The aim of this study was to develop an artificial intelligence model to automatically differentiate between non-neoplastic and neoplastic gallbladder polyps, while also distinguishing benign from malignant polyps. Patients with gallbladder polyps who underwent cholecystectomy from January 2022 to June 2023 were recruited from two hospitals retrospectively. Conventional ultrasound findings and clinical characteristics of patients before cholecystectomy were acquired. Ultrasound image blocks of gallbladder lesions were automatically segmented by the Unet network for diagnosis. A fusion deep learning model based on dual-mode ultrasound (grey-scale ultrasound and colour Doppler flow imaging) was established to diagnose gallbladder polyps and validated in the validation and test set. Finally, we compared the diagnostic efficiency of the model with that of radiologists and guidelines. A total of 339 patients (mean ages 53.17 ± 15.89, 182 females) were enroled in this study. The Dice coefficient and intersection over union (IoU) value of the automatic segmentation based on the Unet-efficientnet-b4 network were 0.912 and 0.838. In differentiating non-neoplastic from neoplastic polyps, the integrative deep learning (IDL) model showed area under the curves (AUCs) of 0.829 and 0.802 in validation and test sets, respectively. In differentiating benign and malignant polyps, the IDL model showed AUCs of 0.844 and 0.839 in validation and test sets, respectively. In the test set, the diagnostic performance of two junior radiologists was improved with the assistance of the IDL model. The IDL model based on dual-mode ultrasound could achieve accurate and automatic segmentation of gallbladder lesions, and showed excellent diagnostic performance for diagnosing gallbladder polyps. We developed a deep learning model based on conventional ultrasound that performs gallbladder segmentation while differentiating neoplastic from non-neoplastic polyps and benign from malignant polyps.

OBJECTIVE:To investigate whether incorporating contrast-enhanced ultrasound (CEUS) features improves the diagnostic performance of conventional ultrasound (US)-based malignant risk stratification model for gallbladder polyps (GBPs). METHODS:A total of 300 consecutive patients with GBPs who underwent surgery at two centers between June 2021 and January 2025 were prospectively enrolled. All underwent pre-operative US and CEUS. Two derivative models were developed based on the conventional US model: a simplified CEUS model (by revising scores of US-derived parameters via CEUS) and a multiparametric CEUS model (by incorporating CEUS perfusion characteristics). Surgical pathology served as the gold standard for comparing the three models' diagnostic efficacy. RESULTS:Eight CEUS indicators demonstrated excellent inter-observer consistency (kappa > 0.75). The simplified CEUS model revised scores for polyp shape, size and gallbladder wall integrity, whereas the multiparametric CEUS model incorporated six additional CEUS indicators: pedicle, base width, vascular morphology, gallbladder wall thickening at the polyp attachment site, arterial-phase hyperenhancement of the gallbladder wall at the attachment site and peak enhancement pattern. In the validation set, the multiparametric CEUS model outperformed the simplified CEUS and conventional US models. For non-neoplastic versus neoplastic polyps, their area under the receiver operating characteristic curve values were 0.820 (US), 0.900 (simplified CEUS) and 0.962 (multiparametric CEUS); for benign versus malignant polyps, the values were 0.893 (US), 0.974 (simplified CEUS) and 0.983 (multiparametric CEUS), respectively. CONCLUSION:Incorporation of CEUS features significantly improves the diagnostic efficacy of the malignant risk stratification model for GBPs, a tool with high clinical value that merits widespread clinical application.

Gallbladder cancer (GBC) represents one of the most prevalent malignancies within the biliary system. In the management of GBC, accurate assessment of tumor invasion extent through histopathology is crucial for staging and treatment strategy formulation. The preoperative ultrasound prediction of histopathological local invasion in GBC holds significant clinical value, but it suffers from the limitation that the human vision is unable to directly decipher high-throughput information from ultrasound images. To this end, we propose M3-Net, a Mamba-empowered Multi-task and Multi-phase network, which represents a non-invasive and efficient approach based on deep learning. This network ingeniously integrates contrast-enhanced ultrasound images from multiple phases, including early arterial phase, peak time, and venous phase, leveraging their dynamic information. Through its unique multi-task learning architecture, our M3-Net is capable of efficiently handling three closely related histopathological prediction tasks simultaneously: muscle layer or perimuscular connective tissue invasion, serosal invasion, and liver invasion. Our M3-Net also employs a tumor-aware highlighting approach to strengthen the features of tumor lesions while mitigate the interference from non-lesion areas. Experimental results demonstrate that our M3-Net outperforms traditional single-task, single-phase models across various performance metrics, achieving areas under the receiver operating characteristic curves (ROC-AUCs) as high as 0.883, 0.936, and 0.898 on these three key histopathological tasks, respectively. Our M3-Net holds promise in providing a novel potent tool for precise diagnosis of GBC, further propelling innovations and advancements in clinical practice for malignancies of the biliary system.

As a rare biliary tract tumor, intraductal papillary neoplasm of the bile duct (IPNB) is most common in elderly men and can progress to cholangiocarcinoma- (CCa) if left untreated. It is reported that IPNB usually communicates with the bile duct. As a result, the downstream bile ducts are imaged asymmetrically dilated. However, a case of IPNB that we report here is different. Enhanced MRI revealed a lack of connectivity with the bile duct in this case. Based on this, the purpose of this case study is to suggest that the majority of imaging doctors should widely understand the various imaging manifestations of the disease to avoid misdiagnosis. In addition, although this feature was not indicated by ultrasound in this case, given previous studies and considering the convenience and non-ionizing radiation damage of CEUS, we recommend its use as a screening method for IPNB to improve diagnostic accuracy.

PURPOSE:We aimed to develop a 4-level risk stratification model using a scoring system based on conventional ultrasound to improve the diagnosis of gallbladder polyp. METHOD:Patients with histopathologically confirmed gallbladder polyps were consecutively recruited from three medical centres. Conventional ultrasound findings and clinical characteristics were acquired prior to cholecystectomy. Risk factors for neoplastic and malignant polyps were used to build a risk stratification system via interobserver agreement and multivariate logistic regression analysis. The model was retrospectively trained using 264 pre-surgical samples and prospectively validated using 106 pre-surgical samples. Model performance was evaluated using the area under the receiver operating characteristic curve (AUC) and malignant polyp rate. RESULTS:In total, 370 patients (mean age, 51.68 ± 14.41 years, 156 men) were enrolled in this study. Size (≥12 mm), shape (oblate or round), single, vascularity, gallbladder stone or sludge were considered risk factors for neoplastic polyps. Size (≥14 mm), shape (oblate), single, disrupted gallbladder wall, and gallbladder stone or sludge were risk factors for malignant polyps (all p < 0.05). In the scoring system, the sensitivity, specificity, and AUC of score ≥ 9 in diagnosing neoplastic polyps were 0.766, 0.788, and 0.876 respectively; and the sensitivity, specificity, and AUC of score ≥ 15 in diagnosing malignant polyps were 0.844, 0.926, and 0.949 respectively. In our model, the malignancy rates at the four levels were 0 % (0/24), 1.28 % (2/156), 9.26 % (5/54), and 70.37 % (38/54), respectively. CONCLUSIONS:The 4-level risk stratification model based on conventional ultrasound imaging showed excellent performance in classifying gallbladder polyps.

目的:探讨胆囊占位性病变超微血管成像(superb microvascular imaging,SMI)的血流特征及其在良恶性鉴别诊断中的应用价值.方法:选取71例胆囊占位性病变患者,所有患者术前均行彩色多普勒血流显像(color Doppler flow imaging,CDFI),并用SMI技术检测血流.采用Adler分级标准,比较CDFI与SMI在胆囊占位性病变中血流分级上的差异.根据血流形态,分析SMI在胆囊占位性病变不同病理学类型中的血流形态差异.结果:71例胆囊占位性病变患者中SMI检出血流56例(78.87％),明显优于CDFI[32例(45.07％)](χ2=17.21,P<0.05).SMI在胆囊良性占位性病变与恶性占位性病变的Ⅱ、Ⅲ级血流显示均优于CDFI(χ2=5.430,P<0.05).SMI显示,胆固醇性息肉最常见血流形态为点状(13/25,52.00％),胆囊腺肌症也以点状为主(7/15,46.67％),胆囊腺瘤主要表现为单支状或分支状(9/12,75.00％),胆囊癌以凌乱型为主(8/19,42.11％).胆囊占位性病变不同病理学类型呈现不同的SMI血流形态(χ2=20.16,P<0.05).结论:与CDFI相比,SMI能更清晰地显示胆囊占位性病变的血管分布特征,可为胆囊占位性病变的良恶性鉴别诊断提供新的参考依据.

目的:探讨国产与进口超声设备高端机型腹部图像质量的差距.方法:选取全国8个省及直辖市18家不同级别医院的受试者955例,每位受试者由同一名医师采用某一国产与某一进口超声设备各采集标准切面26张.根据国家及地区相关法规与行业标准,组织专家制定超声腹部图像质量评价标准,对采集到的超声图像数据进行李克特量表评分,5分为最佳,1分为最差.采用R语言进行数据分析,比较5个国产(Ⅰ、Ⅱ、Ⅲ、Ⅳ、Ⅴ)及6个进口超声设备高端机型(A、B、C、D、E、F)的腹部图像质量.结果:共采集腹部超声图像49660张.国产和进口超声设备腹部图像质量评分总体接近,差异无统计学意义.国产型号Ⅰ、Ⅳ表现相对较好,与评分最高的进口设备B之间无差异.国产型号Ⅱ及Ⅲ灰阶细节分辨力、彩色多普勒血流成像充盈度与平滑度、时间流速曲线中的低速血流敏感度较差.国产与进口超声设备腹部图像总体灰阶、彩色多普勒血流成像、时间流速曲线指标评分之间差异无统计学意义.结论:国产超声设备高端机型腹部图像质量较好,总体与进口设备之间无差异,得到超声评价专家的普遍认可.

超声检查是胆囊息肉样病变首选的影像学检查方法,但仅根据常规超声对病变回声、形态、血流的检测结果来区分真假性息肉是不可靠的,难以满足快速准确的术前鉴别需求.基于计算机辅助分析,对经病理证实的胆囊腺瘤真性息肉31例31个病灶和胆囊胆固醇假性息肉37例38个病灶进行研究.从术前超声图像中提取病灶的空域和形态特征,接着对特征进行统计分析并用于支持向量机分类,以鉴别胆囊腺瘤及胆囊胆固醇性息肉.结果表明,胆囊腺瘤组的均一度显著高于胆固醇息肉组(p＜0.001),胆囊腺瘤组的像素分布较胆固醇息肉更均匀.相较于空域或形态特征对应的分类模型,所有特征对应的集成模型性能提升.分类准确率、敏感性、特异性分别达到90.5％、91.0％、90.0％,曲线下面积为0.927.因此,计算机辅助分析超声图像有助于提高胆囊真性和假性息肉的诊断准确性.

Purpose: To explore the value of ultrasound radiomics in the preoperative identification of true and pseudo gallbladder polyps and to evaluate the associated diagnostic accuracy.Methods: Totally, 99 pathologically proven gallbladder polyps in 96 patients were enrolled, including 58 cholesterol polyps (55 patients) and 41 gallbladder tubular adenomas (41 patients). Features on preoperative ultrasound images, including spatial and morphological features, were acquired for each lesion. Following this, two-stage feature selection was adopted using Fisher's inter-intraclass variance ratios and Z-scores for the selection of intrinsic features important for differential diagnosis achievement with support vector machine use.Results: Eighty radiomic features were extracted from each polyp. Eight intrinsic features were identified after two-stage selection. The contrast 14 (Cont14) and entropy 6 (Entr6) values in the cholesterol polyp group were significantly higher than those in the gallbladder adenoma group (4.063 ± 1.682 vs. 2.715 ± 1.867, p < 0.001 for Cont14; 4.712 ± 0.427 vs. 4.380 ± 0.720, p = 0.003 for Entr6); however, the homogeneity 13 (Homo13) and energy 8 (Ener8) values in the cholesterol polyp group were significantly lower (0.500 ± 0.069 vs. 0.572 ± 0.057, p < 0.001 for Homo13; 0.050 ± 0.023 vs. 0.068 ± 0.038, p = 0.002 for Ener8). These results indicate that the pixel distribution of cholesterol polyps was more uneven than that of gallbladder tubular adenomas. The dispersion degree was also significantly lower in the cholesterol polyp group than the gallbladder adenoma group (0.579 ± 0.054 vs. 0.608 ± 0.041, p = 0.005), indicating a lower dispersion of high-intensity areas in the cholesterol polyps. The long axis length of the fitting ellipse (Maj.Len), diameter of a circle equal to the lesion area (Eq.Dia) and perimeter (Per) values in the cholesterol polyp group were significantly lower than those in the gallbladder adenoma group (0.971 ± 0.485 vs. 1.738 ± 0.912, p < 0.001 for Maj.Len; 0.818 ± 0.393 vs. 1.438 ± 0.650, p < 0.001 for Eq.Dia; 2.637 ± 1.281 vs. 5.033 ± 2.353, p < 0.001 for Per), demonstrating that the cholesterol polyps were smaller and more regular in terms of morphology. The classification accuracy, sensitivity, specificity, and area under the curve values were 0.875, 0.885, 0.857, and 0.898, respectively.Conclusions: Ultrasound radiomic analysis based on the spatial and morphological features extracted from ultrasound images effectively contributed to the preoperative diagnosis of true and pseudo gallbladder polyps and may be valuable in their clinical management.

目的 为比较国产与进口医用超声设备浅表基础图像的质量,本课题进行了全国多区域医院的医用超声成像系统的综合评价.方法 本研究分析包括上海市在内的全国5家医院209例受试者的浅表超声基础图像资料.受训者对同一名受试者的浅表同时行国产和进口设备的检查,检查完成后图像上传至数据系统.结果 采用李克特评分,评分采用Wilcoxon秩和检验进行统计学分析.结果 比较国产与进口设备的浅表基础图像质量,大部分切面中二者评分的差异无统计学意义.部分国产品牌基础图像的评分较高,部分切面的评分高于进口设备,主要表现在彩色多普勒功能、细节分辨力方面,差异有统计学意义.结论 通过国产与进口医用超声设备的比较研究,可客观了解目前国产医用超声设备浅表基础图像质量的优劣,为国家提供国产医用超声设备发展战略的依据和支撑.

目的 探讨黄色肉芽肿性胆囊炎(XGC)临床特点及影像学表现,以及不同影像学检查技术对XGC的诊断能力.方法 回顾分析经手术病理确诊的33例XGC及51例非XGC胆囊炎病例的临床及术前影像学检查资料.结果 XGC与非XGC型胆囊炎相比,出现腹痛、血清CA199升高的比例更高,胆囊壁病变更厚(P＜0.05)、更多侵犯肝、胃肠等周围器官(P＜0.05),而二者在壁增厚类型、有无发热或白细胞升高方面无统计学差异(P＞0.05).超声造影(CEUS)、增强CT、增强MRI及PET-CT对XGC的诊断能力相仿(P＞0.05).结论 XGC与非XGC型胆囊炎相比,在临床症状、肿瘤标志物以及影像学表现上具有不同的特征,超声造影、增强CT、增强MRI及PET-CT诊断XGC的准确率相近.

Objective: To investigate the grayscale ultrasound, color Doppler ultrasound and elastography features of encapsu-lated papillary carcinoma of the breast (EPC). Methods: The sonographic features of twenty-six cases of EPC proved by pathology were retrospectively analyzed. Results: Twenty-two lesions were cystic-solid mixed masses. Five were predominantly cystic com-plex masses, and ten were complex masses with almost equivalent cystic and solid components, and seven were predominantly solid complex masses. Four were solid masses. Fourteen lesions were isoechoic, and nine were slightly hypoechoic, and three were hy-poechoic. Twenty lesions were homogenous, and six were heterogeneous. Twenty-two lesions were regular, and four were irregular. Microcalcifications were found in two lesions, and macrocalcification was found in one lesion. Internal echoes were found in thirteen cystic-solid mixed lesions. Hypervascularity, isovascularity, and hypovascularity were found in 17, 7, and 2 masses respectively. Elastography was performed in 17 masses and the elasticity score was 2 in 7 masses, 3 in 3 masses, and 4 in 7 masses. Red yellow stripes were found in four lesions with elasticity score 2. Twenty-three masses were diagnosed as probably or suspiciously malignant. Conclusion: The sonographic appearances of EPC were various. Conventional ultrasonography and elastography are useful for pre-dicting the malignancy of EPC.

目的:比较声触诊组织量化(virtual touch tissues quantification,VTQ)技术与灰阶超声对肝硬化的诊断效能.方法:102例肝穿刺活检或肝手术前后1周的患者接受灰阶超声检查,对肝包膜、肝实质回声、脾脏大小进行评分;同时采用ACUSONS2000超声仪对患者进行VTQ测量.以病理结果作为金标准,分别绘制受试者工作特征曲线(receiver operatingcharacteristic curve,ROC),比较两种方法对肝硬化的诊断效能.结果:病理学检查结果显示,102例患者中S0期7例、S1期9例、S2期13例、S3期21例、S4期52例.VTQ及灰阶超声诊断肝硬化曲线下面积分别为0.876、0.830,差异有统计学意义(P=0.000).结论:VTQ诊断肝硬化的效能优于灰阶超声,对临床治疗的指导价值可能更大.

目的:探讨乳腺实性乳头状癌(solid papillary carcinoma,SPC)灰阶超声、彩色多普勒超声和超声弹性成像的图像特点及诊断价值.方法:分析18例经病理证实的SPC患者的超声图像资料.结果:病灶呈单发、多发导管内病变或实性病变.17个肿块呈低回声或中等回声,1个呈囊实性.10个肿块内见细小钙化.7个肿块后方回声增强,3个后方回声衰减,8个后方回声不变.12个肿块血流丰富,2个血流中等,4个无血流或少量血流.7例原位癌或微小浸润性癌弹性评分为1～2分,3例浸润性癌弹性评分为4分.4例单发导管内原位癌超声诊断为导管内乳头状瘤;14例多发导管内病灶及单发实性肿块中,超声诊断为恶性肿瘤11例、导管内乳头状瘤3例.结论:SPC有多种超声表现,超声有助于乳腺多发导管内病灶及实性病灶的术前良恶性诊断,灰阶超声与超声弹性成像联合应用有助于提示SPC的浸润程度.

Objective To evaluate the value of S-Detect technology of ultrasonography in the diagnosis of thyroid tumors. Methods Ninety-three thyroid tumors in 93 patients were enrolled in the group.A varied image features of the thyroid masses in gray-scale ultrasonography were analyzed by S-Detect technology and experienced doctor separately.The results were compared and the diagnostic ability were also compared between the two methods. Results There were 44 malignant tumors and 49 benign tumors in these thyroid nodules.The sensitivity of S-Detect technology in the diagnosis of thyroid malignant tumors was higher,up to 88.7%. In the five image features of the thyroid tumors in gray-scale ultrasonography,the result of composition of the mass obtained by S-Detect was the most consistent with that of the doctors and Kappa value was 0.89.Conclusions S-Detect is a kind of computer-aided diagnosis system,which is suitable for the ultrasound beginners in the diagnosis of thyroid tumors.

目的:探讨乳腺富细胞黏液腺癌的超声表现与病理学诊断的关系.方法:分析11例乳腺富细胞黏液腺癌的超声表现及病理学检查结果,包括细胞内有无黏液、分布形态(多发细小囊泡、较大黏液湖、较大黏液湖伴印戒细胞)及细胞外黏液与细胞面积比值(<30％、30％～90％、>90％),并对照研究超声内部回声与病理学检查结果之间的关系.结果:超声图像上肿块均呈实性团块,5个肿块为高回声,2个为等回声,4个为低回声;8个形态不规则;8个边界模糊;10个见中等或丰富彩色血流信号.8例行超声弹性成像检查,5个弹性超声评分为2～3分.病灶细胞内黏液为多发细小囊泡时,超声均表现为高回声;病灶细胞内黏液为黏液湖或无细胞内黏液时,超声表现为等回声或低回声.超声内部回声与细胞外黏液无明显关系.超声均诊断为BI-RADS 4类.结论:乳腺富细胞黏液腺癌常表现为形态不规则、边界模糊的高回声,血流丰富,弹性评分低.其内部回声与细胞内黏液有关,与细胞外黏液无明显关系.超声有助于乳腺富细胞黏液腺癌的术前良恶性诊断.

BACKGROUND: Focal gallbladder adenomyomatosis (ADM) is a common disease that mimics gallbladder cancer (GBC) on ultrasonography. OBJECTIVE: Here we aim to assess the value of contrast-enhanced ultrasound (CEUS) in differentiating ADM from GBC. METHODS: Forty-one histopathologically proven focal ADMs and 34 GBCs (<= T2 stage) were enrolled in the study. Lesion location, blood flow signals, contrast pattern and appearance on contrast-enhanced ultrasound (CEUS) were compared respectively. RESULTS: Lesions were detected in fundus, body, neck at the rates of 61.0% (25/41), 26.8% (11/41) and 12.2% (5/41), respectively, in ADM patients, in comparison to 29.4% (10/34), 32.4% (11/34) and 38.2% (13/34), respectively, in GBC patients (p = 0.009). Blood flow signals were detected in 19.5% (8/41) of cases in ADMs, compared to 58.8% (20/34) in GBCs (p = 0.001). On CEUS, iso-enhancement, hypo-enhancement, intramural anechoic space and intactness of GB wall were detected in 41.5% (17/41), 39.0% (16/41), 56.1% (23/41) and 80.5% (33/41) cases of ADMs, in contrast to 17.6% (6/34), 20.6% (7/34), 20.6% (7/34) and 17.6% (6/34) of GBCs (p =0.001, p = 0.001, p = 0.002, p < 0.001, respectively). The prior Youden's index were 0.81 based on intactness of GB wall on CEUS. CONCLUSION: Combined with CEUS helps improve the differential diagnosis accuracy of focal gallbladder ADMs.

To assess conventional ultrasound (CUS) combined with contrast-enhanced ultrasonography (CEUS) for the differential diagnosis of gallbladder adenomyomatosis (ADM) and early stage gallbladder cancer (GBC). CUS and CEUS data of 21 histopathologically proven ADMs were analyzed and compared with 34 GBCs (≤T2 stage) retrospectively. Significant differences were found in lesion location, intramural anechoic space, echogenic foci, intactness of GB wall on CUS, and enhancement pattern, intramural anechoic space, intactness of GB wall on CEUS. The fundal proportion was 57.1% (12/21) in ADMs and 11.8% (4/34) in GBC (p=0.001). Intramural anechoic space, echogenic foci, intactness of GB wall on CUS and linear or branch blood flow were detected in 42.9% (9/21), 52.4% (11/21), 76.2% (16/21) and 28.6% (6/21) of cases in ADM group, compared to 5.9% (2/34), 11.8% (4/34), 26.5% (9/34) and 73.5% (25/34) in GBC group (p=0.001, p=0.002, p=0.001, p=0.001), respectively. Hypo-enhancement, intramural anechoic space and intactness of GB wall on CEUS were detected in 76.2% (16/21), 71.4% (15/21) and 85.7% (18/21) cases of ADMs, whereas 11.8% (4/34), 5.9% (2/34) and 17.6% (6/34) of GBCs (p<0.001, p=0.001, p=0.001). The prior Youden's index were 0.72, 0.66 and 0.65 based on intramural anechoic space on CEUS, hypo-enhancement and intactness of GB wall on CEUS, respectively. Intramural anechoic space and intactness of GB wall were more detected in ADMs when combined with CEUS which helped to improve the diagnosis of ADMs. Hypo-enhancement, intramural anechoic space on CEUS, intactness of GB wall on CEUS, location in GB fundus, intramural echogenic foci, lack of blood flow on CUS are useful markers for diagnosis of gallbladder ADMs.