A computer-assisted diagnostic (CAD) tool was developed for the diagnosis of acute pulmonary embolism (PE) in perfusion lung scans. Forty-five scans (with angiographic proof) were included in the study. The CAD tool was composed of two modules. The first module performs multifractal texture analysis on the posterior view of the perfusion scan. The second module is a decision algorithm that merges the multifractal parameters into a diagnosis regarding the presence or absence of PE. Linear and non-linear decision models were evaluated for the diagnostic task. A consensus neural network significantly outperformed all decision models including the physicians.