Autostent: A Semi-Automated Approach to Designing Customized 3D-Printed Oral Radiation Stents for Patients with Head and Neck Cancer
Agrawal, A.; Zaid, M. M.; Roach, M. A.; Xiao, L.; Chambers, M. S.; Koay, E. J.
Show abstract
Oral stents may reduce toxicities during radiation therapy for head and neck cancer (HNC). Although customized 3D-printed oral stents are more quickly fabricated and non-inferior (in terms of patient reported outcomes) compared to conventionally-fabricated stents, the design process is still relatively time-consuming, unstandardized, and requires experienced technicians. We hypothesized that semi-automating the 3D-printed stent design process can reduce design time and standardize the workflow. Using oral stent design principles established over decades by oral oncologists, we developed a customized computer program (Autostent) using MATLAB to semi-automate the design process. We then compared a previously described method utilizing non-automated computer-aided design with Autostent. Three users designed stents for four patients with HNC enrolled on a prospective observational study. These patients were selected based on differing dental anatomies, and each user designed stents for each patient thrice, using both the non- and semi-automated methods. The design time and stent volumes for the two methods were statistically analyzed. Semi-automation was found to significantly reduce the average design time by 23.6 minutes (51.2%, p=0.001), independent of user, dental anatomy, and trial number. Additionally, semi-automation reduced the average stent volume by 4.33 mL (12.9%, p=0.016, univariate analysis). While this was not statistically significant after accounting for the other experimental variables (p=0.40, multivariate analysis), semi-automation did reduce the variability in the stent volume across users (overall standard error of the mean reduced by 40%). Thus, the semi-automated workflow significantly reduced the design time and the variability in the stent volume across users compared to the non-automated workflow. This may lead to potential cost benefits, standardization of the device, and increased population-wide access to a device that could help reduce toxicities for HNC.
Matching journals
The top 6 journals account for 50% of the predicted probability mass.
Similar papers in this journal
- Comprehensive Quantitative Evaluation of Inter-observer Delineation Performance of MR-guided Delineation of Oropharyngeal Gross Tumor Volumes and High-risk Clinical Target Therapy: An R-IDEAL Stage 0 Prospective Study 94%
- Multi-institutional Normal Tissue Complication Probability (NTCP) Prediction Model for Mandibular Osteoradionecrosis: Results from the PREDMORN Study 94%
- Initial Feasibility and Clinical Implementation of Daily MR-guided Adaptive Head and Neck Cancer Radiotherapy on a 1.5T MR-Linac System: Prospective R-IDEAL 2a/2b Systematic Clinical Evaluation of Technical Innovation 94%
Similar papers in this journal
- Recent advances in the clinical applications of machine learning in proton therapy 92%
- Breast density prediction from low and standard dose mammograms using deep learning: effect of image resolution and model training approach on prediction quality 89%
- Model uncertainty estimates for deep learning mammographic density prediction using ordinal and classification approaches 86%
Similar papers in this journal
- Quality Assurance Assessment of Intra-Acquisition Diffusion-Weighted and T2-Weighted Magnetic Resonance Imaging Registration and Contour Propagation for Head and Neck Cancer Radiotherapy 92%
- Cell lines of the same anatomic site and histologic type show large variability in intrinsic radiosensitivity and relative biological effectiveness to protons and carbon ions 91%
- Non-Invasive monitoring of normal tissue radiation damage using quantitative ultrasound spectroscopy 91%
Similar papers in this journal
- Large language models to help appeal denied radiotherapy services 92%
- Imaging-Genomics Study Of Head-Neck Squamous Cell Carcinoma: Associations Between Radiomic Phenotypes And Genomic Mechanisms Via Integration Of TCGA And TCIA 91%
- DeepPhe-CR: Natural Language Processing Software Services for Cancer Registrar Case Abstraction 88%
Similar papers in this journal
- Artificial Intelligence Uncertainty Quantification in Radiotherapy Applications - A Scoping Review 92%
- Deep learning NTCP model for late dysphagia after radiotherapy for head and neck cancer patients based on 3D dose, CT and segmentations 92%
- Evaluation of indirect damage and damage saturation effects in dose-response curves of hypofractionated radiotherapy of early-stage NSCLC and brain metastases 92%
"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.