Malaria is a serious public health burden in Senegal. Accurate diagnosis is essential to avoided unnecessary presumptive treatment. Malaria diagnosis currently relies on identifying malaria parasite using microscopy and detecting soluble parasite antigens by Rapid Diagnostic Tests (RDT). These techniques do not detect low-level, sub-patent malaria infections and are inherently hazardous and invasive. To overcome these obstacles, alternatives diagnostics method were explored. In contrast to blood, saliva presents a reduced biohazard and can be painlessly collected in relatively large quantities by individuals with moderate training. The objective of this study was to use saliva collected in OmnIGEN Kits as an alternative sample for malaria detection. Methods: A total of 77 febril patients tested malaria positive by mRDT were enrolled in this study. From each patient, blood sample was collected for dried blood spot and blood smear; and saliva sample on OMNIGEN kit. Parasite density was determined from smear and Plasmodium falciparum DNA was extracted from both dried blood spot and saliva samples collected from the same patients. Extracted DNA was amplied by qPCR machine. Results: Malaria prevalence using qPCR was 98,7% and 60% respectively with blood and saliva. Compared to blood, saliva showed a sensitivity; specicity; positive predictive value of 60%; 100 %; and 100 %. The concordance between parasites detection from saliva and blood was (p = 0.45). In addition, no difference was found between these two methods and the microscopy counting. Conclusions:Saliva could be a non-invasive alternative method for P. falciparum detection and epidemiological surveillance in country with limited ressources.