Baekseok University

PURPOSE:Tooth pattern and surrounding tissue can affect the complexity of third molar (3M) extraction surgery. This study aimed to develop a new, more precise 3M extraction difficulty assessment scoring system by applying image computational techniques and artificial intelligence (AI)-based technologies to panoramic radiography images. METHODS:We used dental panoramic X-ray data of patients aged 16 to 86 years. The mandibular canal (MC), inner area of alveolar bone (IAAB), and 3M were detected and segmented using an AI-based deep learning model. We developed algorithms to evaluate the difficulty of 3M extraction and the risk of inferior alveolar nerve damage. The inclination of the 3M (1-4 points), impaction depth of the 3M (1-4 points), and proximity between the mandibular third molar (M3M) and MC (0-4) were classified and scored. The scores obtained from each algorithm were summed to classify the difficulty index as very easy (2-4), easy (5-6), slightly difficult (7-9), and very difficult (10-12). RESULTS:The average precisions of the maxillary 3M and M3M detection models were 0.93 and 0.97, respectively. Moreover, the average accuracies of the 3M inclination classification algorithm, 3M impaction depth calculation algorithm, and the proximity of M3M and MC algorithms were 0.8544, 0.9515, and 0.8991, respectively. CONCLUSION:The developed extraction difficulty scoring model can aid dentists in assessing the risk of 3M extraction and establish an appropriate surgical plan. Future studies should include more diverse patient data and improve the performance of the AI models to realize clinical implementation.

Deepfake technology poses severe security and trust challenges, demanding robust detection strategies. We propose a unified audio-visual framework that jointly exploits facial dynamics and speech acoustics to capture subtle forgery traces. The system integrates CNN-based face encoders for spatial features, RNN/Temporal-Conformer blocks for temporal cues, and lightweight Transformers for contextual modeling. An adaptive mid-late fusion module aggregates multimodal embeddings via gated attention and a calibration head, ensuring resilience against partial modality corruption. Preprocessing involves face detection and cropping, log-mel spectrogram extraction from 16 kHz audio, and alignment of video-audio segments with a unimodal fallback mechanism for instances of missing modalities. Experiments on FaceForensics++, DFDC, and FakeAVCeleb-with subject-and speaker-disjoint splits-demonstrate strong generalization, achieving up to 97.3% accuracy and 99.0% AUC under clean conditions. Key contributions include a principled multimodal fusion strategy and a plug-and-play ensemble mechanism that stabilizes training across datasets. Limitations such as computational overhead and the need for further adversarial robustness testing are discussed, while reproducibility is facilitated through released configurations and scripts. Overall, this work advances multimodal Deepfake detection by offering an efficient, accurate, and extensible defense framework.

BACKGROUND:Mothers with gestational diabetes mellitus (GDM) face specific challenges in breastfeeding, yet data on the long-term effectiveness of mobile-based interventions remain limited. This study aimed to evaluate the long-term effectiveness of a Mobile-Based Breastfeeding Promotion Program for GDM (M-BFGDM) on breastfeeding knowledge, self-efficacy, and practice rates up to 6 months postpartum. METHODS:A nonequivalent control group pretest-posttest quasi-experimental study was conducted. Participants were recruited from an online community. The intervention group received the M-BFGDM, which included Information-Motivation-Behavioral Skills (IMB) model-based educational videos and KakaoTalk counseling. Data were collected at prenatal, 1-week, 1-month, and 6-month postpartum time points. Data were analyzed using generalized estimating equations (GEE) and repeated-measures ANOVA. RESULTS:The final analysis included 38 participants (experimental group, n = 18; control group, n = 20). The M-BFGDM was effective in improving breastfeeding knowledge among women with GDM (p = 0.003). However, the intervention did not significantly influence the trajectory of breastfeeding self-efficacy or prevent the decline in practice rates over 6 months compared to the control group. CONCLUSIONS:These findings suggest that while mobile education enhances knowledge, sustained breastfeeding requires more intensive, individualized support to address physical barriers, such as low milk supply and latch difficulties.

Anxiety undermines skilled music performance, and mindfulness training and cognitive restructuring are frequently proposed as regulatory strategies. The author argues that their effectiveness depends on performance context and training dose. This study tested whether brief versus repeated interventions differentially influence music performance anxiety (MPA) and music performance quality (MPQ), and examined decentering and reappraisal as candidate mechanisms. Two complementary experiments were conducted with 196 year-2 music majors randomly assigned to mindfulness training (MT), cognitive restructuring (CR), combined MT + CR, or control. Experiment 1 implemented a brief pre-performance micro-session in a low-stakes setting and assessed immediate decentering and reappraisal as mediators. Experiment 2 embedded a multi-week program across four high-stakes juries. Outcomes included state MPA and blinded MPQ ratings. Data were analyzed using analysis of variance (ANOVA), analysis of covariance (ANCOVA), parallel multiple mediation, latent growth modelling, and mixed ANOVA. In the low-stakes context, all active interventions produced small reductions in MPA, with the largest effect observed for MT + CR; differences in MPQ were negligible. Anxiety reductions were mediated by immediate increases in decentering and reappraisal. In the multi-week high-stakes context, interventions yielded larger and more sustained reductions in MPA alongside modest improvements in blinded MPQ, again strongest for MT + CR. The study concludes that brief mindfulness-based interventions primarily reduce MPA, with minimal immediate effects on MPQ. In contrast, repeated practice under sustained evaluative pressure is necessary for these regulatory gains to extend to improvements in performance quality. These findings indicate that the effectiveness of mindfulness training and cognitive restructuring in music education is contingent on both contextual demands and training dose. This study was not preregistered.

This study examines global Hallyu (Korean Wave) fandom as a pluralistic, heterogeneous counter current within the decentralizing flow of global popular culture. It conceptualizes this fandom as a form of digital neo-tribe and interprets its fluid, ever-shifting character through the theoretical lens of liquidity. To better understand the performative practices of these fan communities, the study adopts and applies the concepts of trans-bricolage and post-coding. Hallyu fandom practices can be broadly divided into two types: cultural practices and economic practices. Cultural practices are expressed through the reappropriation of K-content's forms and messages via modes such as quotation/commentary, reference/borrowing, and transformation/modification. In contrast, economic practices are manifested through the production of derivative content based on original material or the creation of halo content-content that gains value through association with K-content. The analysis reveals that global Hallyu fans actively decontextualize and recontextualize original content according to their own cultural logics and sensibilities. Through their creative interventions, Hallyu is continuously transformed and redefined, securing influence as a fluid, hybridized "third culture" within the global popular culture market. These findings challenge the notion of fandom as passive consumption and instead position global Hallyu fans as active cultural agents and co-producers of meaning. Consequently, Hallyu should be understood not as a unidirectional export but as a collaborative cultural formation shaped by transnational participation. Korean creators, in this context, are encouraged to reflect on their role not only as content originators but as reciprocal re-broadcasters within the dynamic global circulation of Hallyu.