Integrated transcriptomics and machine learning reveal diagnostic biomarkers and immune–stromal remodeling in ischemic heart failure

Background Ischemic heart failure (IHF) is a major cause of cardiovascular morbidity worldwide, characterized by complex tissue remodeling and inflammation. However, reliable molecular biomarkers for early diagnosis and a systematic understanding of the associated immune–stromal microenvironment remain limited. Identifying specific transcriptomic signatures may enhance diagnostic precision and reveal novel therapeutic targets. Methods An integrative transcriptomic analysis was performed utilizing IHF datasets from the Gene Expression Omnibus (GEO). Differential expression analysis and Weighted Gene Co-expression Network Analysis (WGCNA) were employed to identify key disease-associated modules. To construct a robust diagnostic model, candidate features were screened using the intersection of four complementary machine learning algorithms: Least Absolute Shrinkage and Selection Operator (LASSO), Random Forest (RF), Support Vector Machine-Recursive Feature Elimination (SVM-RFE), and eXtreme Gradient Boosting (XGBoost). The immune and stromal landscape of IHF was comprehensively characterized using a hybrid approach combining MCP-counter and ssGSEA algorithms to quantify cell-type–specific infiltration patterns. Results Through the integration of machine learning strategies, a robust 6-gene diagnostic signature was identified, comprising FCN3, OGN, ITPK1, HMOX2, MTCH1, and HMGN2. Immune deconvolution analysis revealed pronounced remodeling of the IHF microenvironment, characterized by significantly elevated infiltration of Endothelial cells, Macrophages, Neutrophils, and Natural killer cells, indicating a pro-inflammatory and angiogenic phenotype. Conclusion This study identifies a novel and robust 6-gene diagnostic signature for Ischemic heart failure through a multi-algorithm machine learning framework. These biomarkers are intrinsically linked to pathological alterations in the cardiac stromal and immune microenvironment, particularly fibrosis and innate immune activation. Our findings provide a systems-level view of IHF pathogenesis and offer potential molecular targets for improved diagnosis and therapeutic intervention.