Objective:To develop and validate an artificial intelligence (AI) diagnostic model for coronary artery disease based on facial photos.Methods:This study was a cross-sectional study. Patients who were scheduled to undergo coronary angiography (CAG) at Beijing Anzhen Hospital and Beijing Daxing Hospital from August 2022 to November 2023 were included consecutively. Before CAG, facial photos were collected (including four angles: frontal view, left and right 60° profile, and top of the head). Photo datasets were randomly divided into a training set, a validation set (70%), and a testing set (30%). The model was constructed using Masked Autoencoder (MAE) and Vision Transformer (ViT) architectures. Firstly, the model base was pre-training using 2 million facial photos obtained from the publicly available VGGFace dataset, and fine-tuned by the training and validation sets; the model was validated in the test set. In addition, the ResNet architecture was used to process the dataset, and its outputs were compared with those of the models based on MAE and ViT. In the test set, the area under the operating characteristic curve ( AUC) of the AI model was calculated using CAG results as the gold standard. Results:A total of 5 974 participants aged 61 (54, 67) years were included, including 4 179 males (70.0%), with a total of 84 964 facial photos. There were 79 140 facial photos in the training and validation sets, with 3 822 patients with coronary artery disease; there were 5 824 facial photos in the test set, with 239 patients with coronary artery disease. The AUC value of the MAE and ViT model initialized with pre-training model weights was 0.841 and 0.824, respectively. The AUC of the ResNet model initialized with random weights was 0.810, while the AUC of the ResNet model initialized with pre-training model weights was 0.816. Conclusion:The AI model based on facial photos showes good diagnostic performance for coronary artery disease and holds promise for further application in early diagnosis.

Objective To analyze result of the external quality assessment for laboratories of toxicological pathology diagnosis in organizations in China. Methods A total of 86 organizations that participated in the 2020-2021 external quality assessment in laboratory of toxicological pathology diagnosis (hereinafter referred to as "reference units") were selected as research subjects using convenient sampling method, and the assessment results were analyzed. Results The median of total score was 92, and the 0-100 percentiles were 64-100 in these 86 reference units. Among these reference units, 76 were rated as excellent, 10 as qualified, with the excellent and the qualified rate of 88.4% and 11.6%, respectively. No reference unit was rated as unqualified. The rates of excellence of the reference units in public health institutions, pharmaceutical research institutions, drug safety evaluation centers and testing companies were 95.7%, 84.2%, 85.7% and 86.7%, and the qualified rates were 4.3%, 15.8%, 14.3% and 13.3%, respectively. The distribution of excellence and qualification among the four types of reference units showed no statistical difference (P>0.05). The distribution of sample scores according to the three grades of poor, good, and excellent were 4.9%, 20.7%, and 74.5% in public health institutions, 8.6%, 23.7%, and 67.8% in pharmaceutical research institutions, 12.5%, 25.0%, and 62.5% in drug safety evaluation centers, and 5.4%, 17.5%, and 77.1% in testing companies. The proportion of excellence unit in public health institutions was higher than that in pharmaceutical research institutions (P<0.05). Conclusion The overall toxicological pathology diagnostic capabilities in China are good, and various types of reference units demonstrate comparable technical capabilities. However, there is a need for standardization of diagnostic terminology.

Interleukin-27 (IL-27), a heterodimeric cytokine, plays a protective role in diabetes. Ghrelin, a gastric hormone, provides a hunger signal to the central nervous system to stimulate food intake. The relationship between IL-27 and ghrelin is still unexplored. Here we investigated that signal transducer and activator of transcription 3 (STAT3)-mechanistic target of rapamycin (mTOR) signaling mediates the suppression of ghrelin induced by IL-27. Co-localization of interleukin 27 receptor subunit alpha (WSX-1) and ghrelin was observed in mouse and human gastric mucosa. Intracerebroventricular injection of IL-27 markedly suppressed ghrelin synthesis and secretion while stimulating STAT3-mTOR signaling in both C57BL/6J mice and high-fat diet-induced-obese mice. IL-27 inhibited the production of ghrelin in mHypoE-N42 cells. Inhibition of mTOR activity induced by siRNA or rapamycin blocked the suppression of ghrelin production induced by IL-27 in mHypoE-N42 cells. siRNA also abolished the inhibitory effect of IL-27 on ghrelin. IL-27 increased the interaction between STAT3 and mTOR in mHypoE-N42 cells. In conclusion, IL-27 suppresses ghrelin production through the STAT3-mTOR dependent mechanism.