LMTK2对病理分期的诊断价值可能高于Finkler超声评分.

目的:探讨基于超声造影参数联合常见影响因素建立列线图预测前列腺癌(PCa)治疗后生化复发的价值.方法:选择PCa患者149例,治疗前均接受前列腺特异性抗原(PSA)水平的测定、超声造影以及穿刺活检病理分级,治疗后进行5年随访.随访期间出现生化复发的PCa患者纳入复发组,其余患者纳入未复发组.比较复发组与未复发组的临床指标及超声造影参数的差异.采用多因素COX回归筛选影响PCa患者生化复发的独立影响因素,并建立列线图预测PCa患者生化复发的概率.结果:149例患者中,失访8例,余141例患者治疗后5年内生化复发率为29.1％,41例纳入复发组,100例纳入未复发组.多因素COX回归分析结果显示淋巴结转移(P=0.032)、治疗方式(P=0.035)、Gleason评分(P=0.031)、治疗前PSA(P=0.022)、峰值强度(PI) (P=0.048)及达峰时间(TTP)(P=0.041)是PCa治疗后生化复发的独立影响因子.基于上述因子建立的列线图具有良好的区分度(C-in-dex 0.935)和校准度(x2=9.514,P=0.301,Hosmer-Lemeshow检验),能够准确预测PCa患者生化复发的风险.结论:基于超声造影结合常见影响因素建立的列线图可以精确预测PCa患者治疗后生化复发的概率,有助于临床及时采取预防措施以改善患者预后.

患者女,28岁,孕1产0.于孕24周来我院产检.专科检查:孕妇状态良好,胎心音正常,腹围宫高测值均符合孕周.系统超声检查:双顶径5.9 cm,头围21.2 cm,腹围19.4 cm,股骨径4.1 cm.左侧脐动脉腹内段纤细,内径约0.19 cm,其舒张期血流消失,阻力指数1.0;右侧脐动脉明显增粗,内径约0.35 cm,阻力指数0.62,收缩期最大流速与舒张末期流速比值:2.61.

目的 旨在探讨超声造影及增强CT检查对肝转移瘤的诊断价值.方法 选取我院2016年1月-2018年1月收治的肝转移瘤患者49例,患者均进行了增强CT及超声造影检测,收集患者一般资料、影像学资料等进行回顾性分析,观察肝转移瘤在增强CT及超声造影中的影像学表现特征,比较增强CT及超声造影对肝转移瘤的检出率及诊断符合率.结果 超声造影对肝转移瘤病灶总检出率为96.72％,增强CT对肝转移瘤病灶总检出率为95.08％,两者间比较数据无明显差异(P＞0.05);超声造影对肝转移瘤病灶诊断符合率为93.44％,与增强CT 95.08％比较无明显差异(P＞0.05);在超声增强中,61个肝转移瘤中31个呈现环状增强,12个团状增强,11个延迟增强,7个无明显增强;病灶直径＞3 cm的肝血管瘤增强型信号显著高于≤3cm病灶.CT增强扫描中,动脉期肝内见多发大小不等环形强化灶,瘤体密度与同层腹主动脉相近,静脉期病灶出现明显强化,延迟期则多为轻度持续环形强化.结论 超声造影及增强CT检查均能有效检出并诊断肝转移瘤,对肝转移瘤的典型图像特征显示佳,可为临床提供可靠影像学资料.

目的 探讨彩色多普勒超声、数字X线引导下下肢静脉造影检查在下肢深静脉瓣膜功能不全临床诊断中的价值.方法 选取德阳市人民医院自2014年1月至2016年5月临床诊断为大隐静脉曲张的53例患者(60条患肢)为研究对象.所有患者均接受彩色多普勒超声及数字X线引导下下肢静脉造影检查,客观分析并总结两种影像检查的图像特点;以手术情况为"金标准",评价彩色多普勒超声及数字X线引导下下肢静脉造影的临床诊断价值,包括灵敏度、特异度和诊断符合率.结果 数字X线引导下下肢静脉造影检查的灵敏度、特异度、诊断符合率均高于彩色多普勒超声,但差异无统计学意义(P>0.05).结论 彩色多普勒超声、数字X线引导下下肢静脉造影均具有定位精确、动态显示反流程度等优点,是评估下肢深静脉瓣膜功能不全的高敏感性、高准确性辅助检查方法,各有优劣.

Objective To investigate the value of color Doppler ultrasonography in the examination of perforating veins of the lower extremities by comparing with X-ray antegrade venography.Methods Fifty-three patients (60 lower extremities) clinically diagnosed with varicosis of the great saphenous vein from January 2014 to March 2016 were selected for color Doppler ultrasonography and X-ray antegrade venography.The results were compared and confirmed by surgery.Results Forty-seven lower extremities were diagnosed with perforating venous valve insufficiency by color Doppler ultrasonography,while 50 lower extremities by X-ray antegrade venography.There was no significant difference between the two methods (x2=0.800,P=0.371).A total of 78 perforating veins were detected by color Doppler ultrasonography,with 1-4 perforating veins per one lower extremity.Among them,66 perforating veins had an internal diameter of 2-5 mm,accounting for 84.6％.Except two perforating veins with an internal diameter ＜2 mm,the rest showed reflux.In addition,a total of 73 perforating veins presented distance to the pelma of 10-32 mm (foot-boot),accounting for 93.6％.Fifty-two lower extremities with varicosis of the great saphenous vein underwent laser ablation of the great saphenous vein with high ligation or + devascularization of perforating veins.Compared with operation findings,color Doppler ultrasonography in diagnosis of perforating venous valve insufficiency had 37 true positives,2 false positives,10 true negatives and 3 false negatives,and the sensitivity was 92.5％ (37/40),the specificity was 83.3％ (10/12),the accuracy rate was 90.4％ (47/52).Conclusions Color Doppler ultrasonography can accurately localize perforating veins,measure the internal diameter and observe the degree of reflux.It is of high sensitivity and accuracy in the diagnosis of perforating venous valve insufficiency.Moreover,it is safe,simple and economical,and suitable for preoperative and intraoperative diagnosis and localization.

目的:探讨乳管镜联合乳腺彩超诊断乳头溢液早期乳腺癌的临床效果.方法:我院乳腺外科2010年7月至2014年12月收治42例早期乳腺癌伴乳头溢液的病例,分析比较42例患者行乳腺彩超、乳管镜及乳腺彩超联合乳管镜检查的阳性率.结果:42例乳管溢液早期乳腺癌中,乳腺彩超阳性率47.6％,乳管镜检查阳性率66.7％,联合检查阳性率92.9％;乳管镜检查阳性率高于乳腺彩超阳性率,但差异无统计学意义(x2=2.227,P＞0.05);乳管镜检查阳性率低于联合检查阳性率,差异有统计学意义(x2=9.091,P＜0.01).结论:乳管镜联合乳腺彩超诊断乳头溢液早期乳腺癌更加准确可靠.

Objective Discuss about the ultrasonographic characteristics of acute appendicitis. Methods By means of analyzing the ultrasonography of 56 patients through the pathology proof. Results According to the ultrasonography, there are 48 patients with direct sign of acute appendicitis, account for 85.7%, 6 patients with indirect sign of acute appendicitis, account for 10.7% and 2 patients without abnormal echo, account for 3.6%.Conclusion With the image display, ultrasound can directly show the diagnosis results of acute appendicitis with low price, lit le trauma and high diagnostic accordance rate.

Objective To observe the clinical efficacy of using ultrasound-guided thrombin injection to treat false aneurysm Methods 11 false aneurysm were injected with thrombin under the guidance of ultrasound using 18 GPTC needle 200-500 U of thrombin was injected into each chamber.Results Ultrasound-guided thrombin injection was successful in 11 patients.The time of thrombosis was 30 s ～ 4 min.There was no patient with limb ischemia.Conclusion Ultrasound-guided thrombin injection to treat false aneurysm was simple,safe and effective.

Objective To evaluate the effect of preoperative ultrasonography on success rate of arteriovenous fistula after radio-cephalic end-to-lateral anastomosis operation. Methods Fifty-seven patients with uremia all accepted the operation of radio-cephalic end-to-lateral anastomosis on wrist, 30 cases of them were chosen randomly to receive ultrasound examination(examination group)and the other 27 cases received physical examination only(contral group). All cases received ultrasonography 1~6 months after the operation, the course, the internal diameter, and the patency of the blood vessels were observed. The success rate after operation was compared. Results The success rate of operation was higher in examination group(83.3%) than that in contral group(74.1%), there was significant difference between two groups (P < 0.05). Conclusion Ultrasonography can not only evaluate the blood vessels after operation, but also evaluate the condition of radial artery and cephalic vein before operation, determine the indications of operation and best operation position, decrease the blindness and failure rate of operation.

目的探讨食管贲门癌术后膈疝的临床特征、诊断治疗和预防方法。方法回顾971例食管贲门癌手术切除后发生膈疝11例的临床资料。结果本组发生于术后4~10 d 6例,术后10天~1个月3例,术后1个月~3.2年2例;发生于剧烈咳嗽后6例,大便便秘后2例,前列腺肥大小便后1例,未发现明显诱因2例;手术10例,致疝原因:1例为线结松脱,6例为缝线切割,3例为缝线针距间距过大;严密观察随访1例,全组死亡1例,膈疝发生率1.13%,病死率9.09%。结论食管癌贲门癌术后膈疝主要是由于手术操作技术不良所致,腹内压增高是重要的诱发因素。