Machine Learning for Predicting Stillbirth: A Systematic Review.

Stillbirth is a major global issue, with over 5 million cases each year. The multifactorial nature of stillbirth makes it difficult to predict. Artificial intelligence (AI) and machine learning (ML) have the potential to enhance clinical decision-making and enable precise assessments. This study reviewed the literature on predictive ML models for stillbirth highlighting input characteristics, performance metrics, and validation. The PubMed, Cochrane, and Web of Science databases were searched for studies using AI to develop predictive models for stillbirth. Findings were analyzed qualitatively using narrative synthesis and graphics. Risk of bias and the applicability of the studies were assessed using PROBAST. Model design and performance were discussed. Eight studies involving 14,840,654 women with gestational ages ranging from 20 weeks to full term were included in the qualitative analysis. Most studies used neural networks, random forests, and logistic regression algorithms. The number of predictive features varied from 14 to 53. Only 50% of studies validated the models. Cross-validation was commonly employed, and only 25% of studies performed external validation. All studies reported area under the curve as a performance metric (range 0.54-0.9), and five studies reported sensitivity (range, 60- 90%) and specificity (range, 64 - 93.3%). A stacked ensemble model that analyzed 53 features performed better than other models (AUC = 0.9; sensitivity and specificity > 85%). Available ML models can attain a considerable degree of accuracy for prediction of stillbirth; however, these models require further development before they can be applied in a clinical setting.

Remodeling of T-cell mitochondrial metabolism to treat autoimmune diseases.

T cells are key drivers of the pathogenesis of autoimmune diseases by producing cytokines, stimulating the generation of autoantibodies, and mediating tissue and cell damage. Distinct mitochondrial metabolic pathways govern the direction of T-cell differentiation and function and rely on specific nutrients and metabolic enzymes. Metabolic substrate uptake and mitochondrial metabolism form the foundational elements for T-cell activation, proliferation, differentiation, and effector function, contributing to the dynamic interplay between immunological signals and mitochondrial metabolism in coordinating adaptive immunity. Perturbations in substrate availability and enzyme activity may impair T-cell immunosuppressive function, fostering autoreactive responses and disrupting immune homeostasis, ultimately contributing to autoimmune disease pathogenesis. A growing body of studies has explored how metabolic processes regulate the function of diverse T-cell subsets in autoimmune diseases such as systemic lupus erythematosus (SLE), multiple sclerosis (MS), autoimmune hepatitis (AIH), inflammatory bowel disease (IBD), and psoriasis. This review describes the coordination of T-cell biology by mitochondrial metabolism, including the electron transport chain (ETC), oxidative phosphorylation, amino acid metabolism, fatty acid metabolism, and one­carbon metabolism. This study elucidated the intricate crosstalk between mitochondrial metabolic programs, signal transduction pathways, and transcription factors. This review summarizes potential therapeutic targets for T-cell mitochondrial metabolism and signaling in autoimmune diseases, providing insights for future studies.

Gene polymorphisms of an interleukin-23 receptor associated with susceptibility to rheumatoid arthritis in the Western Chinese Han population.

Rheumatoid arthritis (RA) is a chronic inflammatory disease which is closely related to genetic background. Single-nucleotide polymorphisms (SNPs) have been found to play an important role in the development of RA. This study intends to investigate the links between gene polymorphisms in the interleukin-23 receptor (IL23R) and interleukin 17A (IL17A) and susceptibility to RA in the Western Chinese Han population. Four SNPs (rs6693831 T > C, rs1884444 G > T, and rs7517847 T > G in IL23R gene, and rs2275913 G > A in IL17A gene) were genotyped in 246 RA patients and 362 healthy controls by high resolution melting analysis. The comparative analyses among genotype distributions, clinical indicators, and IL-17A and IL-23R levels in RA patients were also performed. The study revealed that the SNP rs6693831 and rs1884444 of IL23R had a significant association with RA susceptibility. The frequencies of rs6693831 genotype CC and allele C were significantly higher in the RA group and associated with higher RA risk compared with genotype TT and allele T (OR = 7.797, 95% confidence interval [CI] = 4.072-14.932 and OR = 5.984, 95%CI = 3.190-11.224, respectively). The TT genotype of rs1884444 appeared to decrease the RA risk compared with the GG genotype (OR = .251, 95%CI = .118-.536). The genotype CC and allele C of rs6693831 and the genotype GG and allele G of rs1884444 may be risk factors for RA. IL23R gene polymorphisms may be involved in the risk of RA susceptibility in the Western Chinese Han population.

Genetic polymorphisms of rs73620203 in the transforming growth interacting factor gene associated with rheumatoid arthritis in a Chinese population.

OBJECTIVE: To explore the association of single nucleotide polymorphisms (SNPs) in the transforming growth interacting factor (TGIF) gene with bone metabolism markers and rheumatoid arthritis (RA) susceptibility. METHODS: Three SNPs were genotyped in 155 RA patients and 168 healthy controls using high-resolution melting (HRM) analysis. The serum levels of osteocalcin, bone alkaline phosphatase (BALP), and ß type I collagen-crosslinked C telopeptide (ß-CTX) were detected using electrochemical luminescence in 108 patients randomly selected from the RA group. RESULTS: Genotype and allele frequency analysis showed that rs73620203 was associated with bone erosion in RA (P = 0.012 and P = 0.003, respectively), and individuals carrying the T allele for rs73620203 showed a decreased RA risk (OR = 0.59, 95% CI = 0.42-0.84; P = 0.003). In sex-specific analysis, the rs73620203 polymorphism was associated with susceptibility to RA in women (P = 0.022 and P = 0.006, respectively). In addition, RA patients with three genotypes at the rs73620203 locus showed significant differences in serum osteocalcin and BALP (P = 0.006 and P = 0.037, respectively). Haplotype analysis revealed that the haploid ATG and GCA frequencies were significantly lower in the RA group (P = 0.036, OR = 0.693; P = 0.002, OR = 0.189, respectively), while the haploid ACA frequency of the RA group was enhanced (P < 0.01, OR = 5.058). CONCLUSION: Our study provides the first evidence that rs73620203 is associated with RA susceptibility and the relationship between TGIF gene SNPs and the regulation of bone metabolism in RA patients.

Differential expression and correlation of immunoregulation related piRNA in rheumatoid arthritis.

Background: Although PIWI-interacting RNAs (piRNAs) have recently been associated with germline development and many human diseases, their expression pattern and relationship in autoimmune diseases remain indistinct. This study aimed to investigate the presence and correlation of piRNAs in rheumatoid arthritis (RA). Methods: We first analyzed the expression profile of piRNAs using small RNA sequencing in peripheral leukocytes of three new-onset untreated RA patients and three healthy controls (HCs). We then selected piRNAs related to immunoregulation by bioinformatics analysis and verified them in 42 new-onset RA patients and 81 HCs by RT-qPCR. Furthermore, a receiver operating characteristic curve was generated to quantify the diagnostic performance of these piRNAs. A correlation analysis was conducted to observe the link between piRNA expression and RA clinical characteristics. Results: A total of 15 upregulated and 9 downregulated piRNAs among 1,565 known piRNAs were identified in peripheral leukocytes of RA patients. Dysregulated piRNAs were enriched in numerous pathways related to immunity. After selection and validation, two immunoregulation piRNAs (piR-hsa-27620 and piR-hsa-27124) were significantly elevated in RA patients and have good abilities to distinguish patients from controls, which have the potential to serve as biomarkers. PIWI and other proteins implicated in the piRNA pathway were also associated with RA.

Application of Ultrasound Elastography in Assessing Portal Hypertension.

Portal hypertension is a common manifestation in late-to-end-stage liver diseases and can cause severe complications such as ascites, hepatic encephalopathy, etc. However, an early diagnosis of portal hypertension is often difficult as it can be asymptomatic. Though the gold standard to diagnose portal hypertension is hepatic vein catheterization, ultrasound elastography is regarded as a noninvasive alternative that can be used to accurately predict portal hypertension and a few further complications such as gastro-esophageal varices. Since ultrasound elastography is available in most medical centers, and is cheaper and noninvasive, studying its function in predicting portal hypertension is of paramount importance. Therefore, this review generalized the results of recently published articles in order to establish the indicators that were related to diagnostic and prediction efficiency. Our study found that various technologies of ultrasound elastography could be used to predict portal hypertension with satisfactory diagnostic sensitivity, specificity, accuracy, and AUC. Meanwhile, we also recognized similar diagnostic efficiency of ultrasound elastography in gastro-esophageal varices.

Analysis of inflammation-related microRNA expression in patients with ankylosing spondylitis.

Ankylosing spondylitis (AS) is a complex genetic disease characterized by axial skeletal inflammation. Available scientific evidence suggests that a relationship may exist between miRNA expression levels and the pathogenesis of AS. This study investigated the clinical diagnostic value of miR-146a, miR-15a, miR-20a, miR-125a-3p, miR-125a-5p, miR-125b-5p, miR-148a, miR-149a, miR-499, and miR-155a in AS. A total of 44 AS patients and 56 healthy controls (HCs) were included in the study. MiRNA expression levels were detected using fluorescence quantitative PCR (qPCR). Results showed that the expression levels of miR-146a, miR-125a-3p, miR-125a-5p, miR-125b-5p, and miR-155a decreased, whereas miR-499a expression increased significantly in AS patients compared to that in the controls. Logistic regression analysis with receiver operating characteristic (ROC) curves showed that combined miR-146a/miR-125a-5p/miR-125b-5p/miR-499a/miR-155a (area under curve [AUC] = 0.824, 95% confidence interval [CI] = 0.727-0.921) had high sensitivity and specificity for AS diagnosis. C-reactive protein (CRP) levels were positively correlated with the expression of miR-125a-5p (rs = 0.438, p = 0.005) and miR-155a (rs = 0.414, p = 0.006), which indicates that miR-125a-5p and miR-155a can perhaps aggravate AS-induced inflammation. Our findings suggest the association of miR-125a-5p and miR-155a with disease activity in AS patients. Furthermore, miR-146a, miR-125a-5p, miR-125b-5p, miR-499a, and miR-155a could have potential diagnostic value in AS.