Multi-Quantifying Maxillofacial Traits via a Demographic Parity-Based AI Model.

Objective and Impact Statement: The multi-quantification of the distinct individualized maxillofacial traits, that is, quantifying multiple indices, is vital for diagnosis, decision-making, and prognosis of the maxillofacial surgery. Introduction: While the discrete and demographically disproportionate distributions of the multiple indices restrict the generalization ability of artificial intelligence (AI)-based automatic analysis, this study presents a demographic-parity strategy for AI-based multi-quantification. Methods: In the aesthetic-concerning maxillary alveolar basal bone, which requires quantifying a total of 9 indices from length and width dimensional, this study collected a total of 4,000 cone-beam computed tomography (CBCT) sagittal images, and developed a deep learning model composed of a backbone and multiple regression heads with fully shared parameters to intelligently predict these quantitative metrics. Through auditing of the primary generalization result, the sensitive attribute was identified and the dataset was subdivided to train new submodels. Then, submodels trained from respective subsets were ensembled for final generalization. Results: The primary generalization result showed that the AI model underperformed in quantifying major basal bone indices. The sex factor was proved to be the sensitive attribute. The final model was ensembled by the male and female submodels, which yielded equal performance between genders, low error, high consistency, satisfying correlation coefficient, and highly focused attention. The ensemble model exhibited high similarity to clinicians with minor processing time. Conclusion: This work validates that the demographic parity strategy enables the AI algorithm with greater model generalization ability, even for the highly variable traits, which benefits for the appearance-concerning maxillofacial surgery.

Abnormal maxillary sinus diagnosing on CBCT images via object detection and 'straight-forward' classification deep learning strategy.

BACKGROUND: Pathological maxillary sinus would affect implant treatment and even result in failure of maxillary sinus lift and implant surgery. However, the maxillary sinus abnormalities are challenging to be diagnosed through CBCT images, especially for young dentists or dentists in grassroots medical institutions without systematical education of general medicine. OBJECTIVES: To develop a deep-learning-based screening model incorporating object detection and 'straight-forward' classification strategy to screen out maxillary sinus abnormalities on CBCT images. METHODS: The large area of background noise outside maxillary sinus would affect the generalisation and prediction accuracy of the model, and the diversity and imbalanced distribution of imaging manifestations may bring challenges to intellectualization. Thus we adopted an object detection to limit model's observation zone and 'straight-forward' classification strategy with various tuning methods to adapt to dental clinical need and extract typical features of diverse manifestations so that turn the task into a 'normal-or-not' classification. RESULTS: We successfully constructed a deep-learning model consist of well-trained detector and diagnostor module. This model achieved ideal AUROC and AUPRC of 0.953 and 0.887, reaching more than 90% accuracy at optimal cut-off. McNemar and Kappa test verified no statistical difference and high consistency between the prediction and ground truth. Dentist-model comparison test showed the model's statistically higher diagnostic performance than dental students. Visualisation method confirmed the model's effectiveness in region recognition and feature extraction. CONCLUSION: The deep-learning model incorporating object detection and straightforward classification strategy could achieve satisfying predictive performance for screening maxillary sinus abnormalities on CBCT images.

Quantification of the apical palatal bone index for maxillary incisor immediate implant assessment: A retrospective cross-sectional study.

BACKGROUND: Apical palatal bone is important in immediate implant evaluation. Current consensus gives qualitative suggestions regarding it, limiting its clinical decision-making value. OBJECTIVES: To quantify the apical palatal bone dimension in maxillary incisors and reveal its quantitative correlation with other implant-related hard tissue indices to give practical advice for pre-immediate implant evaluation and design. MATERIAL AND METHODS: A retrospective analysis of immediate implant-related hard tissue indices in maxillary incisors obtained by cone beam computed tomography (CBCT) was conducted. Palatal bone thickness at the apex level (Apical-P) on the sagittal section was selected as a parameter reflecting the apical palatal bone. Its quantitative correlation with other immediate implant-related hard tissue indices was revealed. Clinical advice of pre-immediate implant assessment was given based on the quantitative classification of Apical-P and its other correlated immediate implant-related hard tissue indices. RESULTS: Apical-P positively correlated with cervical palatal bone, whole cervical buccal-palatal bone, sagittal root angle, and basal bone width indices. while negatively correlated with apical buccal bone, cervical buccal bone, and basal bone length indices. Six quantitative categories of Apical-P are proposed. Cases with Apical-P below 4 mm had an insufficient apical bone thickness to accommodate the implant placement, while Apical-P beyond 12 mm should be cautious about the severe implant inclination. Cases with Apical-P of 4-12 mm can generally achieve satisfying immediate implant outcomes via regulating the implant inclination. CONCLUSIONS: Quantification of the apical palatal bone index for maxillary incisor immediate implant assessment can be achieved, providing a quantitative guide for immediate implant placement in the maxillary incisor zone.

"Brain-inspired intelligence" in dental implant decision-making / 口腔疾病防治

@#At present, implant surgery robots have basically achieved "surgical intelligence", but "brain-inspired intelligence" of robots is still in the stage of theory and exploration. The formulation of a clinical implantation plan depends on the timing of implantation, implantation area, bone condition, surgical procedure, patient factors, etc., which need to evaluate the corresponding clinical decision indicators and clinical pathways. Inspired by evidence-based medicine and the potential of big data and deep learning, combined with the data characteristics of clinical decision indicators and clinical pathways that can be quantitatively or qualitatively analyzed, this review simulates the cognitive behavior and neural mechanisms of the human brain and proposes a feasible brain-inspired intelligence scheme by predicting the decision indices and executing clinical pathways intelligently, that is, "select clinical indicators and clarify clinical pathways -- construct database -- use deep learning to intelligently predict decision indicators -- intelligent execution of clinical pathways -- brain-inspired intelligence of implant decision-making". Combined with the previous research results of our team, this review also describes the process of realization of brain-inspired intelligence for immediate implant timing decisions, providing an example of the comprehensive realization of brain-inspired intelligence of implant surgery robots in the future. In the future, how to excavate and summarize other clinical decision factors and select the best way to realize the automatic prediction of evidence-based clinical indicators and pathways and finally realize the complete intellectualization of clinical diagnosis and treatment processes will be one of the directions that dental clinicians need to strive for.

"Gingival Soft Tissue Integrative" Lithium Disilicate Glass-Ceramics with High Mechanical Properties and Sustained-Release Lithium Ions.

Due to their good mechanical performances and high biocompatibility, all-ceramic materials are widely applied in clinics, especially in orthopedic and dental areas. However, the "hard" property negatively affects its integration with "soft" tissue, which greatly limits its application in soft tissue-related areas. For example, dental implant all-ceramic abutments should be well integrated with the surrounding gingival soft tissue to prevent the invasion of bacteria. Mimicking the gingival soft tissue and dentine integration progress, we applied the modified ion-exchange technology to "activate" the biological capacity of lithium disilicate glass-ceramics, via introducing OH- to weaken the stability of Si-O bonds and release lithium ions to promote multi-reparative functions of gingival fibroblasts. The underlying mechanism was found to be closely related to the activation of mitochondrial activity and oxidative phosphorylation. In addition, during the ion-exchange process, the larger radius sodium ions (Na+) replaced the smaller radius lithium ions (Li+), so that the residual compressive stress was applied to the glass-ceramics surface to counteract the tensile stress, thus improving the mechanical properties. This successful case in simultaneous improvement of mechanical properties and biological activities proves the feasibility of developing "soft tissue integrative" all-ceramic materials with high mechanical properties. It proposes a new strategy to develop advanced bioactive and high strength all-ceramic materials by modified ion-exchange, which can pave the way for the extended applications of such all-ceramic materials in soft tissue-related areas.

Construction of an end-to-end regression neural network for the determination of a quantitative index sagittal root inclination.

BACKGROUND: Immediate implant placement in the esthetic area requires comprehensive assessments with nearly 30 quantitative indexes. Most artificial intelligence (AI)-driven measurements of quantitative indexes depend on segmentation or landmark detection, which require extra labeling of images and contain possible intraclass errors. METHODS: For the initial attempt, the method was tested on sagittal root inclination measurement. This study had developed an accurate and efficient end-to-end model incorporating a convolutional neural network (CNN) based on unlabeled cone-beam computed tomography (CBCT) images for immediate implant placement diagnosis and treatment. The model took pretrained ResNeXt101 as the backbone and was constructed based on 2,920 CBCT images with corresponding angles of the tooth axis and bone axis. The performance of our CNN model was evaluated on a separate test set. RESULTS: Our model exhibited high prediction accuracy in sagittal root inclination measurements, as evidenced by the low mean average error of 2.16°, the high correlation coefficient of 0.915 to manual measurement, and the narrow 95% confidence interval shown by Bland-Altman plots. The intraclass correlation coefficient further confirmed the measurement accuracy of our model was comparable with that of junior clinicians. The model took merely 0.001 seconds for each CBCT image, making it highly efficient. To better understand the model's quality, we visualized our end-to-end CNN model through Guided Backpropagation, Grad-CAM, and Guided Grad-CAM, and confirmed its effectiveness in region recognition. CONCLUSIONS: We succeeded in taking the first step in constructing the end-to-end immediate implant placement AI tool through sagittal root inclination measurements without intermediate steps and extra labeling on images.

Mesopore Controls the Responses of Blood Clot-Immune Complex via Modulating Fibrin Network.

Formation of blood clots, particularly the fibrin network and fibrin network-mediated early inflammatory responses, plays a critical role in determining the eventual tissue repair or regeneration following an injury. Owing to the potential role of fibrin network in mediating clot-immune responses, it is of great importance to determine whether clot-immune responses can be regulated via modulating the parameters of fibrin network. Since the diameter of D-terminal of a fibrinogen molecule is 9 nm, four different pore sizes (2, 8, 14, and 20 nm) are rationally selected to design mesoporous silica to control the fibrinogen adsorption and modulate the subsequent fibrin formation process. The fiber becomes thinner and the contact area with macrophages decreases when the pore diameters of mesoporous silica are greater than 9 nm. Importantly, these thinner fibers grown in pores with diameters larger than 9 nm inhibit the M1-polorazation of macrophages and reduce the productions of pro-inflammatory cytokines and chemokines by macrophages. These thinner fibers reduce inflammation of macrophages through a potential signaling pathway of cell adhesion-cytoskeleton assembly-inflammatory responses. Thus, the successful regulation of the clot-immune responses via tuning of the mesoporous pore sizes indicates the feasibility of developing advanced clot-immune regulatory materials.

Analysis of the sagittal root angle and its correlation with hard and soft tissue indices in anterior teeth for immediate implant evaluation: a retrospective study.

BACKGROUND: To assess the root angle characteristics of maxillary incisors, and to analyze the relationship between the root angle and other implant-related anatomical indices to use the sagittal root angle as an index for immediate implant evaluation and design. METHODS: A random sample consisting of 400 cone-beam computed tomography (CBCT) images and 65 maxillary plaster models were selected for the present study. CBCT and stereolithography (STL) scan images were imported as DICOM files into coDiagnostiX software for matching the hard and soft tissue. The angle between the long axis of the anterior tooth and the corresponding alveolar bone and implant-related hard and soft tissue indices were measured in the sagittal section. Descriptive statistics, frequency analysis, multi-level comparisons, and correlation analyses were performed. RESULTS: The average sagittal root angles were 15° at the central incisor and 19° at the lateral incisor. The root angle in males was significantly larger than that in females, and increased with age. The largest angle, 22.35°, was found in the lateral incisors of the oldest (> 50 years old) male group. The root angle was found to correlate with coronal buccal bone thickness, coronal palatal bone thickness, apical buccal bone thickness, palatal bone thickness, and the below apex bone thickness. CONCLUSIONS: The sagittal root angle could reflect the distribution of other implant-related anatomical indices, which may provide additional reference for the evaluation of immediate implant placement.