Abstract
BACKGROUND: Oral stents may reduce toxicities during radiation therapy for head and neck cancer (HNC). Customized 3D-printed oral stents offer faster production and achieve comparable patient-reported outcomes to conventionally fabricated stents. However, their design process remains time-consuming, lacks standardization, and relies heavily on skilled technicians. We hypothesized that semi-automating the design process for 3D-printed, mouth-opening, tongue-depressing (MOTD) stents could standardize the design workflow and decrease design time. METHODS: Using oral stent design principles established over decades by oral oncologists, we created a customized computer program (Autostent) using MATLAB to semi-automate the design process of MOTD stents. We subsequently compared Autostent to a previously described method that utilized non-automated computer-aided design. Three users designed stents for four patients with HNC enrolled in a prospective observational study. These patients were selected based on their varying dental anatomies, and each user repeatedly designed an MOTD stent for each patient three times, employing both the non-automated and semi-automated methods. Both methods were compared in terms of design time and stent volume. RESULTS: Semi-automation reduced the average design time by 23.6 min (51.2%, p = 0.001), regardless of user, dental anatomy, or trial number. Additionally, semi-automation decreased the average stent volume by 4.33 mL (12.9%, p = 0.016, univariate analysis). Although this reduction was not statistically significant when considering other experimental variables (p = 0.40, multivariate analysis), semi-automation did lower the variability in stent volume among users (the overall standard error of the mean decreased by 40%). CONCLUSION: Our semi-automated workflow for designing and fabricating customized, 3D-printed MOTD stents significantly improves efficiency and reduces variability in the design. While these results indicate greater consistency compared to manual methods, further development is warranted to achieve full automation and to optimize clinical integration.