Laboratory testing is of great value in the management of autoimmune disease. The results can help confirm a diagnosis, estimate disease severity, aid in assessing treatment effect. But the current autoimmunity laboratory system, including testing standards, quality control and supervision, does not match the national conditions well. As a result, the test reports are not mutual-recognized among laboratories. In the current background of precision medicine, with the advances of technology and the application of deep learning and artificial intelligence in the clinical laboratory field, the autoimmune laboratory has ushered in a new development trend of integration, automation and intelligence.

Antinuclear antibodies (ANA) testing is essential for the diagnosis, classification, and disease activity monitoring of systemic autoimmune rheumatic diseases. In recent years, with the enhancement of computing power and the innovation of algorithms, the newly hip branch, deep learning (DL), practically delivered all of the most stunning achievements and breakthroughs in artificial intelligence (AI) so far. The application of DL to visual tasks, known as computer vision, has revealed significant power within the medical image recognition. Indirect immunofluorescence on HEp-2 cells is the reference method for ANA testing, the results is interpreted manually by specialized physicians. ANA fluorescent pattern classification is based on image recognition, which has a broad prospect of combining with DL to realize automatic interpretation system. This paper reviews the recent research progress and challenges of DL in the field of ANA detection in order to provide references for the standardization of ANA testing in the future.

In recent years, machine learning has become a hot spot in various research fields. Using machine learning can realize the transformation from data driven to knowledge discovery, which is an important development direction of laboratory intelligence in the future. The application of machine learning in laboratory medicine has shown great potential, but t it also has many challenges and difficulties. The direction of our joint efforts is to promote the clinical transformation of machine learning technology, realize the practicality and industrialization in medical laboratories, and achieve the goal of assisting clinical decision-making as soon as possible.

Objective:To develop a prototype artificial intelligence immunofluorescence image recognition system for classification of antinuclear antibodies in order to meet the growing clinical requirements for an automatic readout and classification of immunof luorescence patterns for antinuclear antibody (ANA) images.Methods:Immunofluorescence images with positive results of ANA in Xinhua Hospital Affiliated to Shanghai Jiao Tong University School of Medicine from April 2020 to December 2021 were collected. Three senior technicians independently and in parallel interpreted the Immunofluorescence images to determine the ANA results. Then the images were labeled according to the ANA International Consensus on Fluorescence Patterns (ICAP) classification criteria. There were 7 labeled groups: Fine speckled, Coarse speckled, Homogeneous, nucleolar, Centromere, Nuclear dots and Nuclear envelope. Each group was randomly divided into training dataset and validation dataset at a ratio of 9∶1 by using random number table. On the deep learning framework PyTORCH 1.7, the convolutional neural network (CNN) training platform was constructed based on ResNet-34 image classification network, and the automatic ANA recognition system was established. After the model was established, the test set was set up separately, the judgment results of the model were output by ranking the prediction probability, with the results of the 2 senior technicians was taken as "golden standard". Parameters such as accuracy, precision, recall and F1-score were used as indicators to evaluate the performance of the model.Results:A total of 23138 immunofluorescence images were obtained after segmentation and annotation. A total of 7 models were trained, and the effects of different algorithms, image processing and enhancement methods on the model were compared. The ResNet-34 model with the highest accuracy andswas selected as the final model, with the classification accuracy of 93.31%, precision rate of 91%, and recall rate of 90% and F1-score of 91% in the test set. The overall coincidence rate between the model and manual interpretation was 90.05%, and the accuracy of recognition of nucleolus was the highest, with the coincidence rate reaching 100% in the test set.Conclusion:The current AI system developed based on deep learning of the ANA immunofluorescence images in the present study showed the ability to recognize ANA pattern, especially in the common, typical, simple pattern.

Inflammatory caspase-11 senses and is activated by intracellular lipopolysaccharide (LPS) leading to pyroptosis that has critical role in defensing against bacterial infection, whereas its excess activation under pathogenic circumstances may cause various inflammatory diseases. However, there are few known drugs that can control caspase-11 activation. We report here that scutellarin, a flavonoid from Erigeron breviscapus, acted as an inhibitor for caspase-11 activation in macrophages. Scutellarin dose-dependently inhibited intracellular LPS-induced release of caspase-11p26 (indicative of caspase-11 activation) and generation of N-terminal fragment of gasdermin D (GSDMD-NT), leading to reduced pyroptosis. It also suppressed the activation of non-canonical nucleotide-binding oligomerization domain-like receptor family pyrin domain containing 3 (NLRP3) inflammasome as evidenced by reduced apoptosis-associated speck-like protein containing a CARD (ASC) speck formation and decreased interleukin-1 beta (IL-1β) and caspase-1p10 secretion, whereas the NLRP3-specific inhibitor MCC950 only inhibited IL-1β and caspase-1p10 release and ASC speck formation but not pyroptosis. Scutellarin also suppressed LPS-induced caspase-11 activation and pyroptosis in RAW 264.7 cells lacking ASC expression. Moreover, scutellarin treatment increased Ser/Thr phosphorylation of caspase-11 at protein kinase A (PKA)-specific sites, and its inhibitory action on caspase-11 activation was largely abrogated by PKA inhibitor H89 or by adenylyl cyclase inhibitor MDL12330A. Collectively, our data indicate that scutellarin inhibited caspase-11 activation and pyroptosis in macrophages at least partly via regulating the PKA signaling pathway.