The diagnosis of primary ciliary dyskinesia is based on demonstration of ciliary defects, mainly concerning dynein arms. Whereas the absence of outer dynein arms can be easily distinguished, the absence of inner dynein arms is difficult to confirm because of their low contrast on electron microscopy. Ciliary ultrastructure was studied in 40 patients suffering from respiratory tract infections. Conventional transmission electron microscopy showed normal cilia in 6 patients, confirmed a diagnosis of primary ciliary dyskinesia in 26 patients, and was inconclusive in 8 patients. All doubtful cases were related to inner dynein arm determination. Conventional electron microscopic analysis was able to define the ultrastructural phenotype of inner dynein arms in 40.5% of cases (6 presence of inner dynein arms, 13 absence of inner dynein arms). We developed computer-assisted analysis of electron microscopic micrographs to improve inner dynein arm visualization. Computer-assisted analysis consisted of image transformations designed to enhance the signal/noise ratio, based on the symmetry of ciliary axonemes. The sensitivity and specificity of computer-assisted analysis were 100 and 98%, respectively. The efficiency of computer-assisted analysis to visualize inner dynein arms, evaluated in the patients with undetermined phenotype after electron microscopy, was 86% (three normal cilia, seven primary ciliary dyskinesia with absence of outer dynein arms, three primary ciliary dyskinesia with absence of inner dynein arms, five partial absence of inner dynein arms). Computer-assisted analysis of ciliary micrographs improves the characterization of inherited axonemal defects.