Prediction of Fishman's skeletal maturity indicators using artificial intelligence

Abstract

The present study aimed to evaluate the performance of automated skeletal maturation assessment system for Fishman's skeletal maturity indicators (SMI) for the use in dental fields. Skeletal maturity is particularly important in orthodontics for the determination of treatment timing and method. SMI is widely used for this purpose, as it is less time-consuming and practical in clinical use compared to other methods. Thus, the existing automated skeletal age assessment system based on Greulich and Pyle and Tanner-Whitehouse3 methods was further developed to include SMI using artificial intelligence. This hybrid SMI-modified system consists of three major steps: (1) automated detection of region of interest; (2) automated evaluation of skeletal maturity of each region; and (3) SMI stage mapping. The primary validation was carried out using a dataset of 2593 hand-wrist radiographs, and the SMI mapping algorithm was adjusted accordingly. The performance of the final system was evaluated on a test dataset of 711 hand-wrist radiographs from a different institution. The system achieved a prediction accuracy of 0.772 and mean absolute error and root mean square error of 0.27 and 0.604, respectively, indicating a clinically reliable performance. Thus, it can be used to improve clinical efficiency and reproducibility of SMI prediction.

Figure 1

Overview of the steps involved in the modified SMI-based skeletal maturation assessment system. In the pre-processing step, the image size of hand-wrist radiographs is adjusted, and right-hand radiographs are mirrored. In addition to the detection and classification of each ROI, the evaluation of the whole hand as a holistic image is performed to enhance the performance by including maturity features that are not parts of the ROIs.

Figure 2

Automated detection of regions of interest. In addition to the ROIs used in the TW3 method, all of the six ROIs used for Fishman’s SMI, as well as the carpal bones are detected. (DP3, 3rd distal phalanx; MP3, 3rd middle phalanx; MP5, 5th middle phalanx; PP3MC, 3rd proximal phalanx/metacarpal; PP1, 1st proximal phalanx, MC1, 1st metarcarpal).

Figure 3

Result screen displaying the SMI stage and bone age predicted by the SMI-modified automated skeletal maturation assessment system.

Figure 4

Confusion matrices showing the results of primary validation and accuracy test following fine tuning.