Peripheral nerve injury is a common clinical entity, which may arise due to traumatic, tumorous, or even iatrogenic injury in craniomaxillofacial surgery. Despite advances in biomaterials and techniques over the past several decades, reconstruction of nerve gaps remains a challenge. Autografts are the gold standard for nerve reconstruction. Using autografts, there is donor site morbidity, subsequent sensory deficit, and potential for neuroma development and infection. Moreover, the need for a second surgical site and limited availability of donor nerves remain a challenge. Thus, increasing efforts have been directed to develop artificial nerve guidance conduits (ANCs) as new methods to replace autografts in the future. Various synthetic conduit materials have been testedin vitroandin vivo,and several first- and second-generation conduits are FDA approved and available for purchase, while third-generation conduits still remain in experimental stages. This paper reviews the current treatment options, summarizes the published literature, and assesses future prospects for the repair of peripheral nerve injury in craniomaxillofacial surgery with a particular focus on facial nerve regeneration.

Periapical radiolucencies, which can be detected on panoramic radiographs, are one of the most common radiographic findings in dentistry and have a differential diagnosis including infections, granuloma, cysts and tumors. In this study, we seek to investigate the ability with which 24 oral and maxillofacial (OMF) surgeons assess the presence of periapical lucencies on panoramic radiographs, and we compare these findings to the performance of a predictive deep learning algorithm that we have developed using a curated data set of 2902 de-identified panoramic radiographs. The mean diagnostic positive predictive value (PPV) of OMF surgeons based on their assessment of panoramic radiographic images was 0.69 (±0.13), indicating that dentists on average falsely diagnose 31% of cases as radiolucencies. However, the mean diagnostic true positive rate (TPR) was 0.51 (±0.14), indicating that on average 49% of all radiolucencies were missed. We demonstrate that the deep learning algorithm achieves a better performance than 14 of 24 OMF surgeons within the cohort, exhibiting an average precision of 0.60 (±0.04), and an F₁ score of 0.58 (±0.04) corresponding to a PPV of 0.67 (±0.05) and TPR of 0.51 (±0.05). The algorithm, trained on limited data and evaluated on clinically validated ground truth, has potential to assist OMF surgeons in detecting periapical lucencies on panoramic radiographs.

Background: As artificial intelligence (AI) becomes increasingly important in modern dentistry, we aimed to assess patients’ perspectives on AI in dentistry specifically for radiographic caries detection and the impact of AI-based diagnosis on patients’ trust. Methods: Validated questionnaires with Likert-scale batteries (1: “strongly disagree” to 5: “strongly agree”) were used to query participants’ experiences with dental radiographs and their knowledge/attitudes towards AI as well as to assess how AI-based communication of a diagnosis impacted their trust, belief, and understanding. Analyses of variance and ordinal logistic regression (OLR) were used (p < 0.05). Results: Patients were convinced that “AI is useful” (mean Likert ± standard deviation 4.2 ± 0.8) and did not fear AI in general (2.2 ± 1.0) nor in dentistry (1.6 ± 0.8). Age, education, and employment status were significantly associated with patients’ attitudes towards AI for dental diagnostics. When shown a radiograph with a caries lesion highlighted by an arrow, patients recognized the lesion significantly less often than when using AI-generated coloured overlays highlighting the lesion (p < 0.0005). AI-based communication did not significantly affect patients’ trust in dentists’ diagnosis (p = 0.44; OLR). Conclusions: Patients showed a positive attitude towards AI in dentistry. AI-supported diagnostics may assist communicating radiographic findings by increasing patients’ ability to recognize caries lesions on dental radiographs.

Background: Digital learning environments have become very common in the training of medical professionals, and students often use such platforms for exam preparation. Multiple choice questions (MCQs) are a common format in medical exams and are used by students to prepare for said exams. Objective: We aimed to examine whether particular learning activities contributed more strongly than others to users’ exam performance. Methods: We analyzed data from users of an online platform that provides learning materials for medical students in preparation for their final exams. We analyzed whether the number of learning cards viewed and the number of MCQs taken were positively related to learning outcomes. We also examined whether viewing learning cards or answering MCQs was more effective. Finally, we tested whether taking individual notes predicted learning outcomes, and whether taking notes had an effect after controlling for the effects of learning cards and MCQs. Our analyses from the online platform Amboss are based on user activity data, which supplied the number of learning cards studied and test questions answered. We also included the number of notes from each of those 23,633 users who had studied at least 200 learning cards and had answered at least 1000 test exam questions in the 180 days before their state exam. The activity data for this analysis was collected retrospectively, using Amboss archival usage data from April 2014 to April 2017. Learning outcomes were measured using the final state exam scores that were calculated by using the answers voluntarily entered by the participants. Results: We found correlations between the number of cards studied (r=.22; P<.001) and the number of test questions that had been answered (r=.23; P<.001) with the percentage of correct answers in the learners’ medical exams. The number of test questions answered still yielded a significant effect, even after controlling for the number of learning cards studied using a hierarchical regression analysis (β=.14; P<.001; ΔR2=.017; P<.001). We found a negative interaction between the number of learning cards and MCQs, indicating that users with high scores for learning cards and MCQs had the highest exam scores. Those 8040 participants who had taken at least one note had a higher percentage of correct answers (80.94%; SD=7.44) than those who had not taken any notes (78.73%; SD=7.80; t23631=20.95; P<.001). In a stepwise regression, the number of notes the participants had taken predicted the percentage of correct answers over and above the effect of the number of learning cards studied and of the number of test questions entered in step one (β=.06; P<.001; ΔR2=.004; P<.001). Conclusions: These results show that online learning platforms are particularly helpful whenever learners engage in active elaboration in learning material, such as by answering MCQs or taking notes.