In reference to an article by Ariji et al,1 we would like to thank the authors for an informative study. This article was enlightening owing to the paucity of studies conducted regarding this subject. The authors took great efforts to examine the various imaging features that can be useful for qualitative differential diagnosis between ameloblastomas and keratocystic odontogenic tumours (KCOTs). However, we would like to highlight certain ambiguities in this article. First, in the methodology, all 60 observers enrolled in the study gave a diagnosis after observing the images via a website. The imaging features which helped them come to a diagnosis were not mentioned in the article. It could have added to the authenticity of the study if all 60 observers were given a set of instructions with criteria (imaging features) for evaluation of the images. We feel that it would have been better if all observers specialized in radiology were included in the study and an intraobserver analysis was carried out to remove any observational bias, as in the study by O Ferreira et al.2 Second, in the results it was stated in the article that number of locules was an important imaging feature contributing to correct imaging diagnosis. We fail to understand the basis of this as only 3 ameloblastomas were multilocular, while the other 17 lesions (including all 10 KCOTs) were unilocular. Further, whether all images were viewed by each observer is not clearly stated in the article. The authors have made a very laudable recommendation to use characteristic imaging features to differentiate between these two lesions that present clinical and radiographic similarities but require different treatments. They have also proposed CT as a valuable tool with the possibility of measuring the different density patterns of the lesions, which until today is a poorly investigated subject.