A novel computer algorithm for modeling and treating mandibular fractures: A pilot study

To describe a novel computer algorithm that can model mandibular fracture repair. To evaluate the algorithm as a tool to model mandibular fracture reduction and hardware selection.

Retrospective pilot study combined with cross-sectional survey.

A computer algorithm utilizing Aquarius Net (TeraRecon, Inc, Foster City, CA) and Adobe Photoshop CS6 (Adobe Systems, Inc, San Jose, CA) was developed to model mandibular fracture repair. Ten different fracture patterns were selected from nine patients who had already undergone mandibular fracture repair. The preoperative computed tomography (CT) images were processed with the computer algorithm to create virtual images that matched the actual postoperative three-dimensional CT images. A survey comparing the true postoperative image with the virtual postoperative images was created and administered to otolaryngology resident and attending physicians. They were asked to rate on a scale from 0 to 10 (0 = completely different; 10 = identical) the similarity between the two images in terms of the fracture reduction and fixation hardware.

Ten mandible fracture cases were analyzed and processed. There were 15 survey respondents. The mean score for overall similarity between the images was 8.41 ± 0.91; the mean score for similarity of fracture reduction was 8.61 ± 0.98; and the mean score for hardware appearance was 8.27 ± 0.97. There were no significant differences between attending and resident responses. There were no significant differences based on fracture location.

This computer algorithm can accurately model mandibular fracture repair. Images created by the algorithm are highly similar to true postoperative images. The algorithm can potentially assist a surgeon planning mandibular fracture repair.

4. Laryngoscope, 2016 127:331-336, 2017.