Objective To investigate the value of intraoperative ultrasound radiomics for predicting isocitrate dehydrogenase 1(IDH1)mutation of high-grade glioma.Methods Ninety-five patients with high-grade glioma(WHO grade Ⅲ and Ⅳ)who underwent craniotomy glioma resection and ultrasound assisted tumor localization during operation and then confirmed by pathology were retrospectively enrolled.The patients were divided into training set(n=66,including 24 IDH1 mutation type and 42 IDH1 wild type)and validation set(n=29,including 11 IDH1 mutation type and 18 IDH1 wild type)at the ratio of 7∶3.Based on intraoperative ultrasound,radiomics features were extracted,the best ones were screened,and a radiomics model was established for predicting IDH1 mutation of high-grade glioma using random forest algorithm.Receiver operating characteristic(ROC)curve was plotted,the area under the curve(AUC)was calculated to evaluate the predictive efficacy of the model,and decision curve analysis(DCA)was used to evaluate the clinical value of the model.Results A total of 851 radiomics features were extracted based on intraoperative ultrasound,and finally 5 best ones were screened out to construct a radiomics model.The AUC of the radiomics model for predicting IDH1 mutation of high-grade glioma in training and validation sets was 0.902 and 0.707,respectively,with no significant difference(P=0.097).DCA maps showed that the clinical net benefit of the radiomics model was high.Conclusion Intraoperative ultrasound radiomics could effectively predict IDH1 mutation of high-grade glioma.

Objective To observe the efficacy of a nomogram based on conventional ultrasonic manifestations for predicting malignant risk of breast non-mass lesions(NML).Methods A total of 120 NML diagnosed with ultrasound in 113 female patients were retrospectively included and divided into malignant group(n=44)and benign group(n=76)based on pathological results.Patients' age,ultrasonic manifestations of NML and ipsilateral axillary lymph node were compared between groups.Then a nomogram was constructed using a multivariable logistic regression model,and its diagnostic performance and accuracy were validated.Results The ratios of patients' aged≥45 years,with abnormal ipsilateral axillary lymph nodes,NML with maximum diameter≥2 cm,posterior echo attenuation,rich blood flow signals,architectural distortion and microcalcification were significantly different between groups(all P＜0.05).Patient's age≥45 years,with abnormal ipsilateral axillary lymph nodes,and NML with rich blood flow signals,architectural distortion and microcalcification were all independent predictive factors for high malignant risk of breast NML(all P＜0.05).The constructed nomogram achieved an area under the receiver operating characteristic curve of 0.901 for predicting malignant risk of breast NML,and its predicting probabilities demonstrated good agreement with the actual probabilities(Hosmer-Lemeshow test P＞0.05).Conclusion The nomogram incorporating patient's age,ultrasonic manifestations of NML and ipsilateral axillary lymph node could effectively quantify malignancy risk of breast NML.

Objective To investigate the value of intraoperative ultrasound radiomics for predicting isocitrate dehydrogenase 1(IDH1)mutation of high-grade glioma.Methods Ninety-five patients with high-grade glioma(WHO grade Ⅲ and Ⅳ)who underwent craniotomy glioma resection and ultrasound assisted tumor localization during operation and then confirmed by pathology were retrospectively enrolled.The patients were divided into training set(n=66,including 24 IDH1 mutation type and 42 IDH1 wild type)and validation set(n=29,including 11 IDH1 mutation type and 18 IDH1 wild type)at the ratio of 7∶3.Based on intraoperative ultrasound,radiomics features were extracted,the best ones were screened,and a radiomics model was established for predicting IDH1 mutation of high-grade glioma using random forest algorithm.Receiver operating characteristic(ROC)curve was plotted,the area under the curve(AUC)was calculated to evaluate the predictive efficacy of the model,and decision curve analysis(DCA)was used to evaluate the clinical value of the model.Results A total of 851 radiomics features were extracted based on intraoperative ultrasound,and finally 5 best ones were screened out to construct a radiomics model.The AUC of the radiomics model for predicting IDH1 mutation of high-grade glioma in training and validation sets was 0.902 and 0.707,respectively,with no significant difference(P=0.097).DCA maps showed that the clinical net benefit of the radiomics model was high.Conclusion Intraoperative ultrasound radiomics could effectively predict IDH1 mutation of high-grade glioma.

Objective To observe the efficacy of a nomogram based on conventional ultrasonic manifestations for predicting malignant risk of breast non-mass lesions(NML).Methods A total of 120 NML diagnosed with ultrasound in 113 female patients were retrospectively included and divided into malignant group(n=44)and benign group(n=76)based on pathological results.Patients' age,ultrasonic manifestations of NML and ipsilateral axillary lymph node were compared between groups.Then a nomogram was constructed using a multivariable logistic regression model,and its diagnostic performance and accuracy were validated.Results The ratios of patients' aged≥45 years,with abnormal ipsilateral axillary lymph nodes,NML with maximum diameter≥2 cm,posterior echo attenuation,rich blood flow signals,architectural distortion and microcalcification were significantly different between groups(all P＜0.05).Patient's age≥45 years,with abnormal ipsilateral axillary lymph nodes,and NML with rich blood flow signals,architectural distortion and microcalcification were all independent predictive factors for high malignant risk of breast NML(all P＜0.05).The constructed nomogram achieved an area under the receiver operating characteristic curve of 0.901 for predicting malignant risk of breast NML,and its predicting probabilities demonstrated good agreement with the actual probabilities(Hosmer-Lemeshow test P＞0.05).Conclusion The nomogram incorporating patient's age,ultrasonic manifestations of NML and ipsilateral axillary lymph node could effectively quantify malignancy risk of breast NML.

Objective To observe the correlations of clinical,multi-modal ultrasonic and pathological features of papillary thyroid carcinoma(PTC)with BRAF V600E mutation.Methods Patients with clinically suspected malignant thyroid lesions were prospectively collected.After the results of pathology and gene test were obtained,PTC patients were divided into BRAF V600E mutation(+)(mutant group)and BRAF V600E mutation(-)(wild-type group)according to BRAF V600E gene test.Univariate analysis and multivariate logistic regression were used to compare clinical,conventional ultrasound,contrast-enhanced ultrasound(CEUS)and pathological manifestations between groups,so as to screen the independent risk factors of PTC BRAF V600E gene mutation.Results A total of 116 PTC patients(116 lesions)were enrolled,including 77 in mutant group and 39 in wild-type group.Statistical differences of thyroid peroxidase antibody(TPO-Ab)level,lesion size,orientation,margin and microcalcification observed on conventional ultrasound,changes of lesion size and the mean transit time(MTT)shown on CEUS,as well as of extramembranous invasion,cervical central lymph node metastasis,complicated with benign nodules shown by pathology were found between groups(all P＜0.05).TPO-Ab level,lesion orientation and changes of lesion size after enhancement shown by multi-modal ultrasound,and pathological cervical central lymph node metastasis were all independent predictive factors for PTC BRAF V600E gene mutation(OR=0.175,3.868,5.769,6.943,all P＜0.05).Conclusion Patients'TPO-Ab level,lesion orientation,changes of lesion size after enhancement shown by multi-modal ultrasound and pathological cervical central lymph node metastasis were all independently associated with PTC BRAF V600E gene mutation.

Objective To compare the value of 7 different thyroid imaging reporting and data systems(TI-RADS)for differentiating benign and malignant thyroid nodules under the context of Hashimoto thyroiditis(HT).Methods A total of 338 thyroid nodules in 200 HT patients were enrolled,including 167 benign and 171 malignant ones.Kwak-TIRADS,American Thyroid Association(ATA)guideline,American Association of Clinical Endocrinologists(A ACE)/American College of Endocrinology(ACE)/Associazione Medici Endocrinologi(AME)guideline,K-TIRADS of Korean Society of Thyroid Radiology,EU-TIRADS of European Thyroid Association,American College of Radiology(ACR)-TIRADS and 2020 Chinese guidelines for malignant risk stratification of thyroid nodules by ultrasound proposed by the superficial organs and vessels group of the ultrasound medicine branch of the Chinese Medical Association(C-TIRADS)were used for grading of benign and malignant thyroid nodules.Taken pathological results as gold standards,the diagnostic efficacy of 7 kinds of TI-RADS were analyzed.Results The sensitivity of Kwak-TIRADS,ATA guideline,A ACE/ACE/AME guideline,K-TIRADS,EU-TIRADS,ACR-TIRADS and C-TIRADS for differentiating benign and malignant thyroid nodules under the context of HT was 97.08％,98.25％,99.42％,95.91％,99.42％,90.06％and 99.42％,respectively,the specificity was 88.02％,83.23％,82.04％,88.02％,82.04％,86.83％and 84.43％,respectively,and the area under the curve(AUC)was 0.946,0.913,0.907,0.934,0.909,0.916 and 0.960,respectively.The sensitivity of C-TIRADS,EU-TIRADS and A ACE/ACE/AME guideline were all higher than that of K-TIRADS and ACR-TIRADS(all P＜0.05),and the specificity of Kawk-TIRADS and K-TIRADS were both higher than that of C-TIRADS,ATA guideline,EU-TIRADS and AACE/ACE/AME guideline(all P＜0.05),while AUC of C-TIRADS and Kawk-TIRADS were both higher than that of the rest 5 kinds of TI-RADS(all P＜0.05).According to Kwak-TIRADS,ATA guideline,AACE/ACE/AME guideline,K-TIRADS,EU-TIRADS,ACR-TIRADS and C-TIRADS,the malignant rate of different grades nodules identified with the same TI-RADS were significant different(all P＜0.05),which all raised with the increase of TI-RADS grade.Conclusion C-TIRADS and Kawk-TIRADS had better value for differentiating benign and malignant thyroid nodules under the context of HT,among which C-TIRADS had higher sensitivity and Kawk-TIRADS had higher specificity.

Objective:To assess the value of S-Detect and contrast-enhanced ultrasound (CEUS) in the differential diagnosis of Breast Imaging Reporting and Data System(BI-RADS) 4 breast lesions.Methods:A total of 104 breast lesions in 100 patients diagnosed as BI-RADS category 4 by conventional ultrasound were prospectively enrolled, and all of them were received S-Detect and CEUS examination at the same time. Taking pathology as the gold standard, ROC curve was constructed to compare the diagnostic efficacy of conventional ultrasound, S-Detect, CEUS and their combination.Results:Among the 104 BI-RADS category 4 breast lesions, 63 were benign and 41 were malignant. The sensitivities of conventional ultrasound, S-Detect, CEUS and S-Detect combined with CEUS were 73.17%, 87.80%, 87.80%, 90.24%; the specificities were 57.14%, 60.32%, 68.25%, 77.78%; the positive predictive values were 52.63%, 59.02%, 64.29% and 72.55%; the negative predictive values were 76.60%, 88.37%, 89.59% and 92.45%; the accuracies were 63.46%, 71.15%, 75.96% and 82.69%; and the areas under the ROC curve (AUC) were 0.652, 0.741, 0.780 and 0.840. The sensitivity, specificity, positive predictive value, negative predictive value and accuracy of S-Detect and CEUS diagnosis were improved compared with conventional ultrasound. The AUC of combined diagnosis was higher than that of S-Detect, CEUS alone, and the differences were statistically significant(all P<0.05). The AUC of CEUS was higher than that of conventional ultrasound, and the difference was statistically significant ( P<0.05). There were no significant differences in AUC between any two of other groups (all P>0.05). Conclusions:The combined application of S-Detect and CEUS could achieve complementary advantages, which is of great significance for the differential diagnosis of benign and malignant in BI-RADS 4 breast lesions.

Objective To investigate the correlation between intraoperative ultrasonographic features and expression of immunohistochemical markers in patients with glioma.Methods Totally 116 patients with glioma confirmed pathologically were collected.Ultrasonic features were observed,including the main site of the tumor,maximum diameter,border,cystic degeneration,calcification,the degree of peritumoral edema and CDFI blood flow classification,and the correlation between ultrasonographic features and immunohistochemical results such as Ki-67 and isocitrate dehydrogenase 1 (IDH1) was analyzed.Results Univariate analysis showed that the border,peritumoral edema and CDFI blood flow were significantly different between negative and positive Ki-67 expression patients (P＜ 0.01).The border,cystic degeneration and peritumoral edema were distinct between negative and positive IDH1 patients (P＜0.01).Multivariate analysis showed that gliomas with clear border,high degree of peritumoral edema and rich CDFI blood flow tended to show positive Ki-67 expression,while those with vague border,low degree of peritumoral edema were frequently accompanied by positive IDH1 expression.Conclusion The border of gliomas,peritumoral edema and blood flow showed on ultrasonography may predict the expression of Ki-67 and IDH1.It is of great significance for preliminary evaluation on biological behaviors and prognosis of the tumors before surgical operation.

Objective To evaluate the vulnerability of carotid arteriosclerosis plaque by ultrasound real-time tissue elastography in patients with SSS-TOAST1 style cerebral infarction, and discussing the value of the technique in assessment of the clinical course after cerebral infarction. Methods There were 113 patients of SSS-TOAST1 style cerebral infarction who had carotid arteriosclerosis plaque and 48 patients of contrast group who had carotid arteriosclerosis plaque selected by ultrasound in Department of Ultrasound, Beijing Tiantan Hospital, Capital University of Medical Sciences. The results between two groups were compared. The cerebral infarction group was divided into two sub-groups according to the clinical course of patients after cerebral infarction, and the difference between them was compared. Results The size had no significant difference between cerebral infarction group and contrast group as well as between aggravated group and non-aggravated group (t=15.61, 10.77, 4.52, P<0.05). The real-time tissue elastography of carotid arteriosclerosis plaques were red-green in most patients of cerebral infarction group. The real-time tissue elastography of carotid arteriosclerosis plaques were green-blue in most patients of in control group. The value of elasticity of plaque, vessel wall and stiffness of carotid arteriosclerosis plaque between cerebral infarction group and control group had significant differences (t=15.61, 10.77, 4.52, P<0.05). The value of real-time tissue elastography between aggravated group and non-aggravated group had significant difference (t=6.39, 2.30, 3.80, P<0.05). Conclusion Real-time tissue elastography could evaluate the stiffness of carotid arteriosclerosis plaque, which was related with the vulnerability of carotid arteriosclerosis plaque. The values of elasticity of plaque, vessel wall and stiffness of carotid arteriosclerosis plaque in patients with SSS-TOAST1 style cerebral infarction were lower, and the vulnerability of carotid arteriosclerosis plaque was higher. Real-time tissue elastography had some worth in evaluating the clinical course of patients after cerebral infarction.

Objective To investigate the alteration of peripheral tissue`s temperature of the coagulation zone of microwave ablation in brain tissue, and to provide experimental evidence for clinical application. Methods Twelve canines were treated by microwave ablation in brain tissue. Each was ablated for 180 s with microwave output power of 20 W, 30 W, and 40 W. During the operation the peripheral temperature at the distance of 0.5 cm, 1.0 cm, 1.5 cm and 2.0 cm from the ablation center was recorded respectively. The ultrasound was performed 1 hour after the operation, and then the animals were executed and the microscopic changes of the ablation lesion were observed. Results Eleven canines suffered well for the ablation, while 1 presented abnormal respiration during the operation and died 2 hours later. During the operation, the temperature of the area 0.5 cm from the center rose signiifcantly, with the maximum temperature was (96.40±1.46)℃at the power of 20 W, and 100℃at the power of 30 W and 40 W. The temperature of the area 1.0 cm from the center rose faster, with the maximum temperatures at different powers all above the 46℃. The temperature of the area 1.5 cm from the center rose slower, with the maximum temperature below 46℃at the power of 20 W and 30 W and above 46℃at the power of 40 W. The maximum temperatures of the area 2.0 cm from the center at different powers were all below 46℃. The difference of the maximum temperature at different distances (1.0 cm, 1.5 cm, and 2.0 cm from the center) was signiifcant (F=776.78, 2640.64 and 3025.53, all P<0.05). The length and width of the ablation lesion as well as the area of edema increased with the power. At the power of 20 W, 30 W, and 40 W, the length of the ablation lesion was (29.3±1.8) mm, (32.7±2.1) mm and (34.2±2.4) mm, the width was (22.5±1.5) mm, (23.7±1.7) mm and (27.1±2.0) mm, and the width of the edema zone was (2.3±0.4) mm, (2.6±0.4) mm and (2.7±0.5) mm. The differences of the length and width of the ablation lesion at different powers were signiifcant (F=11.46, 14.49, both P<0.01). The difference of the edema area at different powers was insigniifcant (F=1.94, P=0.169). Conclusions Microwave ablation is a safe therapeutic modality. However, the shorter distance from the ablation center and greater ablation power give rise to larger ablation lesion, higher maximum temperature, and faster temperature increase. Therefore, 2.0 cm from the ablation center is a safe area.