Search

Objective To develop a predictive model using deep learning (DL) techniques based on breast ultrasound grayscale images and elastography, and to explore the diagnostic efficacy of this model in differentiating benign from malignant breast lesions. Methods Data of 1 000 breast lesions from patients who underwent surgical treatment at Linyi People's Hospital Breast Surgery Department from May 2020 to April 2021 were collected retrospectively. including ultrasound images and related clinical pathological information. The ultrasound grayscale image and elastography of the largest section of each lesion were selected and randomly divided into training, validation, and test sets at a ratio of 7∶2∶1. A predictive model was constructed based on neural networks using the training and validation sets, and the diagnostic efficacy of the model was tested with the test set images. Four sonographers were invited to read the test set ultrasound images independently, and their diagnostic efficacies were compared with the model's performance. Results The area under the curve (AUC) value (0.907) of the receiver operating characteristic curve (ROC) of DL model for breast lesion diagnosis was higher than all participating sonographers, with a statistically significant difference (P＜0.05). The average AUC value for the diagnosis by senior sonographers (0.835) was higher than that for junior sonographers (0.719), with a statistically significant difference (P＜0.05). When the model assisted junior sonographers and senior sonographers in diagnosing the test set breast lesions, the average AUC value was 0.806 and 0.864, respectively.〖LM〗 After assistance from the model, the diagnostic efficacy of sonographers of different experience levels improved, with a more significant increase for junior sonographers (P＜0.05). Notably, there was no statistically significant difference in the AUC values between the junior sonographers assisted by the DL model and the senior sonographers reading alone (P＞0.05). Conclusion A predictive model based on dual-modality ultrasound DL can significantly improve the diagnostic efficacy of sonographers in differentiating benign from malignant breast lesions.

Aiming at welding fracture problem of EGR cooler with eccentric intake structure,numerical simulation based on sequential fluid-solid coupling was presented and the improvement of eccentric intake structure was taken. Flow field mesh and solid finite element mesh were set up based on 3 D modeling. The sequential fluid-solid coupling with flow field analysis,heat conduction analysis and thermal stress analysis was taken to study the relation among intake structure,gas side streamline,solid temperature,thermal stress range and the fracture. An improvement scheme with increasing a hump was proposed to improve gas uniformity and reduce local thermal stress range. The results indicate that the sequential fluid-solid coupling is feasible in engineering practice,the partial fracture of EGR cooler will caused by unreasonable eccentric intake structure and will be solved by optimization of flow passage.

Objective To explore whether the non-invasive method of ultrasound elastography can diagnose and grade hepatic steatosis by using a rat model system. Methods Fifty healthy male Wistar rats were divided into a model group (n =40 ) and a control group (n = 10) . The model was induced by high-fat diet (20% lard), alcohol intake (10% in drinking water), and CCl4 subcutaneous injections (40% in olive oil ; 0.2 ml/kg two times a week) . The control rats received normal diet, no alcohol, and olive oil injections. After weeks 6, 8, 10, 12, and 14 of the regimen, rats from each group were randomly selected and subjected to liver ultrasound elastography analysis ( Hitachi HV900 Color Doppler instrument; 6-13 MHz linear probe frequency), first as transabdominal imaging then by fully exposed rat liver imaging, followed by pathological examination with hematoxylin-eosin staining. The imaging techniques were performed in triplicate for each rat and averaged. The imaging findings of controls (B) were compared to those of models (A) and used to calculate the strain ratio (B/A). Hepatic steatosis detected by pathology was graded as percent of fatty degeneration: normal, <5%; mild, 5%-33%; moderate, 34% -66%; severe, >66%. The correlations between B/A and pathology grades were assessed by Kruskal-Wallis and Spearman's coefficient tests. Results Ten rats died during the establishment of the hepatic steatosis model system. Of the 30 surviving model rats assessed by ultrasound and pathology, the hepatic steatosis grade was mild in 10, moderate in 12, and severe in 8; all control rats showed normal liver. The transabdominal imaging results (B/A) were significantly different between control rats (0.25±0.04) and rats with all three grades of hepatic steatosis ( vs. mild: 0.43 ± 0.05, moderate: 0.59 ± 0.17, and severe : 0.89 ± 0.21; P<0.05). The B/A was positively correlated with the pathology grade (r=0.88, P<0.001) . Conclusion Ultrasonic elastography effectively diagnosed hepatic steatosis and was able to quantitate various pathologic grades in a non-invasive manner, suggesting its potential utility for clinical application. [ABSTRACT FROM AUTHOR]