Comparison of ischemic stroke diagnosis models based on machine learning

Autophagy and machine learning: Unanswered questions

Autophagy is a critical conserved cellular process in maintaining cellular homeostasis by clearing and recycling damaged organelles and intracellular components in lysosomes and vacuoles. Autophagy plays a vital role in cell survival, bioenergetic homeostasis, organism development, and cell death regulation. Malfunctions in autophagy are associated with various human diseases and health disorders, such as cancers and neurodegenerative diseases. Significant effort has been devoted to autophagy-related research in the context of genes, proteins, diagnosis, etc. In recent years, there has been a surge of studies utilizing state of the art machine learning (ML) tools to analyze and understand the roles of autophagy in various biological processes. We taxonomize ML techniques that are applicable in an autophagy context, comprehensively review existing efforts being taken in this direction, and outline principles to consider in a biomedical context. In recognition of recent groundbreaking advances in the deep-learning community, we discuss new opportunities in interdisciplinary collaborations and seek to engage autophagy and computer science researchers to promote autophagy research with joint efforts.

Discovery of biomarkers in the psoriasis through machine learning and dynamic immune infiltration in three types of skin lesions

Introduction

Psoriasis is a chronic skin disease characterized by unique scaling plaques. However, during the acute phase, psoriatic lesions exhibit eczematous changes, making them difficult to distinguish from atopic dermatitis, which poses challenges for the selection of biological agents. This study aimed to identify potential diagnostic genes in psoriatic lesions and investigate their clinical significance.

Methods

GSE182740 datasets from the GEO database were analyzed for differential analysis; machine learning algorithms (SVM-RFE and LASSO regression models) are used to screen for diagnostic markers; CIBERSORTx is used to determine the dynamic changes of 22 different immune cell components in normal skin lesions, psoriatic non-lesional skin, and psoriatic lesional skin, as well as the expression of the diagnostic genes in 10 major immune cells, and real-time quantitative polymerase chain reaction (RT-qPCR) and immunohistochemistry are used to validate results.

Results

We obtained 580 differentially expressed genes (DEGs) in the skin lesion and non-lesion of psoriasis patients, 813 DEGs in mixed patients between non-lesions and lesions, and 96 DEGs in the skin lesion and non-lesion of atopic dermatitis, respectively. Then 144 specific DEGs in psoriasis via a Veen diagram were identified. Ultimately, UGGT1, CCNE1, MMP9 and ARHGEF28 are identified for potential diagnostic genes from these 144 specific DEGs. The value of the selected diagnostic genes was verified by receiver operating characteristic (ROC) curves with expanded samples. The the area under the ROC curve (AUC) exceeded 0.7 for the four diagnosis genes. RT-qPCR results showed that compared to normal human epidermis, the expression of UGGT1, CCNE1, and MMP9 was significantly increased in patients with psoriasis, while ARHGEF28 expression was significantly decreased. Notably, the results of CIBERSORTx showed that CCNE1 was highly expressed in CD4+ T cells and neutrophils, ARHGEF28 was also expressed in mast cells. Additionally, CCNE1 was strongly correlated with IL-17/CXCL8/9/10 and CCL20. Immunohistochemical results showed increased nuclear expression of CCNE1 in psoriatic epidermal cells relative to normal.

Conclusion

Based on the performance of the four genes in ROC curves and their expression in immune cells from patients with psoriasis, we suggest that CCNE1 possess higher diagnostic value.

Knowledge structure and global trends of machine learning in stroke over the past decade: A scientometric analysis

Objective

Machine learning (ML) models have been widely applied in stroke prediction, diagnosis, treatment, and prognosis assessment. We aimed to conduct a comprehensive scientometrics analysis of studies related to ML in stroke and reveal its current status, knowledge structure, and global trends.

Methods

All documents related to ML in stroke were retrieved from the Web of Science database on March 15, 2023. We refined the documents by including only original articles and reviews in the English language. The literature published over the past decade was imported into scientometrics software for influence detection and collaborative network analysis.

Results

2389 related publications were included. The annual publication outputs demonstrated explosive growth, with an average growth rate of 63.99 %. Among the 90 countries/regions involved, the United States (729 articles) and China (636 articles) were the most productive countries. Frontiers in Neurology was the most prolific journal with 94 articles. 234 highly cited articles, each with more than 31 citations, were detected. Keyword analysis revealed a total of 5333 keywords, with a predominant focus on the application of ML models in the early diagnosis, classification, and prediction of "acute ischemic stroke" and "atrial fibrillation-related stroke". The keyword "classification" had the first and longest burst, spanning from 2013 to 2018. 'Upport vector machine' got the strongest burst strength with 6.2. Keywords such as 'mechanical thrombectomy', 'expression', and 'prognosis' experienced bursts in 2022 and have continued to be prominent.

Conclusion

The applications of ML in stroke are increasingly diverse and extensive, with researchers showing growing interest over the past decade. However, the clinical application of ML in stroke is still in its early stages, and several limitations and challenges need to be addressed for its widespread adoption in clinical practice.

Analysis and Regulatory Mechanisms of Platelet-Related Genes in Patients with Ischemic Stroke

It was found that ischemic stroke (IS) was associated with abnormal platelet activity and thrombosis. However, the potential significance of platelet-related genes (PRGs) in IS still needs to be more thorough. This study extracted IS-related transcriptome datasets from the Gene Expression Omnibus (GEO) database. The target genes were obtained by intersecting the differentially expressed genes (DEGs), the module genes related to IS, and PRGs, where the key genes of IS were screened by two machine learning algorithms. The key genes-based diagnostic model was constructed. Gene set enrichment analysis (GSEA) and the immune microenvironment analyses were analyzed targeting key genes in IS. The co-expression, TF-mRNA, and competitive endogenous RNAs (ceRNA) regulatory networks were constructed to reveal the potential regulation of key genes. Potential drugs targeting key genes were predicted as well. Totals of eight target genes were obtained and were associated with immune-related functions. Four platelet-related key genes were acquired, which were related to immunity and energy metabolism. The abnormal expressions of DOCK8, GIMAP5, ICOS were determined by the quantitative real-time polymerase chain reaction (qRT-PCR), and the significant correlations among these key genes were identified. Notably, hsa-miR-17-3p, hsa-miR-3158-3p, hsa-miR-423-3p, and hsa-miR-193a-8p could regulate all key genes at the same time. In addition, Caffeine, Carboplatin, and Vopratelimab were the targeted drugs of these key genes. This study identified four platelet-related key genes of IS, which might help to deepen the understanding of the role of platelet-related genes in the molecular mechanism of IS.