Objective:To realize the optimization and upgradation of the detection method of you only look once(YOLO)algorithm model based on the improved YOLO network on the ultrasound image for breast tumor.Methods:A total of 659 images of breast tumor of the Kaggle database were selected as the initially dataset,and the image annotation tool Labelimg was used to conduct pre-labeling for the detection targets in the images.According to a ratio as 7:3,629 images of the 659 images were divided into the train set and validation set,and the other 30 images were used as the test set.The convolutional block attention module(CBAM)and bidirectional feature pyramid network(BiFPN)were introduced into the original YOLO algorithm to underwent structural improvement,which was named as YOLOv5-BiFPN-CBAM.Both the train set and validation set were placed in original YOLO algorithm model and YOLOv5-BiFPN-CBAM model to conduct train,which included 200 rounds of iterative training.The obtained optimal weight files were used in the final test of test set.Results:After 200 rounds of iterative train for two kinds of models,the test results of validation set indicated that the mean values of average precision of two kinds of models were respectively 72.1%and 80.5%for all ultrasound images of breast tumor.The result,that the optimal weight file of improved model was tested by test set,indicated the test ability of improved model was significantly enhanced than that of original model for small target in image.Conclusion:Compared with the original YOLO algorithm model,the improved YOLO algorithm model has higher recognition capability for image,which also enhances precision and sensitivity in identifying small targets of ultrasound images of breast tumor.This model is helpful to improve the diagnostic efficiency in clinical practice for breast tumor.

Objective:To analyze the development status,frontiers and hotspots of radiomics research in the past five years from 2019 to 2023,and to provide theoretical reference and guidance for radiomics research in China.Methods:The relevant literature in the field of radiomics published in the core database of Web of Science(WOS)from January 1,2003 to August 10,2023 were searched.According to the inclusion and exclusion criteria,6,777 eligible literatures were screened and obtained,including 6,254 articles in the past five years from January 1,2019 to August 10,2023.Bibliometric methods were used to analyze the clustering of countries and regions,institutions,journals,authors,keywords and draw visual maps.Results:The 6,777 radiomics-related articles published between 2003 and 2023 were first published in 2011,and the number of papers tended to stabilize in 2018,and then the number showed a significant trend of increasing year by year.Among the 6,254 articles published from 2019 to 2023,China(3,564 articles),United States(1,164 articles),and Italy(530 articles)ranked the top 3 in terms of publication volume,with close cooperation between countries.General Electric of the United States published the most papers(448 articles),and the journal Frontiers in Oncology(704 papers)ranked first in terms of paper publication volume.From 2019 to 2023,the diseases of concern in the field of radiomics are rectal cancer,hepatocellular carcinoma,breast cancer,and lung cancer(especially non-small cell lung cancer).Conclusion:Although China ranks first in the number of national publications,the quality of research still needs to be improved.In the future,the research trend in the field of radiomics may be the diagnosis and differential diagnosis of various diseases,the prediction and evaluation of curative effect,the evaluation of tumor disease metastasis and the identification of gene phenotype based on radiomics combined with multiple imaging techniques.