ObjectiveTo explore the relation between ultrasound images of endometriosis and its clinical symptoms. MethodsChoosing clinical data of 300 patients with endometriosis pathologically diagnosed between January 2009 and January 2014, we retrospectively analyzed ultrasound images and clinical symptoms, using Chisquare test for statistical analysis, and the index P<0.05 was statistically significant. ResultsIn patients with big endometriosis' nidus, the menstrual quantity increased, menstrual cycle prolonged, the incidence of abnormally vaginal bleeding was high (χ2=11.749, P=0.001; χ2=4.847, P=0.028; χ2=5.686, P=0.017). In patients whose endometriosis were located in posterior uterine wall, the menstrual quantity increased, and the incidence of abnormally vaginal bleeding was high (χ2=5.188, P=0.023; χ2=49.691, P<0.001). The size of endometriosis' nidus had nothing to do with dysmenorrhea, constipation and frequent micturition (P>0.05). The position of endometriosis' nidus had nothing to do with menostaxis, dysmenorrhea, constipation and frequent micturition (P>0.05). ConclusionThe size of endometriosis' nidus has a connection with the clinical symptoms of menorrhea, menostaxis and abnormally vaginal bleeding; the position of endometriosis' nidus has a connection with the clinical symptoms of menorrhea and abnormally vaginal bleeding. The results of ultrasonography should be combined with clinical symptoms in diagnosing endometriosis, avoiding missed-diagnosis and misdiagnosis.
Ultrasound guided percutaneous interventional therapy has been widely used in clinic. Aiming at the problem of soft tissue deformation caused by probe contact force in robot-assisted ultrasound-guided therapy, a real-time non-reference ultrasound image evaluation method considering soft tissue deformation is proposed. On the basis of ultrasound image brightness and sharpness, a multi-dimensional ultrasound image evaluation index was designed, which incorporated the aggregation characteristics of the organization. In order to verify the effectiveness of the proposed method, ultrasound images of four different models were collected for experiments, including prostate phantom, phantom with cyst, pig liver tissue, and pig liver tissue with cyst. In addition, the correlation between subjective and objective evaluations was analyzed based on Spearman’s rank correlation coefficient. Experimental results showed that the average evaluation time of a single image was 68.8 milliseconds. The evaluation time could satisfy real-time applications. The proposed method realizes the effective evaluation of real-time ultrasound image quality in robot-assisted therapy, and has good consistency with the evaluation of supervisors.
The registration of preoperative magnetic resonance (MR) images and intraoperative ultrasound (US) images is very important in the planning of brain tumor surgery and during surgery. Considering that the two-modality images have different intensity range and resolution, and the US images are degraded by lots of speckle noises, a self-similarity context (SSC) descriptor based on local neighborhood information was adopted to define the similarity measure. The ultrasound images were considered as the reference, the corners were extracted as the key points using three-dimensional differential operators, and the dense displacement sampling discrete optimization algorithm was adopted for registration. The whole registration process was divided into two stages including the affine registration and the elastic registration. In the affine registration stage, the image was decomposed using multi-resolution scheme, and in the elastic registration stage, the displacement vectors of key points were regularized using the minimum convolution and mean field reasoning strategies. The registration experiment was performed on the preoperative MR images and intraoperative US images of 22 patients. The overall error after affine registration was (1.57 ± 0.30) mm, and the average computation time of each pair of images was only 1.36 s; while the overall error after elastic registration was further reduced to (1.40 ± 0.28) mm, and the average registration time was 1.53 s. The experimental results show that the proposed method has prominent registration accuracy and high computational efficiency.
Existing classification methods for myositis ultrasound images have problems of poor classification performance or high computational cost. Motivated by this difficulty, a lightweight neural network based on a soft threshold attention mechanism is proposed to cater for a better IIMs classification. The proposed network was constructed by alternately using depthwise separable convolution (DSC) and conventional convolution (CConv). Moreover, a soft threshold attention mechanism was leveraged to enhance the extraction capabilities of key features. Compared with the current dual-branch feature fusion myositis classification network with the highest classification accuracy, the classification accuracy of the network proposed in this paper increased by 5.9%, reaching 96.1%, and its computational complexity was only 0.25% of the existing method. The obtained results support that the proposed method can provide physicians with more accurate classification results at a lower computational cost, thereby greatly assisting them in their clinical diagnosis.