RESUMEN
Objective:To explore the predictive value of ultrasound-based radiomics for liver metastasis in pancreatic neuroendocrine tumors (pNEN).Methods:A retrospective analysis was conducted on clinical, pathological, and ultrasound data of 269 pNEN patients confirmed by pathology at Tianjin Medical University Cancer Institute and Hospital from January 2012 to June 2022, including 94 patients with liver metastasis and 175 without liver metastasis. The regions of interest (ROI) were delineated on the maximum diameter section of the tumor using ITKSNAP software, and radiomics features were extracted using Pyradiomics. Radiomics features with an intra-group correlation coefficient greater than 0.90 were retained, and the optimal features were selected using the maximum relevance minimum redundancy (MRMR) algorithm. The dataset was randomly divided into a training set and a validation set in a ratio of 7∶3, and the random forest algorithm (Rfs) was used to predict pNEN liver metastasis. Three models were constructed, including the clinical ultrasound model, the radiomics model, and the comprehensive model that combined clinical ultrasound and radiomics features. The predictive performance of different models for pNEN liver metastasis was analyzed using the ROC curve, and the predictive performance of different models was compared using the Delong test.Results:A total of 874 features were extracted from the ROI, and 12 highly robust radiomics features were retained for model construction based on inter- and intra-observer correlation grading and feature selection. The area under curve(AUC), sensitivity, specificity, and accuracy of the radiomics model, the clinical ultrasound model, and the comprehensive model for predicting liver metastasis in pNEN patients were 0.800, 0.574, 0.789, 0.714; 0.780, 0.596, 0.874, 0.777; and 0.890, 0.694, 0.874, 0.810, respectively. The Delong test showed that the comprehensive model had the best predictive performance, with an AUC superior to that of radiomics model ( Z=3.845, P=0.000 12) and clinical ultrasound model ( Z=3.506, P=0.000 45). Conclusions:The radiomics model based on ultrasound has good performance in predicting liver metastasis in pNEN, and the comprehensive model that combines clinical ultrasound and radiomics features can further improve the predictive performance of the model.
RESUMEN
Objective:This study aimed to investigate the risk features of postoperative tumor recurrence of medullary thyroid carcinoma.Methods:One hundred and seventy two patients with medullary thyroid carcinoma diagnosed at Tianjin Cancer Hospital between Jan 2010 and Jan 2018 were enrolled in this study. Based on the follow-up results, patients were divided into tumor recurrence and non-tumor recurrence group. US features,clinicopathological characteristics and somatic RET mutations were evaluated between the two groups. The cut-off values of pre-and post-operative serum calcitonin were calculated by ROC curve.Univariate and multivariate analysis were adopted between the two groups to determine independent risk factors for tumor recurrence of MTC.Tumor-free survival was determined by Kaplan-Meier analysis.Results:Univariate analysis showed that preoperative serum calcitonin≥1 367 pg/ml ( χ2=18.909, P=0.000), postoperative serum calcitonin ≥61 pg/ml ( χ2=72.278, P=0.000), mulifocality ( χ2=11.787, P=0.001),lesions in both lobes ( χ2=10.452, P=0.003), extrathyroidal invasion ( χ2=14.511, P=0.000), T3+T4-staging ( χ2=11.920, P=0.001)、TNMⅢ+Ⅳ-staging ( χ2=18.915, P=0.000), ACR TI-RADS 5 ( χ2=7.162, P=0.006) and RET mutation ( χ2=10.937, P=0.001) were significantly related to tumor recurrence of medullary thyroid carcinoma. Multivariate analysis demonstrated that postoperative serum calcitonin≥61 pg/ml ( OR=22.323, 95%CI: 6.370-78.236) and RET mutation ( OR=4.054, 95%CI: 1.354-12.139) were the independent factors related to tumor recurrence of medullary thyroid carcinoma.The survival curves of MTC patients showed a significantly lower percentage of surviving patients in the group with postoperative serum calcitonin ≥61 pg/ml ( P=0.000) or RET mutations ( P=0.001). Conclusions:Postoperative serum calcitonin ≥61 pg/ml and oncogenic RET mutation were the independent risk factors for tumor recurrence of MTC.Patients with postoperative serum calcitonin ≥61 pg/ml or a RET mutation tended to have a shorter tumor-free survival.
RESUMEN
Objective:To explore the diagnostic value of strain ultrasonic elastography (SUE) supported by Image Pro Plus (IPP) software in distinguishing benign and malignant Breast Imaging Reporting and Data System (BI-RADS) category 4 nodules.Methods:The clinical and ultrasound imaging data of 192 female patients with BI-RADS category 4 nodules diagnosed by breast ultrasound in Tianjin Medical University Cancer Institute and Hospital from January 2020 to January 2021 were retrospectively analyzed. One nodule was selected for each patient, and a total of 192 nodules were analyzed. The ultrasound images and SUE images of each nodule were acquired before surgery, followed by BI-RADS classification scoring and SUE scoring. The IPP software was applied to outline the region of interest of all nodules, and the software automatically counted the SUE area ratio (SUE-IPP-AR) of all nodules. Using pathological diagnosis as the gold standard, SUE-IPP scoring was performed based on the optimal cut-off value of SUE-IPP-AR for distinguishing benign and malignant nodules according to receiver operating characteristic (ROC) curve. The efficacy of BI-RADS classification, SUE score, SUE-IPP score, and SUE-IPP+BI-RADS combined score in the differentiation of benign and malignant breast nodules were analyzed by ROC curve. Kappa coefficient was used to assess inter-observer agreement for SUE and SUE-IPP-AR.Results:Of the 192 nodules, 58 were benign nodules and 134 were malignant nodules diagnosed by pathology. These nodules were classified by BI-RADS, with 46 nodules in BI-RADS category 4A, 57 nodules in BI-RADS category 4B and 89 nodules in BI-RADS category 4C. SUE images 1-6 level were 12, 14, 41, 51, 42, and 32 nodules, respectively. ROC curve analysis showed that the optimal cut-off value of SUE-IPP for determining benign and malignant BI-RADS category 4 breast nodules was 58% [area under the curve (AUC) = 0.729, sensitivity 73%, specificity 69%]. The AUC of SUE-IPP+BI-RADS combined score for determining benign and malignant nodules was the largest (0.871), which was higher than that of BI-RADS score (AUC = 0.829, Z = 2.51, P = 0.012), SUE-IPP score (AUC = 0.729, Z = 3.56, P < 0.001) and SUE score (AUC = 0.695, Z = 4.37, P < 0.001). The sensitivity of SUE-IPP+BI-RADS combined score ≥ 3 points for diagnosing malignant nodules was 86.6%, while the specificity was 82.8%. Therefore, SUE-IPP+ BI-RADS score had the best efficacy for determining benign and malignant nodules. There was good agreement between sonographer with more than 5 years of experience and those with less than 5 years of experience in applying SUE to diagnose breast malignant nodules (Kappa = 0.768, 95% CI 0.711-0.825), and very good diagnostic agreement in applying SUE-IPP (Kappa = 0.946, 95% CI 0.919-0.974). Conclusions:SUE-IPP can improve the diagnostic ability for BI-RADS category 4 breast malignant nodules and can be used as an adjunct to ultrasound diagnosis.
RESUMEN
Objective:To predict the clinical value of World Health Organization/International Society of Urological Pathology (WHO/ISUP) grading of clear cell renal cell carcinoma (ccRCC) pre-scholarly based on ultrasound imaging group.Methods:Clinical and ultrasound imaging data of patients with surgically pathologically confirmed ccRCC at Tianjin Medical University Cancer Institue and Hospital from January 2021 to October 2022 were retrospectively collected and divided into a low grade group (grade Ⅰ and Ⅱ, 105 cases) and a high grade group (grade Ⅲ and Ⅳ, 70 cases) using WHO/ISUP pathological grading criteria. The clear image of the largest diameter of the tumor was selected and imported into ITK-SNAP software for manual segmentation of the image and extraction of ultrasonographic radiomics features. The patients were randomly divided into a training group and a test group in the ratio of 7∶3, with 122 cases in the training group and 53 cases in the test group. Stable radiomics features were obtained by dimensionality reduction. The support vector machines (SVM) algorithm was applied to predict the pathological grading of ccRCC. Finally, a clinical-ultrasound imaging model, an ultrasonographic radiomics model and a comprehensive model combining the two were constructed. The predictive effects of the three models were analyzed by the area under the ROC curve (AUC). The performance of each model was evaluated by applying the calibration curve. The net benefit of patients was obtained by applying the decision curve.Results:A total of 873 radiomics features were extracted, and 10 features were finally obtained for model construction after dimensionality reduction. Final test results showed that the AUC, sensitivity, specificity and accuracy of the clinical-ultrasound imaging model were 0.68, 0.47, 0.78, 0.66. The AUC, sensitivity, specificity and accuracy of the ultrasonographic radiomics model were 0.74, 0.53, 0.88, 0.74. The AUC, sensitivity, specificity and accuracy of the comprehensive model were 0.84, 0.63, 0.86, 0.77. The AUC of the comprehensive model being larger than that of the clinical-ultrasound imaging model ( Z=-3.224, P=0.001) and ultrasonographic radiomics model ( Z=-2.594, P=0.009). The calibration curves showed that the comprehensive model was more stable than the other two models. The decision curve showed a higher net clinical benefit for the comprehensive model than for the other two models within a threshold of 0.1-1.0. Conclusions:The preoperative prediction of ccRCC pathological grading by the radiomics model based on ultrasound images is effective. The comprehensive model constructed by combining relevant clinical and ultrasound parameters has better performance, which can help predict ccRCC pathological grading preoperatively to a certain extent. It is crucial to help physicians choose the best management plan in the era of personalized medicine.
RESUMEN
Objective:To investigate the prediction of National Institute of Healthy (NIH) risk stratification of gastrointestinal stromal tumor(GIST) based on clinical ultrasound model, ultrasonographic radiomics model and combined model by oral contrast enhanced ultrasonography.Methods:The clinical and ultrasound imaging data of 204 gastric GIST patients attending Tianjin Medical University Cancer Institute and Hospital from June 2021 to June 2022 were retrospectively analyzed, among whom a total of 101 patients with high and moderate NIH risk stratification GIST confirmed by postoperative pathology were included in the high risk group, and a total of 103 patients with low and extremely low NIH risk stratification GIST were in the low risk group. The ultrasound images of the largest diameter of the GIST were manually segmented by ITK-SNAP software, and Pyradiomics (v3.0.1) module in Python 3.8.7 was applied to extract ultrasonographic radiomics features from the ROI segmented images. The patients were randomly divided into training and validation sets in the ratio of 7∶3. The XGBoost of Sklearn module was applied to construct the clinical ultrasound imaging model, ultrasonographic radiomics model, and combined model. Then the area under ROC curve (AUC), sensitivity, specificity, and accuracy were evaluated; the predictive ability of the three models was compared by Delong test. Calibration Curve was applied to evaluate the model performance, and the clinical Decision Curve Analysis was applied to determine the net benefit to patients.Results:A total of 578 ultrasonographic radiomics features were extracted from ROI, and 8 ultrasonographic radiomics features were finally retained for modeling after regression and dimensionality reduction. Finally, test results showed that AUC, sensitivity, specificity and accuracy of clinical ultrasound imaging model, ultrasonographic radiomics model and combined model were 0.75, 69.3%, 68.9%, 69.1%; 0.87, 79.2%, 81.6%, 80.4%; 0.91, 80.2%, 83.5%, 81.9%, respectively. Delong test showed that the difference of AUC between ultrasonographic radiomics model and clinical ultrasound imaging model was statistically significant ( Z=2.698, P<0.001), and the combined model was significantly better than clinical ultrasound imaging model ( Z=4.062, P<0.001) and ultrasonographic radiomics model ( Z=2.225, P=0.026). Calibration Curve showed the high performance of combined model, and Decision Curve Analysis showed the superior clinical usefulness of combined model. Conclusions:It is feasible to construct an ultrasonographic radiomics model for GIST NIH risk stratification based on oral contrast enhanced ultrasonography images, and the combined model has more advantageous diagnostic performance, which can identify high risk NIH GIST objectively and stably for clinical purposes.
RESUMEN
Objective:To evaluate the clinical and ultrasonographic features for early diagnosis and prediction of lateral cervical lymph node metastasis of medullary thyroid microcarcinoma.Methods:From Jan 2010 to Jan 2020 233 patients undergoing primary surgery were categorized as "medullary thyroid microcarcinoma" and "medullary thyroid macrocarcinoma". The preoperative clinical and ultrasonographic characteristics, the diagnostic positive rate of preoperative serum calcitonin and fine needle aspiration (FNA) were investigated between two groups. All patients with medullary thyroid microcarcinoma were divided on the basis of wether there was lateral cervical lymph node metastasis.Results:There were statistically significant differences in initial diagnostic method(χ 2=32.290, P=0.000), TNM staging(χ 2=50.300, P=0.000) between medullary thyroid microcarcinoma and medullary thyroid macrocarcinoma. Medullary thyroid microcarcinoma showed more malignant ultrasonic features. The diagnostic accuracy of preoperative serum calcitonin was higher than FNA for medullary thyroid micro carcinoma(χ 2=47.933, P=0.000). Multivariate regression analysis demonstrated that the abutment/perimeter ≥1/4( OR=25.475, 95%CI: 2.320-279.771), preoperative serum calcitonin >65 ng/L( OR=32.663, 95%CI:2.433-438.409) were the independent factor for lateral cervical lymph node metastases of medullary thyroid microcarcinoma. Conclusions:The combination of ultrasonography and serum calcitonin helps establish early diagnosis of medullary thyroid microcarcinoma. Medullary thyroid microcarcinoma with the abutment/perimeter ≥1/4, serum calcitonin >65 ng/L predicts lateral cervical lymph node metastases.
RESUMEN
Objective To compare the diagnostic efficiency of superb micro-vascular imaging ( SMI) and power Doppler imaging ( PDI) in differentiating different size breast tumors . Methods Retrospective analysis were performent in a total of 170 cases of breast tumors of 144 patients screened with SMI and PDI in our hospital from August 2016 to July 2017 ,and pathological results were finally obtained . These tumors were divided into two groups according to the maximum diameter 2 .0 cm . SMI and PDI diagnostic efficiency were compared by showing blood flow distribution ,vascular morphology and Alder grade . Results There were 68 benign tumors and 102 malignant tumors .When the maximum diameter was no more than 2 .0 cm ,there were no significant difference in the blood flow distribution ,vascular morphology and Alder grade between PDI and SMI of benign tumors( all P > 0 .05) ;However ,there were significant differences in the vascular morphology and Alder grade of malignant tumors( all P < 0 .05) ,but no difference in blood flow distribution( P = 0 .174) ;The area under the curve of ROC by PDI and SMI were 0 .724 and 0 .844 ,thus , the difference was statistically significant( P = 0 .024) . When the maximum diameter was more than 2 .0 cm ,the between PDI and SMI differences in blood flow distribution ,vascular morphology and Alder grade were not statistically significant ( all P > 0 .05) in benign tumors ,but there was statistical difference in vascular morphology ( P = 0 .001) in malignant tumors ,and there was no significant difference in blood flow distribution ,Alder grade in malignant tumors between PDI and SMI ( all P > 0 .05) . The area under the curve of ROC by PDI and SMI were 0 .768 and 0 .802 ,the difference was not statistically significant ( P = 0 .447) . Conclusions SMI can show the blood flow characteristics more clearly in breast tumors . When the maximum diameter is no more than 2 .0 cm ,SMI diagnostic efficiency is better than PDI .
RESUMEN
Objective To compare the difference of colour Doppler flow imaging(CDFI)and superb-microvascular imaging(SMI)for detecting blood flow in cystic renal mass,and explore the consistency of conventional ultrasound combined with SMI and Contrast-enhanced ultrasonography(CEUS) in the diagnosis of renal cystic masses based on the Bosniak classification system,so as to evaluate the diagnosis value of SMI in renal cystic masses.Methods Fifty-five patients with renal cystic mass were enrolled in this study and underwent conventional ultrasound,CDFI and SMI.Patients with renal cystic mass at category Ⅱ for higher received contrast-enhanced ultrasonography additionally.The diagnostic performance of SMI and CEUS were evaluated based on final diagnosis obtained by follow-up or pathology diagnosis after surgery.Results In 55 cases,44 cases got the pathological diagnosis afer surgical resection including 38 cases of malignant masses,and 6 cases of benign masses.CDFI and SMI showed significant difference in tumor flow imaging(P <0.05).Kappa Value of SMI and CEUS was 0.866.The sensitivity,specificity, accuracy,positive predictive value and negative predictive value of SMI were 94.8%,75.0%,89.1%,90.2% and 85.7% respectively.Conclusions SMI is superior to CDFI in displaying micro-vascular of separated and solid structure in renal cystic masses.SMI has a better consistency with CEUS in Bosniak classification.SMI can improve the accuracy of non-invasive ultrasound in the diagnosis of renal cystic lesions.
RESUMEN
Objective To evaluate the values of contrast-enhanced ultrasound in diagnosis of cystic renal cell carcino-ma. Methods A total of 73 patients with renal cystic lesions were included in this study. The image features of ultrasound and contrast-enhanced ultrasound examination were analysed. All of patients underwent surgical treatment and had patholog-ical results. The diagnostic values of the ultrasound and contrast-enhanced ultrasound were analyzed by evaluating the im-age features of cystic renal cell carcinoma. Results There were 64 cases of cystic renal cell carcinoma, 9 cases of benign cyst. With ultrasound and color doppler ultrasound,irregular shape, thickness wall, solid ingredients, divisions and more blood flow signals were found in cystic renal cell carcinoma. Renal cyst showed regular shape, few solid component and thin separation and inconspicuous blood flow signals. In contrast-enhanced ultrasound, cystic renal cancer contrast agent appear-ing time was (15.13±4.21)s, and reached the peak time (23.42±5.68)s, fade time was (28.42±4.27)s. The enhanced mode for fast in and fast out was found in 22 cases (34.3%), fast in and slow out in 30 cases (46.8%), slow in and fast out in 2 cases (3.2%), slow in and slow out in 4 cases (6.4%), and synchronously in and out in 6 cases (9.3%). The hyper-enhancement was found in 42 cases (65.6%), the iso-enhancement and hypo-enhancement in 22 cases (34.4%). In renal cyst, There were three cases out of contrast filling. In other 6 cases, the contrast agent appearing time was (16.67±2.73)s, the peak time was (25.83±3.06)s and fade time was (34.17±4.26)s. The enhanced mode for fast in and fast out was found in 1 case (16.7%), fast in and slow out in 1 case (16.7%) and synchronously in and out in 4 cases (66.6%). The hyper-enhancement was found in 2 cases (33.3%), the iso-enhancement and hypo-enhancement in 4 cases (66.7%). The sensitivity, specificity, positive predic-tive value, negative predictive value and accuracy of ultrasound were 85.9%, 66.7%, 94.8%, 40.0%and 83.6%. The sensitiv-ity, specificity, positive predictive value, negative predictive value and accuracy of contrast-enhanced ultrasound were 92.2%, 77.8%, 96.7%, 58.3%and 90.4%. Conclusion Contrast-enhanced ultrasound can be used in benign and malignan-cy identification of renal cystic lesion.
RESUMEN
Renal cell carcinoma is one of the most common tumors in the urinary system. Most of these tumors are malignant. The incidence of renal tumors has increased in recent years. Contrast-enhanced ultrasound examination is increasingly applied in clinics and has become a common diagnostic method for renal lesions. Contrast-enhanced ultrasound can clearly show the organization of tu-mor blood vessels and capillaries and provide dynamic observation of the blood perfusion status of normal human tissues and tumor tis-sues. This review discusses the situation and value of contrast-enhanced ultrasound in renal tumor treatment.