Search Results (165)

Thermal facial imaging enables non-contact measurements of face heat patterns that are valuable for healthcare and affective computing, but common occluders (glasses, masks, scarves) and the single-channel, texture-poor nature of thermal frames make robust landmark localisation and visibility estimation challenging. We propose M³MSTL, a multi-modal, multi-stage, multi-task framework for occlusion-aware landmarking on thermal faces. M³MSTL pairs a ResNet-50 backbone with two lightweight heads: a compact fully connected landmark regressor and a Vision Transformer occlusion classifier that explicitly fuses per-landmark temperature cues. A three-stage curriculum (mask-based backbone pretraining, head specialisation with a frozen trunk, and final joint fine-tuning) stabilises optimisation and improves generalisation from limited thermal data. On the TFD68 dataset, M³MSTL substantially improves both visibility and localisation: the occlusion accuracy reaches 91.8% (baseline 89.7%), the mean NME reaches 0.246 (baseline 0.382), the ROC–AUC reaches 0.974, and the AP is 0.966. Paired statistical tests confirm that these gains are significant. Our approach aims to improve the reliability of temperature-based biometric and clinical measurements in the presence of realistic occluders. Full article

Reconstructing hand and object shapes from a single view during interaction remains challenging due to severe mutual occlusion and the need for high physical plausibility. To address this, we propose a novel framework for hand–object interaction reconstruction based on holistic, multi-stage collaborative optimization. Unlike methods that process hands and objects independently or apply constraints as late-stage post-processing, our model progressively enforces physical consistency and geometric accuracy throughout the entire reconstruction pipeline. Our network takes an RGB-D image as input. An adaptive feature fusion module first combines color and depth information to improve robustness against sensing uncertainties. We then introduce structural priors for 2D pose estimation and leverage texture cues to refine depth-based 3D pose initialization. Central to our approach is the iterative application of a dense mutual attention mechanism during sparse-to-dense mesh recovery, which dynamically captures interaction dependencies while refining geometry. Finally, we use a Signed Distance Function (SDF) representation explicitly designed for contact surfaces to prevent interpenetration and ensure physically plausible results. Through comprehensive experiments, our method demonstrates significant improvements on the challenging ObMan and DexYCB benchmarks, outperforming state-of-the-art techniques. Specifically, on the ObMan dataset, our approach achieves hand CD_h and object CD_o metrics of 0.077 cm² and 0.483 cm², respectively. Similarly, on the DexYCB dataset, it attains hand CD_h and object CD_o values of 0.251 cm² and 1.127 cm², respectively. Full article

To achieve precise in-hand manipulation and feedback control using soft robotic fingers, it is essential to accurately measure their deformable structures. In particular, estimating the three-dimensional shape of a soft finger under contact conditions is a critical challenge, as the deformation state directly affects manipulation reliability. However, nonlinear deformations and occlusions arising from interactions with external objects make the estimation difficult. To address these issues, we propose a soft finger structure that integrates small magnets and magnetic sensors inside the body, enabling the acquisition of rich deformation information in both contact and non-contact states. The design provides a 15-dimensional time-series signal composed of motor angles, motor currents, and magnetic sensor outputs as inputs for shape estimation. Built on the sensing signals, we propose a mode-selection-based learning approach that outputs multiple candidate shapes and selects the correct one. The proposed network predicts the three-dimensional positions of four external markers attached to the finger, which serve as a proxy representation of the finger’s shape. The network is trained in a supervised manner using ground-truth marker positions measured by a motion capture system. The experimental results under both contact and non-contact conditions demonstrate that the proposed method achieves an average estimation error of approximately 4 mm, outperforming conventional one-shot regression models that output coordinates directly. The integration of magnetic sensing is demonstrated to be able to enable accurate recognition of contact states and significantly improve stability in shape estimation. Full article

Background and Clinical Significance: Furuncular myiasis is a tropical parasitic skin infestation caused by dipterous fly larvae, most commonly affecting travellers to endemic regions. While returning travellers typically present with one or few lesions, extensive parasitism is rare. Increased global mobility and expanding ecological range of myiasis-causing species underscores the need for clinicians in endemic and non-endemic regions to recognise, diagnose, and manage this condition promptly. Awareness of exposure risks—including soil contact, infested clothing, and poor living conditions—is essential to reducing morbidity and preventing complications like secondary bacterial infection. Case Presentation: A healthy male in his forties returned to the UK after a month-long visit to rural Ethiopia, during which he slept on dirt floors and hung his washing on a line. He developed pruritic papular lesions that progressed to erythematous furuncles with central puncta and purulent discharge, accompanied by sensations of movement. The patient self-extracted 12 larvae in Ethiopia and subsequently sought local medical attention, receiving Albendazole, after which emerging larvae were non-motile. On UK presentation, he had 27 lesions at varying stages, 3 with signs of secondary infection. Laboratory investigations revealed elevated inflammatory markers, and wound swabs grew scanty Streptococcus pyogenes. Management included wound occlusion and systemic antibiotics. No further larvae were retrieved, precluding definitive speciation. All lesions improved over subsequent reviews. Conclusions: This case illustrates an unusually extensive presentation of presumed Cordylobia spp. myiasis in a returning traveller, highlighting potential complications following larvicidal therapy. Clinicians should maintain a high index of suspicion for myiasis in patients with compatible cutaneous lesions and relevant history. Increasing travel and shifting vector distributions make familiarity with tropical dermatoses and provision of effective safety measures essential in clinical practice. Full article

This cross-sectional experimental study compared a digital intraoral-scanner-based method with a traditional wax-registration method for the quantitative assessment of static occlusal contacts. Twenty adults with natural dentition were evaluated using an intraoral scan analyzed through a Java-based software (PixCount.java, version 1.0, version 1.0, University of Insubria, Varese, Italy) and wax registration analyzed with Z_TMJ software (Z_TMJ, version 1.0, University of Insubria, Varese, Italy). The primary outcome was the percentage distribution of static occlusal contacts between hemi-arches. A paired t-test and the intraclass correlation coefficient (ICC) were used to evaluate differences and agreement. Mean contact distribution was 49.75 ± 3.44% for the digital method and 48.02 ± 5.31% for the wax method. No statistically significant difference was observed (p > 0.05), and agreement analysis showed moderate concordance (ICC ≈ 0.43). Digital analysis provided superior visualization and workflow efficiency, whereas wax registration remained a practical, low-cost option. These findings indicate that both methods provide clinically meaningful information, with the digital approach offering additional practical advantages. The observed consistency between the two techniques supports the expanding role of digital tools in routine occlusal assessment. Full article

The Objective Grading System (OGS) developed by the American Board of Orthodontics (ABO-OGS) provides an objective method to evaluate the quality of orthodontic treatment outcomes. Initially designed to assess individual orthodontists, it is now widely adopted by institutions to evaluate treatment results. However, access to digital cast analysis remains limited in developing countries due to the high cost of specialized software. Objectives: This study aimed to compare physical and digital models based on ABO-OGS parameters in finished treatments and to determine the percentage of cases that met the ABO case category specifications in the graduate Orthodontics program at Universidad Cooperativa de Colombia (Bogotá campus) between 2017 and 2021. Methods: A retrospective descriptive study analyzed clinical records from 32 patients who completed orthodontic treatment between 2017 and 2021. Standardized plaster casts, digitized casts, and panoramic radiographs were evaluated. Manual assessment was performed using the ABO-OGS gauge on physical casts, while digital assessment was conducted using software on scanned models. Eight ABO-OGS parameters were scored following established guidelines. Results: Manual and digital ABO-OGS assessment demonstrated almost perfect agreement. The intraclass correlation coefficient was ICC (A,1) = 0.999 (p < 0.0001), and Bland–Altman analysis revealed a negligible mean bias of 0.34 points with narrow 95% limits of agreement (–0.60 to 1.29). Although the Wilcoxon signed-rank test detected a statistically significant difference (p = 0.001), the median scores were clinically equivalent (23.0 vs. 23.5). Overall, 69% of cases met the ABO-OGS passing threshold (≤30), while 31% did not (>30). The greatest differences between manual and digital methods were observed in occlusal contacts, marginal ridges, and buccolingual inclination. Occlusal relationships, overjet, and alignment contributed the most to the total ABO-OGS scores. Both linear (Least Absolute Shrinkage and Selection Operator regression—Lasso) and non-linear (Random Forest) models consistently identified the same core predictors, confirming the robustness of digital and manual workflows in capturing key determinants of treatment outcomes. Conclusions: Manual and digital methods of ABO-OGS assessment are clinically interchangeable. Despite small statistical differences, digital models provided reproducible results, with 69% of cases meeting ABO-OGS passing criteria. These findings support the validity of digital models as a reliable alternative for orthodontic outcome evaluation. Full article

Non-contact heart rate estimation technology based on frequency-modulated continuous wave (FMCW) radar has garnered extensive attention in single-target scenarios, yet it remains underexplored in multi-target environments. Accurate discrimination of multiple targets and precise estimation of their heart rates constitute key challenges in the multi-target domain. To address these issues, we propose a novel scheme for multi-target heart rate estimation. First, a high-precision distance-bin selection (HDBS) method is proposed for target localization in the range domain. Next, multiple-input multiple-output (MIMO) array processing is combined with the Root-multiple signal classification (Root-MUSIC) algorithm for angular domain estimation, enabling accurate discrimination of multiple targets. Subsequently, we propose an efficient method for interference suppression and vital sign extraction that cascades variational mode decomposition (VMD), local mean decomposition (LMD), and wavelet thresholding (WT) termed as VLW, which enables high-quality heartbeat signal extraction. Finally, to achieve high-precision and super-resolution heart rate estimation with low computational burden, an improved fast iterative interpolated beamforming (FIIB) algorithm is proposed. Specifically, by leveraging the conjugate symmetry of real-valued signals, the improved FIIB algorithm reduces the execution time by approximately 60% compared to the standard version. In addition, the proposed scheme provides sufficient signal-to-noise ratio (SNR) gain through low-complexity accumulation in both distance and angle estimation. Six experimental scenarios are designed, incorporating densely arranged targets and front-back occlusion, and extensive experiments are conducted. Results show this scheme effectively discriminates multiple targets in all tested scenarios with a mean absolute error (MAE) below 2.6 beats per minute (bpm), demonstrating its viability as a robust multi-target heart rate estimation scheme in various engineering fields. Full article

Background/Objectives: Occlusal splints are widely used for managing sleep bruxism (SB), providing uniform contact across the entire dentition in the centric relation. Nonetheless, different guidance schemes, such as bilateral balanced occlusion (BBO) and canine guidance (CG), are used during eccentric movements, and the optimal design remains unclear. This study compared the effects of BBO and CG on masticatory muscle activity, sleep architecture, and subjective outcomes during sleep. Methods: This non-blinded randomized crossover trial enrolled 24 healthy adults diagnosed with SB (16 men and 8 women; mean age, 26.1 years) who were randomly assigned to either a BBO-first or CG-first sequence. Individual splints of both types were milled from the polymethyl methacrylate discs. After a 5-night baseline period, each splint was worn for 33 nights in a home environment, and data from nights 29 to 33 were analyzed. Masseter muscle activity was assessed using single-channel electromyography (EMG), yielding EMG parameters, including integrated EMG per hour, number of episodes and bursts per hour, mean episode duration, and total episode duration per hour. Sleep architecture was assessed using portable polysomnography with automatic scoring, and subjective outcomes were assessed for sleep disturbance, morning symptoms, and splint comfort. Differences between splints were analyzed using Wilcoxon signed-rank tests (α = 0.05). Results: Twenty-three participants completed the study. No statistically significant differences were found between the BBO and CG splints for any EMG parameters, sleep variables, or subjective measures. Conclusions: Splint guidance design differences showed no significant effects; however, smaller, potentially clinically relevant effects cannot be excluded. Full article

Background/Objectives: Artificial intelligence (AI) is increasingly embedded in CAD/CAM workflows to address persistent challenges in restorative dentistry, including unpredictable color outcomes and time-intensive crown design steps. Yet, evidence on its accuracy and efficiency remains fragmented across heterogeneous study designs and metrics. This systematic review and meta-analyses aimed to evaluate the accuracy and performance of AI for color prediction and automated crown design in CAD/CAM ceramics. Methods: A systematic review with random-effects meta-analyses. The outcomes included design time, internal fit, finish-line accuracy, color-prediction acceptability using ΔE₀₀ (AT₀₀), morphology deviation, and occlusal and proximal contacts. Results: Fifteen studies met the inclusion criteria. The meta-analyses showed that AI-equipped CAD reduced crown design time compared to conventional CAD (MD −88.7 s; 95% CI −134.5 to −42.9; I² = 72%). The internal fit showed a small advantage for AI (MD −17.1 µm; 95% CI −26.2 to −7.9; I² = 90%). For finish-line identification, the pooled mean Hausdorff distance was ~0.35 mm (95% CI 0.316–0.382; I² = 0%). For color prediction, the pooled proportion of predictions within each study’s prespecified acceptability threshold (AT₀₀) was near-universal (0.996; 95% CI 0.988–0.999; I² = 0%). Morphology and functional contacts were not pooled due to incompatible metrics and units. Narrative synthesis indicated AI performance comparable to, or favorable over, conventional/technician workflows in selected regions. Conclusions: AI for CAD/CAM dentistry shows practical promise, most clearly for design-time efficiency and with encouraging signals for internal fit, finish-line identification, and color-prediction acceptability under study thresholds. However, clinical translation should proceed cautiously. Full article

Sitting posture recognition, defined as automatically localizing and categorizing seated human postures, has become essential for large-scale ergonomics assessment and longitudinal health-risk monitoring in classrooms and offices. However, in real-world multi-person scenes, pervasive occlusions and overlaps induce keypoint misalignment, causing global-attention backbones to fail to localize critical local structures. Moreover, annotation scarcity makes small-sample training commonplace, leaving models insufficiently robust to misalignment perturbations and thereby limiting cross-domain generalization. To address these challenges, we propose LAViTSPose, a lightweight cascaded framework for sitting posture recognition. Concretely, a YOLOR-based detector trained with a Range-aware IoU (RaIoU) loss yields tight person crops under partial visibility; ESBody suppresses cross-person leakage and estimates occlusion/head-orientation cues; a compact ViT head (MLiT) with Spatial Displacement Contact (SDC) and a learnable temperature (LT) mechanism performs skeleton-only classification with a local structural-consistency regularizer. From an information-theoretic perspective, our design enhances discriminative feature compactness and reduces structural entropy under occlusion and annotation scarcity. We conducted a systematic evaluation on the USSP dataset, and the results show that LAViTSPose outperforms existing methods on both sitting posture classification and face-orientation recognition while meeting real-time inference requirements. Full article

Background/Objectives: Immediate implant placement in the esthetic zone can shorten treatment time but maintaining peri-implant soft tissue stability is challenging. Conventional multi-stage workflows require multiple visits and may disturb peri-implant tissues. Placing a definitive one-time abutment at surgery can preserve soft tissue contours by avoiding multiple abutment changes. This case report introduces a digital one-stage approach delivering a definitive zirconia crown with delayed occlusal veneering at surgery to streamline treatment and preserve tissue stability. Methods: A 60-year-old male with a failing maxillary canine underwent immediate implant placement using guided surgery. A customized healing abutment preserved the emergence profile for the definitive restoration. A zirconia crown with an occlusal cut-back was fabricated and delivered at surgery on the one-time abutment without occlusal contact. After 12 weeks, a ceramic overlay was bonded extraorally to the crown to restore the occlusal surface. Results: At 2-year follow-up, the implant exhibited stable bone and healthy peri-implant soft tissues, with no complications. The one-time approach preserved tissue contours by eliminating provisional stages, and delayed occlusal veneering provided excellent esthetic integration. The patient was satisfied with the immediate result and fewer visits. This one-stage approach required fewer interventions than conventional provisional workflows. Conclusions: Immediate implant placement with a one-time abutment and delayed occlusal loading preserved peri-implant tissue architecture and achieved excellent functional and esthetic outcomes at 2 years. This one-stage workflow is a tissue-preserving alternative to multi-stage protocols; further studies are needed to confirm its long-term efficacy. Full article

Background: Superior semicircular canal dehiscence (SSCD) is characterized by a bony defect of the superior semicircular canal (SSC), leading to vestibular and auditory symptoms. A process of spontaneous “auto-plugging,” in which the overlying dura mater progressively occludes the SSC, may replicate the effects of surgical canal plugging but remains under-recognized. The present study reports diverse clinical, instrumental, and 3d High Resolution MRI findings in patients with SSCD and subsequently confirmed to present with spontaneous complete or partial auto-plugging. Methods: We retrospectively reviewed 11 patients with SSCD diagnosed on high-resolution CT and suspected auto-plugging based on clinical atypia and large dehiscence (>4 mm). Patients underwent comprehensive neurotological assessment, including pure-tone audiometry, vestibular testing, and HR MRI with 3D labyrinthine reconstructions to identify partial or complete auto-plugging. Auto-plugging was classified as partial (Canalis semicircularis superior depressus) or complete (absence of endolymph fluid signal; Canalis semicircularis superior obturatus). Results: Among 13 ears with auto-plugging, 6 were partial and 7 complete. The mean SSCD size in auto-plugged ears was 5.5 mm. Most ears had normal or near-normal vestibular function on VHIT, with minimal air-bone gaps and preserved VEMP responses. Imaging demonstrated varying degrees of dural contact with the SSC, confirming partial or complete canal occlusion. Conclusions: Spontaneous auto-plugging of the SSC is a plausible, under-recognized phenomenon that may reproduce functional effects of surgical plugging. Dedicated 3D labyrinthine MRI enhances detection and characterization. Prospective multimodal studies are needed to clarify the pathophysiology, progression, and clinical implications, optimizing patient selection for surgical versus conservative management. Full article

Non-contact vital sign monitoring in Pediatric Intensive Care Units is challenged by frequent occlusions, data scarcity, and the need for temporally stable anatomical tracking to extract reliable physiological signals. Traditional detectors produce unstable tracking, while video transformers are too computationally intensive for deployment on resource-limited clinical hardware. We introduce Divided Space–Time Mamba, an architecture that decouples spatial and temporal feature learning using State Space Models to achieve linear-time complexity, over 92% lower than standard transformers. To handle data scarcity, we employ self-supervised pre-training with masked autoencoders on over 50 k domain-specific video clips and further enhance robustness with multimodal RGB-D input. Our model demonstrates superior performance, achieving 0.96 mAP@0.5, 0.62 mAP50-95, and 0.95 rotated IoU. Operating at 23 FPS (43 ms latency), our method is approximately 1.9× faster than VideoMAE and 5.7× faster than frame-wise YOLOv8, demonstrating its suitability for real-time clinical monitoring. Full article

Background/Objectives: Occlusal trauma does not initiate periodontitis but may accelerate its progression when inflammation is present. Chlorhexidine (CHX) mouthwash is widely used as an adjunct to periodontal therapy, although its effectiveness in patients with occlusal trauma is insufficiently documented. This study aimed to evaluate the effect of CHX mouthwash on periodontal status in patients with deflective occlusal contacts, in the absence of occlusal adjustments. Materials and Methods: This observational prospective study analyzed data from 52 patients (20 males, 32 females; mean age 41.35 years). Periodontal status was assessed using the Community Periodontal Index of Treatment Needs (CPITN) at baseline, 3 months, and 6 months. Patients were divided into groups based on CHX use and concentration. Statistical analysis evaluated intra- and intergroup variations. Results: Patients using CHX demonstrated statistically significant improvements in CPITN scores at 3 months for all sextants except sextant 5 (p < 0.05). Between 3 and 6 months, further significant improvement was observed only for sextant 6 and for the overall score. In contrast, patients without CHX showed a slight trend toward worsening CPITN values, with no statistically significant differences over the same period. Conclusions: CHX mouthwash significantly improved periodontal parameters in patients with occlusal trauma during the first three months of use. However, improvements plateaued after this period, highlighting the short-term benefits and limitations of CHX. These findings support the adjunctive role of CHX in managing periodontal disease associated with occlusal trauma but reinforce the necessity of mechanical occlusal correction for long-term stability. The CPITN index provides moderate clinical utility compared with full-mouth clinical periodontal measurements. Full article

Background/Objectives: Occlusal analysis is an important component of oral rehabilitation with a determining role in the prognosis of restorations. Over time, several qualitative and quantitative occlusal analysis methods have been proposed, starting with occlusion wax up to the most advanced digital systems. The objective of the present study was to evaluate and compare the data obtained through dental occlusion analysis using the Medit i700 and Exocad Elefsina v3.2 in a group of subjects, in order to establish the reliability or compatibility between the two occlusal analysis systems. Methods: The study was conducted on 20 subjects, aged between 24 and 53 years, who presented in the Dental Prosthetics Clinic of the University of Medicine and Pharmacy of Craiova. Digital impressions were acquired using the Medit Link v.3.3.6 intraoral scanner, and the digital files were subsequently uploaded from the Medit i700 into the Medit Occlusion Analyzer application and the Dental CAD Exocad software. For the analysis of occlusion in dynamics, mandibular movements and data acquisition, positions of edge-to-edge in protrusion, edge-to-edge in right laterotrusion and edge-to-edge in left laterotrusion were recorded, using the corresponding print screens. The 2D occlusal contact images generated by the two software programs were converted into .jpeg format and subsequently imported into Adobe Photoshop CS6 (2021) for comparative analysis. The data were statistically processed for each software used and the obtained data were subsequently compared. Results: The occlusal surfaces recorded with the Medit Occlusion Analyzer application represent 94% of the occlusal surfaces recorded with the Exocad software for the maxilla and 90% of the occlusal surfaces recorded for the mandible. In maximum intercuspation, the highest values were recorded by the Medit i700 software, whereas in edge-to-edge protrusion and both right and left edge-to-edge laterotrusion positions, the highest values were reported by the Exocad software. The discrepancy between maxillary and mandibular values arises from the conversion of the data from a three-dimensional to a two-dimensional format during image processing. Conclusions: The occlusal areas recorded by the DentalCAD Exocad software show higher values than those provided by the Medit Link software with the Medit Occlusion Analyzer application. The differences in recorded values, in the case of the digital flow of prosthetic restorations, require the intervention of the dentist to perform clinical adjustments to optimize occlusal relationships after the fabrication and cementation of restorations. Full article