Qi Zhou

Objectives From the viewpoint of ultrasound (US) physicians, an ideal thyroid US computer-assisted diagnostic (CAD) system for thyroid cancer should perform well in suspicious thyroid nodules with atypical risk features and be able to output explainable results. This study aims to develop an explainable US CAD model for suspicious thyroid nodules. Methods A total of 2992 solid or almost-solid thyroid nodules were analyzed retrospectively. All nodules had pathological results (1070 malignancies and 1992 benignities) confirmed by ultrasound-guided fine-needle aspiration cytology and histopathology after thyroidectomy. A deep learning model (ResNet50) and a multiple risk features learning ensemble model (XGBoost) were used to train the US images of 2794 thyroid nodules. Then, an integrated AI model was generated by combining both models. The diagnostic accuracies of the three AI models (ResNet50, XGBoost, and the integrated model) were predicted in a testing set including 198 thyroid nodules and compared to the diagnostic efficacy of five ultrasonographers. Results The accuracy of the integrated model was 76.77%, while the mean accuracy of the ultrasonographers was 68.38%. Of the risk features, microcalcifications showed the highest contribution to the diagnosis of malignant nodules. Conclusions The integrated AI model in our study can improve the diagnostic accuracy of suspicious thyroid nodules and output the known risk features simultaneously, thus aiding in training young ultrasonographers by linking the explainable results to their clinical experience and advancing the acceptance of AI diagnosis for thyroid cancer in clinical practice. Key Points • We developed an artificial intelligence (AI) diagnosis model based on both deep learning and multiple risk feature ensemble learning methods. • The AI diagnosis model showed higher diagnostic accuracy for suspicious thyroid nodules than ultrasonographers. • The AI diagnosis model showed partial explainability by outputting the known risk features, thus aiding young ultrasonic doctors in increasing the diagnostic level for thyroid cancer.

The aim of the present study was to investigate the changes in cardiac function and myocardial damage in rats with cirrhosis. In addition, a secondary aim was to explore any potential changes in the expression levels of β1‑adrenergic (β1) and muscarinic acetylcholine (M2) receptors . A cirrhotic cardiomyopathy (CCM) rat model was established by CCL4‑oil solution for subcutaneous injection into the neck. Pathological changes in the liver and myocardial tissues were detecting by H&E staining and Masson trichrome staining. Furthermore, changes in the levels of myocardial enzymes lactate dehydrogenase (LDH), creatine kinase isoenzyme (CK‑MB) and troponin in serum were measured by ELISA. The myocardial samples were homogenized and centrifuged. Subsequently, the supernatant was collected for detecting the expression of interleukins in myocardial tissue. Changes in the levels of inflammatory factors, IL‑1, IL‑2 and IL‑6 both in the serum and myocardial tissue were determined by ELISA. Changes in echocardiographic measurements were evaluated using high‑frequency ultrasound and the expression levels of β1 and M2 receptors in myocardial tissues were determined by western blotting. The normal lobular structure in liver tissues was found to be disappeared 8 weeks after modeling, which was replaced by pseudolobules in the rats in the CCM group. In addition, the myocardial cells were observed to be swollen and disorderly arranged. Compared with those in the control group, the left ventricular end‑systolic and end‑diastolic dimensions, interventricular septal dimension and LAD in rats in the CCM8 group were found to be significantly increased. The levels of myocardial enzymes LDH, CK‑MB and cardiac troponin in the serum were also revealed to be significantly increased in the CCM8 group. Additionally, the levels of IL‑1 and IL‑6 in both serum and myocardial tissues were significantly increased in rats in the CCM8 group. However, the levels of IL‑2 in both serum and myocardial tissues were decreased, which were observed alongside reductions in myocardial β1 and M2 receptor protein expression in the myocardial tissues. Taken together, these results indicate that inflammatory factors may be involved in mediating damage to the myocardium in rats with cirrhosis. During cirrhosis‑induced cardiac dysfunction, there may exist a mechanism for downregulation of autonomic nerve system.

Although the connection between SEPS1 gene variants and Hashimoto's thyroiditis (HT) has been established in Europeans, the relationship between the SEPS1 gene and HT remains unclear in Han Chinese. Here we aimed to investigate the potential association between SEPS1 variants and HT in the Han population. In addition, the effects of SEPS1 haplotypes on the susceptibility of the development of immune-mediated diseases with an inflammatory component will also be evaluated. Seven single-nucleotide polymorphisms (SNPs) with minor allele frequency ⩾0.05 were genotyped in 1013 HT patients and 2998 healthy controls from genetically independent Han Chinese individuals. We identified that the rs28665122 SNP was significantly associated with HT, both in the female group (allelic P=0.002644 and genotypic P=0.010326) and the combined data set (allelic P=0.000518 and genotypic P=0.002731). Further analyses based on haplotypes indicated that a two-SNP haplotype (rs2009895-rs28665122) was significantly associated with HT (global P=0.0036), which was also observed in females (global P=0.0162) but not in males. Our findings provide further supporting evidence that confirms the results of previous studies, which suggested potential roles of the SEPS1 gene in the pathogenesis and etiology of HT.