Skeletal-versus soft-tissue-based cephalometric analyses: is the correlation reproducible?

Is automatic cephalometric software using artificial intelligence better than orthodontist experts in landmark identification?

BACKGROUND

To evaluate the techniques used for the automatic digitization of cephalograms using artificial intelligence algorithms, highlighting the strengths and weaknesses of each one and reviewing the percentage of success in localizing each cephalometric point.

METHODS

Lateral cephalograms were digitized and traced by three calibrated senior orthodontic residents with or without artificial intelligence (AI) assistance. The same radiographs of 43 patients were uploaded to AI-based machine learning programs MyOrthoX, Angelalign, and Digident. Image J was used to extract x- and y-coordinates for 32 cephalometric points: 11 soft tissue landmarks and 21 hard tissue landmarks. The mean radical errors (MRE) were assessed radical to the threshold of 1.0 mm,1.5 mm, and 2 mm to compare the successful detection rate (SDR). One-way ANOVA analysis at a significance level of P < .05 was used to compare MRE and SDR. The SPSS (IBM-vs. 27.0) and PRISM (GraphPad-vs.8.0.2) software were used for the data analysis.

RESULTS

Experimental results showed that three methods were able to achieve detection rates greater than 85% using the 2 mm precision threshold, which is the acceptable range in clinical practice. The Angelalign group even achieved a detection rate greater than 78.08% using the 1.0 mm threshold. A marked difference in time was found between the AI-assisted group and the manual group due to heterogeneity in the performance of techniques to detect the same landmark.

CONCLUSIONS

AI assistance may increase efficiency without compromising accuracy with cephalometric tracings in routine clinical practice and research settings.

The Airway Volume Related to the Maxillo-Mandibular Position Using 3D Analysis

Objective

The aim of this research was to compare three cephalometric analyses and their correlation with the airway volume in subjects with different skeletal classes using 2D and 3D images. Study Design. Cross-sectional descriptive study. Material and Method. Steiner, McNamara, and Ricketts analyses and the airway volume were compared in 115 subjects who were candidates for orthognathic surgery under diagnosis using cone beam computed tomography (CBCT); 46 males (40%) and 69 females (60%) were included. The sagittal positions of the maxilla and mandible, the angulation of the mandibular plane, the sagittal positions of the upper and lower incisors, measurements of the largest or shortest airway area, and the volume were compared using Spearman's test considering a p value < 0.05.

Results

Differences were observed between the Steiner and McNamara techniques for the sagittal position of the maxilla (p = 0.01). For mandibular angulation, there was a greater difference between values for Steiner and Ricketts techniques (p = 0.001). In the upper incisor, the results for McNamara and Ricketts techniques were significantly different (p = 0.004). Analysing the airway, subjects with a class II skeletal pattern had a smaller volume than those with a class III pattern (p = 0.034).

Conclusion

It may be concluded that skeletal class II patients have a significantly smaller airway volume than class III patients. The skeletal parameter does not always relate to the airway volume; however, the high mandibular angle could be related to the airway conditions.

Facial Versus Skeletal Landmarks for Anterior-Posterior Diagnosis in Orthognathic Surgery and Orthodontics: Are They the Same?

PURPOSE

The purpose of this investigation was to evaluate diagnostic agreement in anterior-posterior (AP) categorization of the maxilla and mandible between a skeletal-landmark analysis and a facial-landmark analysis for treatment planning of orthognathic surgery and orthodontics.

MATERIALS AND METHODS

This retrospective, consecutive case series of adult patients who presented to the Mayo Clinic orthodontic department compared maxillary and mandibular AP diagnoses. Steiner's analysis of the sella-nasion-A point angle and sella-nasion-B point angle was used for a skeletal-landmark diagnosis. Element II of Andrews' 6 elements of orofacial harmony was used for a facial-landmark diagnosis. Both diagnoses were categorized as either deficient, optimal, or excessive for each jaw. Categorization of the skeletal landmark was determined by normative data, whereas the facial landmark provides a customized categorization unique to each individual.

RESULTS

Weighted κ statistics were completed to test agreement between the categories determined by the skeletal and facial landmarks. The maxilla showed poor agreement, and the mandible showed slight agreement.

CONCLUSIONS

No agreement was found for AP categorization of the maxilla and mandible between skeletal-landmark and facial-landmark analyses. Most mandibles were diagnosed as retrognathic by the facial landmark, whereas most were diagnosed as optimal by the skeletal landmark. When the 2 landmarks disagreed, the facial landmark defined the optimal position farther anterior. The landmark chosen for diagnosis will impact the optimal jaw position and can affect orthognathic and orthodontic outcomes.