Objective: The aim of this study was to develop a low-dose protocol suitable for cone beam CT (CBCT) examination of an impacted maxillary canine in children by using a combination of dosimetry with subjective and objective image quality assessment. Methods: Radiation dose and image quality measurements were made using a dental CBCT machine. An image quality phantom was used to investigate the relationship between objective image quality and dose-area product (DAP) for a broad range of exposure settings. Subjective image quality assessment was achieved using a paediatric skull phantom submerged in a water bath for the same range of exposure settings. Eight clinicians assessed each CBCT data set for nine aspects of image quality using a five-point rating scale of agreement. Results: Acceptable image quality, defined using subjective judgements by the clinicians of the skull phantom images, was achievable with DAP values of 127mGycm2 or greater and a polytetrafluoroethylene (PTFE) contrast-tonoise ratio (CNR) of 4.8 or greater. A cautious choice was made to recommend a low-dose protocol of 80kV and 3mA for implementation into clinical practice, corresponding to a DAP value of 146mGy cm2 and a PTFE CNR of 5.0. Conclusion: A low-dose protocol for this particular CBCT machinewas established which represents as much as a 50% reduction compared with manufacturer's recommendations. Advances in knowledge: To the authors' best knowledge, this is the first study that addresses dose optimization in paediatric clinical protocols in dental CBCT. Furthermore, this study explores the relationship between radiation dose, objective and subjective image quality.