RESUMO
BACKGROUND: Image text is an important text data in the medical field at it can assist clinicians in making a diagnosis. However, due to the diversity of languages, most descriptions in the image text are unstructured data. The same medical phenomenon may also be described in various ways, such that it remains challenging to conduct text structure analysis. The aim of this research is to develop a feasible approach that can automatically convert nasopharyngeal cancer reports into structured text and build a knowledge network. METHODS: In this work, we compare commonly used named entity recognition (NER) models, choose the optimal model as our triplet extraction model, and present a Chinese structuring algorithm. Finally, we visualize the results of the algorithm in the form of a knowledge network of nasopharyngeal cancer. RESULTS: In NER, both accuracy and recall of the BERT-CRF model reached 99%. The structured extraction rate is 84.74%, and the accuracy is 89.39%. The architecture based on recurrent neural network does not rely on medical dictionaries or word segmentation tools and can realize triplet recognition. CONCLUSIONS: The BERT-CRF model has high performance in NER, and the triplet can reflect the content of the image report. This work can provide technical support for the construction of a nasopharyngeal cancer database.
Assuntos
Idioma , Neoplasias Nasofaríngeas , China , Humanos , Imageamento por Ressonância Magnética , Carcinoma Nasofaríngeo/diagnóstico por imagem , Neoplasias Nasofaríngeas/diagnóstico por imagemRESUMO
The study objectives were to investigate the relationship between early exposure to genistein and obesity in young adulthood and to evaluate changes in reproductive health during puberty and adulthood following in utero exposure to genistein. Thirty-two female rats were randomized into four groups; low dose 400 mg genistein/kg diet group (LG), mid-dose 1200 mg genistein/kg diet group (MG), high dose 3600 mg genistein/kg diet group (HG), and control group without genistein diet (CON). Rats were fed genistein at the beginning of pregnancy along with a high-fat diet. Pups were sacrificed at week 4 and week 8 after birth. High performance liquid chromatography (HPLC) results showed a correlation between maternal genistein intake and genistein concentration in pups' plasma. Compared to CON, body weight reduced significantly in male HG group at week 8. No statistical differences were found in plasma estradiol (E2), testosterone (T), interleukin (IL)-6, and C-reactive protein (CRP) levels with early genistein exposure. Furthermore, uterine histopathology showed notable changes in groups HG and MG compared with CON at week 4 and week 8. In conclusion, maternal genistein supplement could reduce body weight in male pups and alter uterine histopathology in female pups.