OBJECTIVES: We propose a Contrastive Regional Attention and Prior Knowledge-Infused U-Transformer model (CRAKUT) to address the challenges of imbalanced text distribution, lack of contextual clinical knowledge, and cross-modal information transformation to enhance the quality of generated radiology reports. METHODS: The CRAKUT model comprises 3 key components, including an image encoder that utilizes common normal images from the dataset for extracting enhanced visual features, an external knowledge infuser that incorporates clinical prior knowledge, and a U-Transformer that facilitates cross-modal information conversion from vision to language. The contrastive regional attention in the image encoder was introduced to enhance the features of abnormal regions by emphasizing the difference between normal and abnormal semantic features. Additionally, the clinical prior knowledge infuser within the text encoder integrates clinical history and knowledge graphs generated by ChatGPT. Finally, the U-Transformer was utilized to connect the multi-modal encoder and the report decoder in a U-connection schema, and multiple types of information were used to fuse and obtain the final report. RESULTS: We evaluated the proposed CRAKUT model on two publicly available CXR datasets (IU-Xray and MIMIC-CXR). The experimental results showed that the CRAKUT model achieved a state-of-the-art performance on report generation with a BLEU-4 score of 0.159, a ROUGE-L score of 0.353, and a CIDEr score of 0.500 in MIMIC-CXR dataset; the model also had a METEOR score of 0.258 in IU-Xray dataset, outperforming all the comparison models. CONCLUSIONS The proposed method has great potential for application in clinical disease diagnoses and report generation.
Humans ; Radiology Information Systems ; Radiology

Objective:To investigate the application effect of bullet screen interaction in the teaching of internal medicine of traditional Chinese medicine.Methods:A randomized controlled trial was conducted, and 150 students were randomly divided into bullet screen interaction group and traditional interaction group and received teaching with bullet screen interaction and traditional interaction, respectively. Flanders Interaction Analysis System (FIAS), class rating, questionnaire survey, and performance test were used to evaluate the effect of classroom interaction and teaching achievement. SPSS 22.0 was used to perform the t-test and the chi-square test. Results:Compared with the traditional interaction group, the bullet screen interaction group had a significantly lower teacher's language ratio [(62.63±2.83)% vs. (71.05±3.19)%] and significantly higher student's language ratio [(32.68±2.62)% vs. (22.79±1.32)%], teacher's indirect/direct influence ratio [(96.63±9.59)% vs. (69.84±3.48)%], and teacher's positive/negative influence ratio [(122.89±6.43)% vs. (50.58±2.35)%]. Compared with the traditional interaction group, the bullet screen interaction group had significantly higher scores of teacher's emotional atmosphere (23.82±6.54 vs. 21.01±6.51), quality of classroom activities (25.67±5.51 vs. 22.56±11.95), and information transmission of teacher's classroom interactive activities (25.46±10.30 vs. 18.44±6.52). The questionnaire survey showed that compared with the traditional interaction group, the bullet screen interaction group had a significantly higher number of the students who selected excellent and good for student interest in classroom, dullness of classroom, and the mastery of classroom knowledge [104 (69.33%)/110 (73.33%)/106 (70.67%) vs. 72(48.00%)/74 (49.33%)/84(56.00%)], and the bullet screen interaction group had significantly higher scores of basic knowledge and case analysis than the traditional interaction group (84.30±4.13/78.53±7.21 vs. 79.26±5.67/72.56±4.22).Conclusions:The application of bullet screen interaction teaching in the teaching of internal medicine of traditional Chinese medicine can help to improve interactive effect and teaching achievement.

An accurate, fast and sensitive method based on high performance liquid chromatography was established for the simultaneous determination of eleven ultra-violet absorbents in cosmetics. Eleven ultra-violet absorbents were baseline separated on an Acclaim C18 column within 11. 5 min using acetonitrile-0. 1%formic acid in water ( V/V) mobile phase, and detected at 361 nm with UV detection. Under the optimized conditions, the relative standard derivations ( RSDs) of the eleven ultra-violet absorbents were all less than 0. 1% for retention time, and less than 1. 2% for peak area; good linearity was obtained from 5 to and 500 mg/L with the correlation coefficients of above 0 . 9990 for these analytes; the recoveries spiked in a cosmetic sample were in the range of 77% -116%. Benzophenone-3, butylmethoxydibenzoylmethane, ethylhexylsalicylatec and homosalate were found in the detected cosmetic samples, and the concentration of homosalate was the highest. The results indicated that this method had potential for applications due to its convenience, accuracy and sensitivity. Oxybenzone, butylmethoxydibenzoylmethane, 2-ethylhexyl salicylatec and homosalate were found in the detected cosmetic samples, and the concentration of homosalate was the highest.