Background: Schistosomiasis is a major public health challenge and one of the neglected tropical diseases globally. Schistosoma haematobium, the causative agent of urogenital schistosomiasis, is primarily endemic in African countries, with school-aged children (7 to 15 years) considered the most vulnerable population. The current diagnostic method relies on identifying parasite eggs in urine through microscopy, which is labor-intensive, requires specialized skills, and is often limited in sensitivity, especially in mild infections. Disease-related biomarkers offer a promising avenue for advancing disease diagnosis and detection. Method: A total of 135 school-aged children (aged 7—15 years) from Zanzibar were recruited for this study. To identify potential host protein biomarkers, data independent acquisition (DIA) proteomics combined with machine learning was used to screen for differentially expressed proteins in the urine samples from individuals infected with schistosome haematobium. Machine learning was used to pinpoint the most discriminative proteins, which were subsequently validated using enzyme-linked immunosorbent assay (ELISA). Results: Proteomic analysis identified 823 common host proteins in urine samples from the infection group, with 269 proteins showing significant differential expression. Of these, 149 were up-regulated and 120 were down-regulated in the infected group. Machine learning further highlighted SYNPO2, CD276, α2M, LCAT, and hnRNPM as the most discriminating biomarkers for Schistosomiasis haematobium infection. ELISA validation confirmed the differential expression trends of these proteins, underscoring their diagnostic potential. Conclusions: This study identified key urinary protein biomarkers associated with Schistosoma haematobium