Dataset Information

Deep learning to diagnose Hashimoto's thyroiditis from sonographic images.

ABSTRACT: Hashimoto's thyroiditis (HT) is the main cause of hypothyroidism. We develop a deep learning model called HTNet for diagnosis of HT by training on 106,513 thyroid ultrasound images from 17,934 patients and test its performance on 5051 patients from 2 datasets of static images and 1 dataset of video data. HTNet achieves an area under the receiver operating curve (AUC) of 0.905 (95% CI: 0.894 to 0.915), 0.888 (0.836-0.939) and 0.895 (0.862-0.927). HTNet exceeds radiologists' performance on accuracy (83.2% versus 79.8%; binomial test, p < 0.001) and sensitivity (82.6% versus 68.1%; p < 0.001). By integrating serologic markers with imaging data, the performance of HTNet was significantly and marginally improved on the video (AUC, 0.949 versus 0.888; DeLong's test, p = 0.004) and static-image (AUC, 0.914 versus 0.901; p = 0.08) testing sets, respectively. HTNet may be helpful as a tool for the management of HT.

SUBMITTER: Zhang Q

PROVIDER: S-EPMC9243092 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:ObjectiveHashimoto's thyroiditis, also known as chronic lymphocytic thyroiditis, is a common autoimmune thyroiditis, which mostly occurs in young and middle-aged women. It can be manifested as hyperthyroidism in the early stage; hypothyroidism may appear with the progression of the disease. Studies have shown that multiple factors such as heredity, environment, and autoimmunity are involved in the pathogenesis, but the specific mechanism is not clear. In our study, we tried to find key genes and potential molecular mechanisms of Hashimoto's thyroiditis to provide new ideas for the therapeutic targets of Hashimoto's thyroiditis.MethodGSE138198 and GSE54958 were downloaded from the GEO database, and two datasets were combined for analysis. The combined data were normalized to identify the differentially expressed genes (DEGs), and GO and KEGG enrichment analyses were performed. Protein-protein interaction (PPI) networks and hub genes between DEGs were identified. We also used the miRWalk database to identify regulatory miRNAs associated with expressions of DEGs.ResultWe identified 182 DEGs (160 upregulated and 22 downregulated) between Hashimoto's disease patients and the healthy control group. GO analysis showed that DEGs were mostly concentrated in detection of chemical stimulus involved in sensory perception, intermediate filament cytoskeleton, and olfactory receptor activity. KEGG pathway analysis showed that DEGs were mainly related to olfactory transduction. Some members of the KRTAP family and HTR5A, KNG1, DRD3, HTR1D, TAS2R16, INSL5, TAS2R42, and GRM7 are the most important hub genes in the PPI network. In addition, we recognized that OTUD4, LLPH, and ECHDC1 were the most important hub genes in the miRNA-target gene network.ConclusionIn this study, a series of bioinformatics analyses of DEGs were performed to identify the key genes and pathways associated with Hashimoto's thyroiditis. These genes and pathways provide a more detailed understanding of the pathogenesis of Hashimoto's disease and provide new ideas for the therapeutic targets of Hashimoto's thyroiditis.

Dataset Information

Deep learning to diagnose Hashimoto's thyroiditis from sonographic images.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets