Automated Identification of Cephalometric Landmarks: Part 2- Might It Be Better Than human?

Abstract

Objectives: To compare detection patterns of 80 cephalometric landmarks identified by an automated identification system (AI) based on a recently proposed deep-learning method, the You-Only-Look-Once version 3 (YOLOv3), with those identified by human examiners. Materials and Methods: The YOLOv3 algorithm was implemented with custom modifications and trained on 1028 cephalograms. A total of 80 landmarks comprising two vertical reference points and 46 hard tissue and 32 soft tissue landmarks were identified. On the 283 test images, the same 80 landmarks were identified by AI and human examiners twice. Statistical analyses were conducted to detect whether any significant differences between AI and human examiners existed. Influence of image factors on those differences was also investigated. Results: Upon repeated trials, AI always detected identical positions on each landmark, while the human intraexaminer variability of repeated manual detections demonstrated a detection error of 0.97 ± 1.03 mm. The mean detection error between AI and human was 1.46 ± 2.97 mm. The mean difference between human examiners was 1.50 ± 1.48 mm. In general, comparisons in the detection errors between AI and human examiners were less than 0.9 mm, which did not seem to be clinically significant. Conclusions: AI showed as accurate an identification of cephalometric landmarks as did human examiners. AI might be a viable option for repeatedly identifying multiple cephalometric landmarks.

Deep Convolutional Neural Networks Based Analysis of Cephalometric Radiographs for Differential Diagnosis of Orthognathic Surgery Indications

“…This method is not only influenced by the measurement value that is entered, but also has the disadvantage of introducing many errors in the process of considering the landmark points. Recently, an automated landmark detection method has been introduced, but the performance is comparable or slightly less than manual detection by specialists [31]. Furthermore, there is another disadvantage of overfitting, which can easily occur when the measurement values have a similar meaning input in the machine- In cases of successful prediction, the ROI identified by class activation mapping (CAM) was mainly focused on the maxillary and mandibular teeth, mandibular symphysis, and mandible.…”

Evaluation of an automated superimposition method for computer-aided cephalometrics

“…The present study greatly increased the number (n ¼ 283) of images that were taken from various malocclusion patients. 14,17 This was somewhat different from the methods used in most previous studies. Since the tracings of the same image were superimposed, any difference could indicate an error either because of tracing error or superimposition error.…”

“…In a recent study regarding fully automatic landmark identification by artificial intelligence (AI), it took only 0.05 seconds to identify 80 cephalometric landmarks per image. 14,17 Likewise, the mathematical computation required in the proposed method can be solved without much difficulty.…”

“…Details on the subject characteristics can be found elsewhere. 14,17 Proposed Superimposition Method…”

“…Computer-aided cephalometrics, automatic identification of cephalometric landmarks via artificial intelligence (AI), and cephalometric analysis using smartphone apps all have gotten attention. [14][15][16][17] Among various superimposition methods, Björk's structural method has been considered as the gold standard. 18 However, this method was used to perform manual superimposition.…”