Deep learning segmentation of mandible with lower dentition from cone beam CT.

OBJECTIVES: This study aimed to train a 3D U-Net convolutional neural network (CNN) for mandible and lower dentition segmentation from cone-beam computed tomography (CBCT) scans. METHODS: In an ambispective cross-sectional design, CBCT scans from two hospitals (2009-2019 and 2021-2022) constituted an internal dataset and external validation set, respectively. Manual segmentation informed CNN training, and evaluations employed Dice similarity coefficient (DSC) for volumetric accuracy. A blinded oral maxillofacial surgeon performed qualitative grading of CBCT scans and object meshes. Statistical analyses included independent t-tests and ANOVA tests to compare DSC across patient subgroups of gender, race, body mass index (BMI), test dataset used, age, and degree of metal artifact. Tests were powered for a minimum detectable difference in DSC of 0.025, with alpha of 0.05 and power level of 0.8. RESULTS: 648 CBCT scans from 490 patients were included in the study. The CNN achieved high accuracy (average DSC: 0.945 internal, 0.940 external). No DSC differences were observed between test set used, gender, BMI, and race. Significant differences in DSC were identified based on age group and the degree of metal artifact. The majority (80%) of object meshes produced by both manual and automatic segmentation were rated as acceptable or higher quality. CONCLUSION: We developed a model for automatic mandible and lower dentition segmentation from CBCT scans in a demographically diverse cohort including a high degree of metal artifacts. The model demonstrated good accuracy on internal and external test sets, with majority acceptable quality from a clinical grader.

World Workshop on Oral Medicine VIII: Development of a core outcome set for oral lichen planus: a consensus study.

OBJECTIVE: A core outcome set (COS) is the minimum agreed-on data set required to be measured in interventional trials. To date, there is no COS for oral lichen planus (OLP). This study describes the final consensus project that brought together the results of the previous stages of the project to develop the COS for OLP. STUDY DESIGN: The consensus process followed the Core Outcome Measures in Effectiveness Trials guidelines and involved the agreement of relevant stakeholders, including patients with OLP. Delphi-style clicker sessions were conducted at the World Workshop on Oral Medicine VIII and the 2022 American Academy of Oral Medicine Annual Conference. Attendees were asked to rate the importance of 15 outcome domains previously identified from a systematic review of interventional studies of OLP and a qualitative study of OLP patients. In a subsequent step, a group of OLP patients rated the domains. A further round of interactive consensus led to the final COS. RESULTS: The consensus processes led to a COS of 11 outcome domains to be measured in future trials on OLP. CONCLUSION: The COS developed by consensus will help reduce the heterogeneity of outcomes measured in interventional trials. This will allow future pooling of outcomes and data for meta-analyses. This project showed the effectiveness of a methodology that could be used for future COS development.

World Workshop on Oral Medicine VIII: Development of a core outcome set for oral lichen planus: the patient perspective.

OBJECTIVE: This study aimed to explore the lived experience of patients with oral lichen planus (OLP) and investigate what treatment-related outcomes are the most important to them and should be included in a core outcome set (COS) for OLP. STUDY DESIGN: A qualitative study involving focus group work with 10 participants was conducted. Interviews with each focus group were held twice: session 1 explored the lived experience of patients with OLP, and session 2 allowed patients to review a summary of the outcome domains used in the OLP literature to date. The discussions were recorded, transcribed verbatim, and analyzed using framework analysis. RESULTS: In session 1, 4 themes and 8 sub-themes emerged from the data analysis. An additional outcome, 'knowledge of family and friends,' was suggested in session 2. CONCLUSIONS: We have gained valuable insight into the lived experience of patients with OLP via this qualitative study. To our knowledge, this study is the first to explore the patient perspective on what should be measured in clinical trials on OLP, highlighting an important additional suggested outcome. This additional outcome will be voted upon in a consensus process to determine a minimum COS for OLP.

World Workshop on Oral Medicine VIII: Development of a core outcome set for oral lichen planus: a systematic review of outcome domains.

OBJECTIVE: There is a lack of consensus regarding clinician- and patient-reported oral lichen planus (OLP) outcomes. The World Workshop on Oral Medicine Outcomes Initiative for the Direction of Research (WONDER) Project aims to develop a core outcome set (COS) for OLP, which would inform the design of clinical trials and, importantly, facilitate meta-analysis, leading to the establishment of more robust evidence for the management of this condition and hence improved patient care. STUDY DESIGN: Ovid MEDLINE, Embase, CINAHL, CENTRAL, and Clinicaltrials.gov were searched for interventional studies (randomized controlled trials, controlled clinical trials, and case series including ≥5 participants) on OLP and oral lichenoid reactions published between January 2001 and March 2022 without language restriction. All reported primary and secondary outcomes were extracted. RESULTS: The searches yielded 9,135 records, and 291 studies were included after applying the inclusion criteria. A total of 422 outcomes were identified. These were then grouped based on semantic similarity, condensing the list to 69 outcomes. The most frequently measured outcomes were pain (51.9%), clinical grading of the lesions (29.6%), lesion size/extension/area (27.5%), and adverse events (17.5%). CONCLUSION: As a first step in developing a COS for OLP, we summarized the outcomes that have been used in interventional studies over the past 2 decades, which are numerous and heterogeneous.