1.Application value of tissue dispersion quantitative analysis technique in different stages of SAT
Chuanju ZHANG ; Chunping NING ; Jiawei TIAN ; Bowen ZHAO ; Jiang ZHU ; Jianghong LV ; Haishan XU ; Jinduo SHOU ; Liming YANG ; Ran CHEN
Chinese Journal of Ultrasonography 2017;26(5):419-423
Objective To investigate the application value of the ultrasonic elastic tissue dispersion quantitative analysis technique in different stages of subacute thyroiditis (SAT).Methods One hundred and forty-four SAT lesions detected from 81 patients were enrolled in the patient group.They were further divided into three subgroups,including acute group (group Ⅰ),medium group (group Ⅱ) and recovery group (group Ⅲ).Another 59 healthy volunteers were collected as control group.All the participants accepted conventional ultrasound and elastographic examinations.Eleven parameters were obtained by the tissue dispersion quantitative analysis software.These parameters were compared between groups and among subgroups by ANOVA.The correlation between all the parameters and the course of SAT were analyzed by Spearman and Multiple linear regression methods.Results Between groups and among subgroups,the complexity (COMP) and correlation (CORR) were not statistically different(all P >0.05).Differences of kurtosis (KURT) and angular secon moment (ASM) among the three subgroups were not significant (all P >0.05).Differences between groups and among subgroups were significantly different among the value of all the other seven indexes (all P <0.01).Moreover,they were all correlated with the clinical staging,with the highest coefficient in area ration of low-strain region (% AREA)(r =-0.881).Regression model was constructed and only % AREA was selected into the regression equation.ROC curves were constructed to estimate the clinic value of % AREA in staging patients of SAT,the areas under ROC curves were0.986(group Ⅰ vs group Ⅱ-Ⅲ) and 0.988 (group Ⅰ-Ⅱ vs group Ⅲ[) for %AREA,respectively.Conclusions The tissue dispersion quantitative analysis technique is helpful in estimating the stiffness of thyroid in patients with SAT.
2.The application value of the tissue dispersion quantitative analysis technique in differentiating thyroid nodules
Chuanju ZHANG ; Bowen ZHAO ; Jianghong LYU ; Haishan XU ; Jinduo SHOU ; Lilong XU ; Liming YANG ; Jiang ZHU
Chinese Journal of Ultrasonography 2020;29(10):870-874
Objective:To investigate the application value of the ultrasonic elastic tissue dispersion quantitative analysis technique in differentiating thyroid nodules.Methods:A total of 164 nodules in 143 patients with thyroid nodules were examined by elastography ultrasound at Sir Run Run Shaw Hospital of Zhejiang University School of Medicine from January to November 2014. Eleven parameters were obtained by the tissue dispersion quantitative analysis software. These parameters were compared between benign and malignant groups by Mann-Whitney U test. The correlations between all the parameters and the pathologic results of thyroid nodules were analyzed by Spearman analysis. The receiver operating characteristic(ROC) curve of the parameter with the highest correlation coefficient was constructed. The cut-off value was calculated. Results:All parameters except correlation (CORR) had statistically significant differences between the groups of benign and malignant thyroid nodules(all P<0.01). Moreover, except CORR, the other parameters were correlated with the pathologic results of thyroid nodules(all P<0.05), with the highest coefficient in area ration of low-strain region (%AREA)( r s=0.818). ROC curves were constructed to estimate the clinic values of %AREA in diagnosis of thyroid cancer, the area under ROC curve was 0.991 for %AREA, the cut-off point was 74.83%, the sensitivity and specifity was 98.1% and 89.8%, respectively. Conclusions:The tissue dispersion quantitative analysis technique has high value in the differential diagnosis of benign and malignant thyroid nodules.
3.Preliminary study on thyroid ultrasound image restoration algorithm based on deep learning
Min ZHANG ; Chiming NI ; Jiaheng WEN ; Ziye DENG ; Haishan XU ; Haiya LOU ; Mei PAN ; Qiang LI ; Ling ZHOU ; Chuanju ZHANG ; Yu LING ; Jiaoni WANG ; Juanping CHEN ; Gaoang WANG ; Shiyan LI
Chinese Journal of Ultrasonography 2023;32(6):515-522
Objective:To explore the feasibility of deep learning-based restoration of obscured thyroid ultrasound images.Methods:A total of 358 images of thyroid nodules were retropectively collected from January 2020 to October 2021 at Sir Run Run Shaw Hospital, Zhejiang University School of Medicine, and the images were randomly masked and restored using DeepFillv2. The difference in grey values between the images before and after restoration was compared, and 6 sonographers (2 chief physicians, 2 attending physicians, 2 residents) were invited to compare the rate of correctness of judgement and detection of image discrepancies. The ultrasound features of thyroid nodules (solid composition, microcalcifications, markedly hypoechoic, ill-defined or irregular margins, or extrathyroidal extensions, vertical orientation and comet-tail artifact) were extracted according to the Chinese Thyroid Imaging Reporting and Data System (C-TIRADS). The consistency of ultrasound features of thyroid nodules before and after restoration were compared.Results:The mean squared error of the images before and after restoration ranged from 0.274 to 0.522, and there were significant differences in the rate of correctness of judgement and detection of image discrepancies between physicians of different groups(all P<0.001). The overall accuracy rate was 51.95%, the overall detection rate was 1.79%, there were significant differences also within the chief physicians and resident groups (all P<0.001). The agreement rate of all ultrasound features of the nodules before and after image restoration was higher than 70%, over 90% agreement rate for features such as solid composition and comet-tail artifact. Conclusions:The algorithm can effectively repair obscured thyroid ultrasound images while preserving image features, which is expected to expand the deep learning image database, and promote the development of deep learning in the field of ultrasound images.