The differentiation between benign and malignant thyroid nodules is pivotal in clinical practice. However, current diagnostic systems face challenges, such as heterogeneity in ultrasound interpretation, the invasiveness of fine-needle aspiration cytology and indeterminate of results. Artificial intelligence (AI) has evolved from traditional machine learning to deep learning, enabling advances in image recognition, feature extraction, and multimodal data integration. Nevertheless, AI applications remain constrained by limitations, including the insufficiency of high-quality annotated data and inadequate cross-center generalizability. Future progress will focus on algorithm refinement, multimodal data fusion, and multicenter clinical validation to enhance model generalizability and diagnostic accuracy.

The differentiation between benign and malignant thyroid nodules is pivotal in clinical practice. However, current diagnostic systems face challenges, such as heterogeneity in ultrasound interpretation, the invasiveness of fine-needle aspiration cytology and indeterminate of results. Artificial intelligence (AI) has evolved from traditional machine learning to deep learning, enabling advances in image recognition, feature extraction, and multimodal data integration. Nevertheless, AI applications remain constrained by limitations, including the insufficiency of high-quality annotated data and inadequate cross-center generalizability. Future progress will focus on algorithm refinement, multimodal data fusion, and multicenter clinical validation to enhance model generalizability and diagnostic accuracy.

Objective:To assess the accuracy of lung ultrasound(LUS) to predict pneumonia in pediatric patients using meta-analysis.Methods:The PubMed, the Cochrane Library, EMbase databases from January 2015 to March 2020 were searched.The retrieved outcome data to evaluate the efficacy of LUS for the diagnosis of pneumonia in patients under 18 years of age were included.Two reviewers independently screened literature, extracted data and assessed the risk of bias of included studies.Meta-analysis was then conducted using MetaDisc 1.4, RevMan 5.3 and Stata 15.0 softwares.Results:Twelve diagnostic studies were included, which involved 2 484 patients.The results of meta-analysis showed that compared with the gold standard, the sensitivity, specificity, positive and negative likelihood ratios, diagnostic odds ratio of LUS were 0.90(95% CI 0.88-0.91), 0.88(95% CI 0.85-0.90), 8.64(95% CI 3.79-19.72), 0.12(95% CI 0.06-0.26) and 77.58(95% CI 28.39-211.99), respectively.The area under the summary receiver operating characteri stic curve was 0.96.Subgroup analysis showed that there was no difference in LUS′s diagnostic accuracy for pneumonia with different department, different diagnostic gold standard, and different level of sonographer training. Conclusion:Current evidence shows that LUS has a high accuracy in the diagnosis of pneumonia in children.

Objective To explore the significance of GBA gene mutation and gene detection in diagnosis of Gaucher disease. Method The clinical data and genetic testing results of 3 probands from 3 unrelated Gaucher families and their family members were analyzed. Results A compound heterozygous mutation of c.907C>A and c.1448T>C was found in the proband of the first family, which was inherited from parents respectively. Another complex heterozygous mutation of c.1174delC and c.1226A>G was found in the proband of second family, which was inherited from parents respectively, and the variant c.1174delC was a new mutation, which has not been reported in the literature according to the search by HGMD. The homozygous nucleotide variation of c.1342G>C and heterozygous nucleotide variation of c.1263_1317del was found in the proband of the third family and the c.1263_1317del heterozygous mutation was inherited from father. Conclusion The mutation of GBA gene was the cause of Gaucher disease in these 3 families and Gaucher disease can be diagnosed by molecular genetics in clinic.