OBJECTIVES: We propose a multimodal model integrating social media text and image data for automated assessment of psychological stress in college students to support the development of intelligent mental health services in higher education institutions. METHODS: Based on deep learning technology, we designed an evaluation framework comprising a text sentiment modeling module, an image sentiment modeling module, and a multimodal fusion prediction module. Text sentiment features were extracted using Bi-LSTM, and image semantic cues were extracted via U-Net. A feature concatenation strategy was used to enable cross-modal semantic collaboration to achieve automatic identification of 3 psychological stress levels: mild, moderate, and severe. We constructed a multimodal annotated dataset using social platform data from 1577 students across multiple universities in Guangdong Province. After data cleaning, 252 samples were randomly selected for model training and testing. RESULTS: In the 3-classification task, the model demonstrated outstanding performance on the test set, and achieved an accuracy of 92.86% and an F1 score of 0.9276, exhibiting excellent stability and consistency. Confusion matrix analysis further revealed the model's ability to effectively distinguish between different pressure levels. CONCLUSIONS The multimodal psychological stress assessment model developed in this study effectively integrates unstructured social behavior data to enhance the scientific rigor and practical applicability of psychological state recognition, and thus provides support for developing intelligent psychological service systems.
Humans ; Stress, Psychological/diagnosis* ; Students/psychology* ; Universities ; Social Media ; Deep Learning

Objective To investigate the awaken effect of naloxon on dexmedetomidine anesthetized mice and its mechanism. Methods Thirty Kunming mice of clean grade were randomly divided into 3 groups which included NAL group (Naloxon group), ATI group(Atipamezole group)and NS group (Normal Saline group). All groups were given dexme?detomidine 1 mg·kg-1 intraperitoneally. Naloxon 2 mg·kg-1, atipamezole 2 mg·kg-1 and normal saline 10 mL·kg-1 were ran?domly given intraperitoneally to the NAL, ATI and NS group respectively 90 minutes after dexmedetomidine administration. At timepoints prior to dexmedetomidine administration and 5, 15, 30, 60, 90, 95, 105, 120, 180 minutes after it, the sedative and analgesic effects besides recovery time (based on restore of righting reflex loss) were assessed. Results Sedation and analgesia effects became apparent within 5 minutes, and peaked at approximately 60 minutes then spontaneously recovered at 180 minutes after injection of dexmedetomidine. The sedative and analgesic effects were reduced in both ATI and NAL groups. Compared with ATI group, the sedation scores were higher at 95, 105 and 120 minutes after dexmedetomidine admin?istration than those in NAL group (P<0.05) but the scores were not statistically significant at 180 minutes between these two groups. Compared with NS group, the sedation scores were lower at time points of 95, 105, 120 and 180 minutes than those in NAL group (P>0.05). The analgesic scores were not statistically significant at time points of 95, 105, 120 and 180 min?utes between NAL group and ATI group, but they were lower in NAL group compared with NS group at timepoints of 95, 105 and 120 minutes (P>0.05). The recovery time in ATI and NAL group were shorter than that in NS group (F=1 793.368, P<0.05), but it showed no statistical difference between ATI group and NAL group (P>0.05). Conclusion Naloxone had a certain awaken effect on dexmedetomidine anesthetized mice.

OBJECTIVE:To analyze the relative factors of ADR in our hospital.METHODS:A total of 185 ADR case reports collected in 2006 were analyzed statistically.RESULTS:Of the total 185 cases,33(17.8%) were reported by physicians,12(6.5%) were new and severe ADR,62.7% were reports from pharmacy department,infusion room and department of derma-tology.For a few departments,the ADR case reports were zero.121(65.4%) were caused by anti-infective drugs;138(74.60%) were caused by intravenous drip and 108(58.4%) manifested as lesions of skin and its appendages.CONCLUSION:The incidence of ADR is related to multi-factors;however,rational use of drugs is the key conducive to the reduction of ADR incidence.