ABSTRACT
Objective:To construct an intelligent foreign bodies detection model based on Faster R-convolutional neural network in posterior-anterior chest X-ray and evaluate the performance of the model.Methods:Totally 5 567 adult posterior-anterior DR chest radiographs from Zhejiang Provincial People′s Hospital and Chun′an County People′s Hospital from June 2019 to March 2020, with 4 247 foreign body-containing chest radiographs were analyzed retrospectively. All data were randomly divided into training set (2 911 foreign body-containing), validation set ( n=1 456, 733 foreign body-containing, 723 free of foreign body) and testing set ( n=1 200, 603 foreign body-containing, 597 free of foreign body). The reference gold standard was set as the results of each chest radiography with foreign body annotated by two radiology residents and reviewed and corrected by a senior radiographer. The receiver operating characteristic (ROC) curve and the area under the curve were used to analyze the efficiency of the deep learning model to distinguish the presence or absence of foreign bodies on chest radiography in the testing set. The precision-recall curve and mean precision (mAP) were used to analyze the stability of the model at different levels. Finally, the influence of different locations, patient gender, and patient age on the foreign body recall of the deep learning model were analyzed. Results:In the testing set, the sensitivity of the deep learning model in diagnosing whether chest radiograph contained foreign bodies was 93.2%(562/603), the specificity was 92.6%(553/597), and the F1 score was 0.94. The area under the ROC curve was 0.97, and the mAP value was 0.69. For foreign bodies in different locations, the recall rates of foreign bodies in lung field and outside lung field were 91.2% (674/739) and 89.0% (1 411/1 585), respectively. For different genders, the recall rates for male and female foreign body detection were 87.3% (337/386) and 90.0%(1 745/1 938), respectively. For different age ranges, the recall rate of foreign body detection was 92.5% (1 041/1 126) for 18-38 years old, 89.7%(505/563) for 39-58 years old, 83.5%(335/401) for 59-78 years old and 85.9% (201/234) for patients ≥79 years old.Conclusion:The constructed deep learning-based foreign body detection model for adult posterior-anterior chest X-ray provides high sensitivity and stability, which can identify foreign bodies in chest radiography quickly and accurately.
ABSTRACT
BACKGROUND:Gastric cancer mesenchyal stem cel s from clinical stomach cancer specimens and tumorigenic tissues in nude mice are similar to the bone marrow mesenchymal stem cel s in biological characteristics, which have been proved to be an important component of tumor microenvironment to promote tumor growth. It is speculated that biological characteristic of bone marrow mesenchymal stem cel s may change in stomach cancer microenvironment. OBJECTIVE:To observe the effect of stomach cancer microenvironment on morphology and proliferation of bone marrow mesenchymal stem cel s and expressions of CD34 and CD44. METHODS:Rat bone marrow mesenchymal stem cel s were cultured alone as control group. In the test group, rat bone marrow mesenchymal stem cel s were co-cultured with human stomach cancer BGC-823 cel s using Transwel chamber assay to establish the stomach cancer microenvironment. Then, cel morphology, proliferation, cel cycle and CD34, CD44 expressions were observed and detected using inverted phase contrast microscope, MTT assay, and flow cytometry, respectively. RESULTS AND CONCLUSION:In the test group, bone marrow mesenchymal stem cel s were similar to human stomach cancer cel s BGC-823 that arranged disorderly and irregularly, were interconnected loosely, became thinner and longer, and grew in clusters with smal er nuclei. The cel proportion in G 1 phase significantly decreased, but that in S and G 2/M phases significantly increased (P<0.01, P<0.05). The positive rate of CD44 significantly declined, and the CD34 expression significantly raised (P<0.01). In conclusion, stomach cancer microenvironment by non-contact co-culture with BCG-823 cel s has an obvious effect on the morphology, proliferation and surface antigens expressions of bone marrow mesenchymal stem cel s that wil tend to be malignant gastric cancer cel s.