Using AI to Diagnose Ischemic Stroke Early
Author Information
Author(s): Lin Ling, Guo Chunmao, Jin Hanna, Huang Haixiong, Luo Fan, Wang Ying, Li Dongqi, Zhang Yuanxin, Xu Yuqian, Zhu Chanyan, Zeng Fengshan, He Huahua, Chen Jie, Zhang Wei, Yu Wenlin
Primary Institution: Huizhou Hospital of Guangzhou University of Chinese Medicine
Hypothesis
Can a multi-omics approach improve the early diagnosis and understanding of immune infiltration in ischemic stroke?
Conclusion
The study identifies specific genes and immune cells involved in ischemic stroke, highlighting their potential as targets for understanding and treating the disease.
Supporting Evidence
- 26 significant differentially expressed genes (DEGs) associated with ischemic stroke were identified.
- The area under the ROC curve (AUC) for the classification model was found to be near 1, indicating high accuracy.
- Immune-related networks were elevated in ischemic stroke patients compared to healthy controls.
Takeaway
Researchers found important genes that can help doctors diagnose strokes faster and understand how the immune system affects strokes.
Methodology
The study used gene expression data from GEO databases to identify differentially expressed genes and employed Random Forest and artificial neural network models for classification.
Limitations
The study's practical application in rapid diagnosis may be limited by the time required for RNA extraction and quantification of DEGs.
Participant Demographics
The study included 39 ischemic stroke patients and 24 healthy controls in the discovery cohort, and 20 ischemic stroke patients and 20 matched controls in another cohort.
Statistical Information
P-Value
<0.05
Statistical Significance
p<0.05
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website