The differentiation between benign and malignant thyroid nodules is pivotal in clinical practice. However, current diagnostic systems face challenges, such as heterogeneity in ultrasound interpretation, the invasiveness of fine-needle aspiration cytology and indeterminate of results. Artificial intelligence (AI) has evolved from traditional machine learning to deep learning, enabling advances in image recognition, feature extraction, and multimodal data integration. Nevertheless, AI applications remain constrained by limitations, including the insufficiency of high-quality annotated data and inadequate cross-center generalizability. Future progress will focus on algorithm refinement, multimodal data fusion, and multicenter clinical validation to enhance model generalizability and diagnostic accuracy.

The differentiation between benign and malignant thyroid nodules is pivotal in clinical practice. However, current diagnostic systems face challenges, such as heterogeneity in ultrasound interpretation, the invasiveness of fine-needle aspiration cytology and indeterminate of results. Artificial intelligence (AI) has evolved from traditional machine learning to deep learning, enabling advances in image recognition, feature extraction, and multimodal data integration. Nevertheless, AI applications remain constrained by limitations, including the insufficiency of high-quality annotated data and inadequate cross-center generalizability. Future progress will focus on algorithm refinement, multimodal data fusion, and multicenter clinical validation to enhance model generalizability and diagnostic accuracy.

Objective Explore the comprehensive emotion adjustment pattern that combines non-contact physiological and psychological detection methods in fasting and low metabolism scenarios.This study aims to verify the accuracy of non-contact physiological and psychological detection algorithms and evaluate the effectiveness of multimodal emotion adjustment schemes for addressing negative emotional states such as depression and anxiety.Methods Deploy non-contact physiological and psychological detection algorithms and emotion adjustment plans to build a multimodal emotion adjustment system.Collect physiological and psychological data from volunteers participating in the 15-days complete fasting human low metabolism experiment of"Green Star Travel Ⅷ".Utilize finger clip oximeters and scales to verify the accuracy of existing non-contact physiological and psychological methods within the system.Design an emotion adjustment experiment featuring four groups:sound adjustment,acupoints adjustment,magnetism adjustment,and combination adjustment.Compare the volunteers'scale scores before and after the adjustments to verify the effectiveness of the system's emotion adjustment capabilities.Results The experimental results demonstrate that the average difference in the Bland-Altman plot for the non-contact heart rate detection model is ﹣0.497 bpm,with 95.3%of the error values falling within the 95%consistency interval.The non-contact psychological detection model achieved an accuracy rate of over 80%in identifying stress,anxiety,and depression,and an accuracy rate of over 70%in identifying fatigue and anger.Following emotion adjustment,the stress levels of the subjects significantly improved(P?

Objective To systematically evaluate the short-term outcomes of Da Vinci robotic surgical system and laparoscopy in pancreaticoduodenectomy.Methods Literatures were researched using PubMed,Embase,Medline,VIP database,Chinese Journal Fulltext Database from January,2013 to October,2016 with the key words including laparoscopic,robotic,Da Vinci,pancreaticoduodenectomy,腹腔锐,达芬奇机器人,胰十二指肠切除术”.The cohort studies about comparison of short-term outcomes of Da Vinci robotic surgical system and laparoscopy in pancreaticoduodenectomy were received and enrolled.Patients undergoing pancreaticoduodenectomy using Da Vinci robotic surgical system and laparoscopy were respectively allocated into the Da Vinci group and laparoscopy group.Two reviewers independently screened literatures,extracted data and assessed the risk of bias.Count data were described as odd ratio (OR) and 95％ confidence interval (CI).Measurement data were represented as weighted mean difference (WMD) and 95％CI.The heterogeneity of the studies was analyzed using the I2 test.Results Five retrospective cohort studies were enrolled in the Meta analysis,and the total sample size was 364 patients,including 159 in the Da Vinci group and 205 in the laparoscopy group.The results of Meta analysis showed that there were statistically significant differences in the rate,of conversion to open surgery,spleen-preserving rate,operation time and duration of postoperative hospital stay between Da Vinci group and laparoscopy group (OR=0.18,3.80,WMD =-37.54,-4.47,95％CI:0.05-0.60,2.01-7.18,-47.46--27.62,-6.70--2.24,P＜0.05).There were no statistically significant difference in the incidence of pancreatic fistula and overall incidence of complications between Da Vinci group and laparoscopy group (OR=0.95,0.55,95％CI:0.59-1.54,0.29-1.03,P＞0.05).Conclusions Da Vinci robotic surgical system and laparoscopy are safe and feasible in pancreaticoduodenectomy,with the same incidence of postoperative complications.Compared with laparoscopy,Da Vinci robotic surgical system can not only reduce the rate of conversion to open surgery,operation time and duration of postoperative hospital stay,but also increase spleen-preserving rate,meanwhile,it does not increase the incidence of postoperative complications.