This paper proposes a methodology based on Bayesian decision theory and electrochemical impedance spectroscopy for rapid and reliable detection of Salmonella typhimurium bacteria. The proposed technique has been validated using a cost-effective and portable biosensor. Results indicate a detection time of approximately 6 minutes (5 minutes for data acquisition;1 minute for analysis) for pre-infectious bacterial concentrations. The technique has demonstrated high accuracy and can be generalized for the detection of other pathogens with healthcare relevance. Furthermore, the technique has a low computational complexity and is therefore well-suited for use in hand-held pathogen detectors.