Semantic Segmentation of Maxillary Teeth and Palatal Rugae in Two-Dimensional Images.

The superimposition of sequential radiographs of the head is commonly used to determine the amount and direction of orthodontic tooth movement. A harmless method includes the timely unlimited superimposition on the relatively stable palatal rugae, but the method is performed manually and, if automated, relies on the best fit of surfaces, not only rugal structures. In the first step, motion estimation requires segmenting and detecting the location of teeth and rugae at any time during the orthodontic intervention. Aim: to develop a process of tooth segmentation that eliminates all manual steps to achieve an autonomous system of assessment of the dentition. Methods: A dataset of 797 occlusal views from photographs of teeth was created. The photographs were manually semantically segmented and labeled. Machine learning methods were applied to identify a robust deep network architecture able to semantically segment teeth in unseen photographs. Using well-defined metrics such as accuracy, precision, and the average mean intersection over union (mIoU), four network architectures were tested: MobileUnet, AdapNet, DenseNet, and SegNet. The robustness of the trained network was additionally tested on a set of 47 image pairs of patients before and after orthodontic treatment. Results: SegNet was the most accurate network, producing 95.19% accuracy and an average mIoU value of 86.66% for the main sample and 86.2% for pre- and post-treatment images. Conclusions: Four architectural tests were developed for automated individual teeth segmentation and detection in two-dimensional photos that required no post-processing. Accuracy and robustness were best achieved with SegNet. Further research should focus on clinical applications and 3D system development.

Three-dimensional evaluation of the effects of injectable platelet rich fibrin (i-PRF) on alveolar bone and root length during orthodontic treatment: a randomized split mouth trial.

BACKGROUND: The role of injectable platelet rich fibrin (i-PRF) in orthodontic treatment has not been investigated with focus on its effect on dental and bony periodontal elements. OBJECTIVE: To evaluate the efficacy of i-PRF in bone preservation and prevention of root resorption. METHODS: A randomized split-mouth controlled trial included 21 patients aged 16-28 years (20.85 ± 3.85), who were treated for Class II malocclusion with the extraction of the maxillary first premolars. Right and left sides were randomly allocated to either experimental treated with i-PRF or control sides. After the leveling and alignment phase, the canines were retracted with 150gm forces. The i-PRF was prepared from the blood of each patient following a precise protocol, then injected immediately before canine retraction on the buccal and palatal aspects of the extraction sites. Localized maxillary cone beam computed tomography scans were taken before and after canine retraction to measure alveolar bone height and thickness and canine root length (indicative of root resorption), and the presence of dehiscence and fenestration. Paired sample t-tests and Wilcoxon signed rank tests were used to compare the changes between groups. RESULTS: No statistically significant differences in bone height, bone thickness were found between sides and between pre- and post-retraction period. However, root length was reduced post retraction but did not differ between sides. In both groups, postoperative dehiscence was observed buccally and palatally and fenestrations were recorded on only the buccal aspect. CONCLUSIONS: I-PRF did not affect bone quality during canine retraction or prevent canine root resorption. I-PRF did not reduce the prevalence of dehiscence and fenestration. Trial registration ClinicalTrials.gov (identifier number: NCT03399760. 16/01/2018).

Effect of injectable platelet-rich fibrin (i-PRF) in accelerating orthodontic tooth movement : A randomized split-mouth-controlled trial.

BACKGROUND: The role of platelet-rich fibrin (PRF) in accelerating orthodontic tooth movement has been controversially discussed in available clinical studies. OBJECTIVE: To investigate the effectiveness of i­PRF in accelerating maxillary canine retraction. MATERIALS AND METHODS: A split-mouth design was applied in 21 participants (6 men, 15 women; mean age: 20.85 ± 3.85 years) whose class II division I malocclusion required the extraction of both maxillary first premolars. The right and left canines were randomized into intervention and control sides. After the initial leveling and alignment phase and immediately before canine retraction, i­PRF obtained from the brachial vein was injected into the mucosa on the buccal and palatal aspects of the intervention sides. The injection was repeated one month later. Study casts were taken at the initiation of canine retraction (T0) and at monthly visits up to 5 months (T1 through T5). The paired t­test was used to compare the total and monthly rates of canine retraction, canine rotation, and anchorage loss. RESULTS: The average rates of canine retraction were greater on the experimental side at T2, T3, and T4, but this difference with the control side was statistically significantly different only at T2 (P < 0.05). Differences in canine rotation and anchorage loss were not statistically significant. CONCLUSION: The rates of canine retraction following the injection of platelet-rich fibrin were not statistically significantly greater on the experimental than the control sides except at the second month (T2). This apparently transient rate of tooth movement indicates that repeated injections might be needed for sustained effects, a premise meriting more focused research.

Potential and limitations of orthodontic biomechanics: recognizing the gaps between knowledge and practice.

The perennial goals of efficient biomechanics are to obtain more controlled and faster movement and using more discrete appliances. The most recent technological advances have buttressed these goals. Temporary anchorage devices have revolutionized orthodontic practice and loom as a solid cornerstone of orthodontic science, along with the use of light forces, facilitated by "smart" archwires for optimal tooth movement. Accelerated tooth movement with decortication has been successful because of decreasing the resistance of cortical bone but micro-osteoperforation has not matched the same results. Clear aligners and preprogrammed regular or lingual appliances reflect the importance of three-dimensional technology in appliance design based on treatment outcome. These mechanical developments have inched the science closer to the traditional goals, but advances lack regarding their enhancement by biomaterials in a system where the physical stimulus is exerted on the teeth but the expression of tooth displacement is through the biological processes within the surrounding tissues. In this article, present tenets, applications, and advances are explored along with the gaps between knowledge and practice and the possibilities to bridge them. Anchorage control remains the major widely used development but slower is the development of faster noninvasive treatment.

Effect of force direction and tooth angulation during traction of palatally impacted canines: A finite element analysis.

INTRODUCTION: Treatment of a palatally impacted canine (PIC) is associated with demanding anchorage control, increased treatment duration, and undesirable side effects. Accurate PIC localization and force application impact treatment success. The objective of this research was to determine the stresses on the PIC when subjected to initial force activation in various directions (buccal, vertical, and distal) and relative to impaction severity. METHODS: Thirty PICs from 21 scans underwent finite element modeling. A prototype 3D model was reconstructed and segmented into its anatomic components. Each PIC was precisely positioned in the prototype model according to impaction position. Stresses in response to a (1.0 N) force in the distal, vertical, and buccal directions were evaluated at different levels of the root (apical, middle, and cervical). RESULTS: Distal and buccal forces yielded higher stress (6.64 and 6.41 kPa, respectively) than the vertical force (5.97 kPa) on the total PIC root and the apical and cervical root levels, but not at midroot. Statistically significant differences between severity groups were found mostly at the apical level among all force directions, except between distal and buccal forces in the higher severity group. In this group, stress was greatest at the cervical level with the buccal force significantly different from the stresses generated by either the distal or the vertical force. CONCLUSIONS: Vertical forces generated the lowest stresses. Differentially distributed stresses over the root reflected an initial tipping movement. Greater cervical stresses from the buccal force indicate resistance to movement, suggesting treatment initiation with vertical and distal forces over buccal forces, particularly with severely inclined canines.

Finite element analysis of stresses on adjacent teeth during the traction of palatally impacted canines.

OBJECTIVE: To evaluate stresses on maxillary teeth during alignment of a palatally impacted canine (PIC) under different loading conditions with forces applied in vertical and buccal directions. MATERIALS AND METHODS: A three-dimensional finite element model of the maxilla was developed from a cone beam computed tomographic scan of a patient with a left PIC. Traction was simulated under different setups: (1) palatal spring extending from a transpalatal bar (TPB) anchored on the first molars (M1) and alternatively combined with different archwires (0.016 × 0.022-inch; 0.018 × 0.025-inch) with and without engaging second molars and (2) a buccal force against 0.018-inch, 0.016 × 0.022-inch, and 0.018 × 0.025-inch archwires with and without engaging the left lateral incisor (I2). RESULTS: Without fixed appliances, stresses were assumed by M1; with fixed appliances, stresses were distributed on all teeth, decreasing mesially toward the midline. Direct buccal pull exerted most stress on neighboring I2 (19-20% with different wire sizes) and first premolar (12-17%), decreasing distally, along a similar pattern with different archwire sizes. When I2 was bypassed, stresses on adjacent teeth increased only by 3-6%. Higher stresses occurred with the lighter round wire. CONCLUSIONS: This first research on stresses on adjacent teeth during PIC traction provided needed quantitative data on the pattern of stress generation, suggesting the following clinical implications: use of distal-vertical pull from posterior anchorage (TPB) as initial movement and when using a buccal force, bypassing the lateral incisor and using heavier wires that would minimize side effects.

Palatally impacted canines: A new 3-dimensional assessment of severity based on treatment objective.

INTRODUCTION: The severity of a palatally impacted canine (PIC) is gauged radiographically on 2-dimensional and 3-dimensional positional components: eg, angulation and height. We hypothesized that the position of a PIC relative to its virtual alignment in the arch is a better indication of impaction severity and treatment requirements. The aims of this research were to evaluate variations in PIC location on 3-dimensional images and to determine positional components associated with impaction severity. METHODS: Linear and angular measurements of 38 PICs from 28 cone-beam computed tomography scans were made on the panoramic, coronal, sagittal, and axial sections. Measurements included angulation of the PIC to the virtually aligned canine, midline, and palatal plane; and distances between cusp tip and apex to various reference planes-eg, occlusal and midpalatal. Statistical assessments comprised t tests for group comparisons based on PIC and virtually aligned canine severity (cutoff at 30°) and Pearson product moment correlations for associations among variables. RESULTS: Angulations of the PIC to the virtually aligned canine were 32.5° ± 15.5° (range, 9°-59°) and 19.6° ± 6.9° and 45.37° ± 9.6°, respectively, in the less severe and more severe groups (P <0.001). Group differences were significant (0.023