Background: The aim of this study is to assess the validity of computer software prediction of soft tissue response to orthognathic surgery using cone beam computed tomography data.Methods: The sample of the study consisted of twelve adult female patients, with age ranging from 21 to 27 years with an average age of 24 years. They were selected on the basis that the main problem was an excess of the anterior part of the maxilla which required surgical correction by anterior maxillary osteotomy. Records were selected retrospectively where patients had available pre and post operative CBCTs which were taken at least 6 months postoperatively by the same operator using the same machine. All patients were treated at the Department of Orthodontics and Oral and Maxillofacial Surgery Cairo University between 2010 and 2013. Virtual models for the patient's soft tissue, maxilla and mandible were created from pre- and post- operative CBCT. Maxilla and mandible virtual models were moved to mimic the actual surgeries after Pre- and Post- operative CBCT were superimposed on the anterior cranial fossa, then soft tissue prediction was performed using InVivoDental v5.3 (Anatomage, San Jose, CA USA). Superimposition analysis and geometric morphometric analysis were used to quantify differences between predicted and actual soft tissue. Multiple landmarks were identified in actual and predicted soft tissue models. Distances between these models were measured at each landmark. Measurements for actual and predicted soft tissue were calculated then compared.Results: Over all 12 patients, the mean absolute error was equal to 1.86 mm. The overall median error was equal to 1.48 mm. The error was lower than 1, 2, 3 mm in 37%, 62% and 79% of the simulations, respectively. The 90th and 95th percentiles were 3.89 and 5.35 mm, respectively. Differences between post and predicted soft tissue were statistically significant less than Test Value = 4 mm at all landmarks. The perioral area had the largest prediction errors where the software over-predicted the position of the upper lip, while lower lip position was under-predicted.Conclusion: The predictions of soft tissue response to orthognathic surgery using In Vivo dental computer software were not reliable.