No abstract

Background To establish the normal reference range of fetal thorax by two-dimensional (2D) and three-dimensional (3D) ultrasound VOCAL technique and evaluate the application in diagnosing fetal thoracic malformations. Methods A prospective cross-sectional study was undertaken involving 1077 women who have a normal singleton pregnancy at 13–40 weeks gestational age (GA). 2D ultrasound and 3D ultrasound VOCAL technique were utilized to assess fetal thoracic transverse diameter, thoracic anteroposterior diameter, thoracic circumference, thoracic area, lung volume, thoracic volume and lung-to-thoracic volume ratio. The nomograms of 2D and 3D fetal thoracic measurements were created to GA. 50 cases were randomly selected to calculate intra- and inter-observer reliability and agreement. In addition, the case groups including congenital skeletal dysplasia (SD) (15), congenital diaphragmatic hernia (CDH) (30), pulmonary sequestration (PS) (25) and congenital cystic adenomatoid malformation (CCAM) (36) were assessed by the nomograms and followed up subsequently. Results Both 2D and 3D fetal thoracic parameters increased with GA using a quadratic regression equation. The intra- and inter-observer reliability and agreement of each thoracic parameter were excellent. 2D fetal thoracic parameters could initially evaluate the fetal thoracic development and diagnose the skeletal thoracic deformity, and lung volume, thoracic volume and lung-to-thorax volume ratio were practical to diagnose and differentiate CDH, PS and CCAM. Conclusion We have established the normal fetal thoracic reference range at 13–40 weeks, which has a high value in diagnosing congenital thoracic malformations.

Ultrasound is one of the critical methods for diagnosis and treatment in thyroid examination. In clinical application, many reasons, such as large outpatient traffic, time-consuming training of sonographers, and uneven professional level of physicians, often cause irregularities during the ultrasonic examination, leading to misdiagnosis or missed diagnosis. In order to standardize the thyroid ultrasound examination process, this paper proposes using a deep learning method based on residual network to recognize the Thyroid Ultrasound Standard Plane (TUSP). At first, referring to multiple relevant guidelines, eight TUSP were determined with the advice of clinical ultrasound experts. A total of 5,500 TUSP images of 8 categories were collected with the approval and review of the Ethics Committee and the patient’s informed consent. Then, after desensitizing and filling the images, the 18-layer residual network model (ResNet-18) was trained for TUSP image recognition, and five-fold cross-validation was performed. Finally, through indicators like accuracy rate, we compared the recognition effect of other mainstream deep convolutional neural network models. Experimental results showed that ResNet-18 has the best recognition effect on TUSP images with an average accuracy rate of 91.07%. The average macro precision, average macro recall, and average macro F1-score are 91.39%, 91.34%, and 91.30%, respectively. It proves that the deep learning method based on residual network can effectively recognize TUSP images, which is expected to standardize clinical thyroid ultrasound examination and reduce misdiagnosis and missed diagnosis.

In this paper, a multiple-input-multiple-output (MIMO) system with finite-bit feedback first proposed by Love-Heath is considered, where a transmitted signal consists of a precode followed by an orthogonal space-time block code (OSTBC), such as Alamouti code. A new design criterion and a corresponding design method of precoders are proposed. Simulations show that the precoders obtained by our proposed criterion and method perform better than the existing ones. Furthermore, since our proposed precoders have a layered structure, their designs can be implemented in the simplest Grassmannian manifold. Moreover, a fast encoding algorithm can be applied, which can greatly reduce the complexity of codeword selection. In this paper, we also propose non-unitary precoders and their design criterion and method based on the performance analysis and the special property of an OSTBC. Interestingly, non-unitary precoders can significantly improve performance over unitary precoders.