Pattern recognition of hematological profiles of tumors of the digestive tract: an exploratory study

Aims In this study, we aimed to apply laboratory blood analysis identify the hematological (based on hemoglobin concentration, erythrocytes, hematocrit, and RDW count) profiles associated with most prevalent forms of digestive tract malignancies. Furthermore, evaluate how these contributed distinguishing tumors at diagnosis. Methods We collected data from date ICD-10 diagnostic coding for C15 esophagus, C16 stomach, C18 colon, C19 rectum 184 individuals. The statistical visualization approaches, notably heat map principal component (PCA), allowed creating a summary profile identifying parameters each pathologic state. Univariate multivariate modeling ROC were performed in both SPSS Python. Results Our reveal unique patterns based tumor development anatomical location, clustering colon esophagus stomach. found significant difference between stomach carcinoma other tumors, which substantially correlated raised conjunction low hematocrit counts. contrast, had higher red cell count, allowing best classification metrics test set binary logistic regression (LR) model, accounting an AUC 0.77 94% sensitivity 52% specificity. Conclusion This study emphasizes significance adding diagnosing malignancies, could path further investigations regarding profiling monitoring point care.